Effect of Aggregates - BM - STP1147

Copyright by ASTM Int'l (all rights reserved); Thu Dec 22 07:29:10 EST 2011
Downloaded/printed by
Indian Institute of Technology Kharagpur pursuant to License Agreement. No further reproductions authorized.
S T P 1147
Effects of Aggregates and

Mineral Fillers on Asphalt
Mixture Performance
Richard C. Meininger, editor
ASTM Publication Code Number (PCN)

04-011470-08
AsTM
1916 Race Street
Philadelphia, PA 19103
Library of Congress Cataloging-in-Publication Data
Effects of aggregates and mineral fillers on asphalt

mixture performance.
(STP ; 1147)
i. Asphalt emulsion mixtures--Testing. 2. Aggregates
(Building materials). 3. Fillers (Materials).
3. Minerals. I. Meininger, Richard C. II. Series: ASTM
special technical publication ; 1147.
TE275.E34 1992 625.8'5 92-31159
ISBN 0-8031-1468-0 CIP
Copyright 9 AMERICAN SOCIETY FOR TESTING AND MATERIALS, Philadelphia, PA. All
rights reserved. This material may not be reproduced or copied, in whole or in part, in any printed,
mechanical, electronic, film, or other distribution and storage media, without the written consent of
the publisher.
Photocopy Rights
Authorization to photocopy items for internal or personal use, or the internal or personal use of
specific clients, is granted by the AMERICAN SOCIETY FOR TESTING AND MATERIALS for users
registered with the Copyright Clearance Center (CCC) Transactional Reporting Service, provided
that the base fee of $2.50 per copy, plus $0.50 per page is paid directly to CCC, 27 Congress St.,
Salem, MA 01970; (508) 744-3350. For those organizations that have been granted a photocopy
license by CCC, a separate system of payment has been arranged. The fee code for users of the
Transactional Reporting Service is 0-8031-1468-2/92 $2.50 + .50.
Peer Review Policy
Each paper published in this volume was evaluated by three peer reviewers. The authors
addressed all of the reviewers' comments to the satisfaction of both the technical editor(s) and the
ASTM Committee on Publications.
The quality of the papers in this publication reflects not only the obvious efforts of the authors and
the technical editor(s), but also the work of these peer reviewers. The ASTM Committee on
Publications acknowledges with appreciation their dedication and contribution to time and effort on
behalf of ASTM.
Printed in Ann Arbor, MI

October 1992
Foreword
This publication, Effects of Aggregates and Mineral Fillerson Asphalt Mixture Performance,
contains papers presented at the symposium of the same name, held in San Diego, CA on
10 Dec., 1991. The symposium was sponsored by ASTM Committee D-4 on Road and
Paving Materials. Richard C. Meininger of the National Ready Mixed Concrete Association
in Silver Spring, MD presided as symposium chairman and is the editor of the resulting
publication.
Contents
O v e r v i e w - - R . C. MEININGER vii
AGGREGATES
T h e U s e o f Steel Slag in A s p h a l t i c C o n e r e t e - - N . A. ALl, J. S. S. CHAN,

T. PAPAGIANNAKIS, E. G. THERIAULT, AND A. T. BERGAN 3
D e v e l o p m e n t and Analysis of C e m e n t - C o a t e d A g g r e g a t e s for A s p h a l t M i x t u r e s - -

F. M. BAYOMY 19
Effects of A g g r e g a t e Absorption and C r u s h P e r c e n t a g e on Bituminous C o n c r e t e - -

G. P. BOUCHARD 35
Selection of A g g r e g a t e P r o p e r t i e s to M i n i m i z e Rutting of Heavy Duty

P a v e m e n t s - - s . A. CROSS AND E. R. BROWN 45
Effects of A g g r e g a t e P r o p e r t i e s o n Flexible P a v e m e n t R u t t i n g in A l a b a m a - -
F. PARKER AND E. R. BROWN 68
Evaluation of P e r c e n t F r a c t u r e of Hot Mix Asphalt Gravels in lndiana--c. A.

SANDERS AND E. L. DUKATZ 90
MINERAL FILLERS
Effect of Filler Type and Content on P r o p e r t i e s o f A s p h a l t C o n c r e t e M i x e s - - A . AL-

SUHAIBANI, J. AL-MUDAIHEEM, AND F. AL-FOZAN 107
R h e o l o g i c a l P r o p e r t i e s of M i n e r a l F i l l e r - A s p h a l t Mastics and Its Importance to

Pavement P e r f o r m a n c e - - D . A. ANDERSON, H. U. BAHIO, AND R. DONGRE 131
Effects of M i n u s N o . 200-Sized A g g r e g a t e o n F r a c t u r e B e h a v i o r of D e n s e - G r a d e d
H o t - M i x A s p h a l t - - D . A. ANDERSON, R. DONGRE, D. W. CHRISTENSEN, III,
AND E. L. DUKATZ 154
A Simple M e t h o d for D e t e r m i n i n g t h e Surface A r e a of F i n e A g g r e g a t e s a n d Fillers

in B i t u m i n o u s M i x t u r e s - - R . P. CHAPUIS AND P. P. LEGARE 177
Effect of Quality and Quantity of Locally Produced Filler (Passing Sieve No. 200)
on Asphaltic Mixtures in Dubai--M. A. SHAHROUR AND B. G. SALOUKEH 187
MIXTURE EVALUATION
Laboratory Evaluation of Rutting Potential of Asphalt MixturesuT. F. FWA AND

S. A. TAN 211
Providing Sufficient Void Space for Asphalt Cement: Relationship of Mineral

Aggregate Voids and Aggregate Gradation--G. A. HUBER AND T. S. SHULER 225
Gyratory Evaluation of Aggregate Blends to Determine Their Effect On Shear

Resistance and Sensitivity to Asphalt ConcretemB. E. RUTH, X. SHEN, AND
L. H. WANG 252
Effect of Aggregate Size and Other Factors on Refusal Density of Asphalt

Concrete by Vibratory CompactionmT. s. SHULER AND G. A. HUBER 265
FATIGUE, MODELING, AND THEORETICAL
Effect of Aggregate on Asphalt Mixture Cracking Using Time-Dependent Fracture

Mechanics A p p r o a c h - - o . A B D U L S H A F I 281
A New Approach for Prediction of Permanent DeformationuD. PERDOMO,

J. W. BUTTON, AND R. L. LYTTON 295
Effects of Aggregate Type and Gradation on Fatigue and Permanent Deformation

of Asphalt Concrete--Y R. KIM, N. KIM, A N D N. P. K H O S L A 310
The Effect of Aggregate Gradation on the Creep Response of Asphalt Mixtures

and Pavement Rutting Estimates--J. M. MATTHEWSAND C. L. MONISMITH 329
Author Index 349
Subject Index 351
Overview
Background
ASTM Committee D-4 on Road and Paving Materials is active in sponsoring symposia
and the publication of technical papers related to the standardization work of the Committee.
Subcommittee D04.92 is the coordinating group which works with Committee D-4's Officers,
Executive Subcommittee, and the ASTM Staff in selecting topics of interest, presenting the
symposium during an ASTM committee week, and then publishing peer-reviewed papers,
usually in a Special Technical Publication (STP) such as this.
This STP, Effects of Aggregates and Mineral Fillers on Asphalt Mixture Performance,
resulted from the Committee D-4 Symposium held on 10 Dec., 1991, at the ASTM Stan-
dardization Meetings in San Diego, CA. This critical topic was chosen because of the recent
interest and research concerned with the importance and influence which aggregates and
fine mineral fillers have on the long-term durability and performance of hot mix asphalt
pavements, particularly resistance to rutting of the pavement and stripping of asphalt from
aggregates when moisture is present. The call for papers brought in 40 abstracts from authors
all over the world who wished to present papers at the symposium. Of the 19 papers included
in the STP, twelve were accepted for presentation at the symposium and, due to time
limitations, an additional seven were accepted for publication only.
SHRP and Other Research

Prior to, and during the time of the organization of this symposium, a great deal of
research had been in progress relating to asphalt cement binders (including fillers), methods
of asphalt mixture evaluation, and the properties of fine and coarse aggregates, principally
grading, particle shape, and the interaction at the asphalt-aggregate interface. Much of the
recent research in the United States has been sponsored by the Strategic Highway Research
Program (SHRP), the National Cooperative Highway Research Program (NCHRP), and
various state highway agencies. This certainly has heightened interest in the symposium and
the technical papers included herein. Also, in light of the prospects for additional research,
such as the Long Term Pavement Performance Study and highway funding aimed at helping
repair the highway infrastructure, it is important to make best use of aggregate resources
in asphalt overlays which can represent some of the most critical performance challenges
for asphalt-aggregate mixtures.
New ASTM standards are needed and existing standards for tests and specifications must
adapt to new technology reported here, and yet to come, as other recent research initiatives
are reported. There is much urgency now in the highway materials community to understand
and incorporate improved technology as it is developed because of the economic benefits
to be derived through better highway pavement surfaces and reduced maintenance expend-
itures.
This volume provides, in one place, a cross section of research and practice on the use
of aggregates and mineral fillers in asphalt paving mixtures. It has international perspective
as well as providing insight into SHRP and state agency research and technology needs. It
also presents the recent work and thinking of important research organizations, such as the
National Center for Asphalt Technology at Auburn University and the laboratories of the
Asphalt Institute.
vii
viii OVERVIEW
The papers are arranged in four groups designed to aid the reader in locating papers of
interest and to compare and contrast the range of work and opinions presented:
(1) Aggregates--The first section relates directly to the practical use of aggregates in
asphalt pavements and includes papers by aggregate and hot mix asphalt producers,
as well as researchers that examine the performance of aggregates both in the lab-
oratory and in existing pavements.
(2) Mineral Fillers--The next grouping includes researchers who have made extensive
investigations of the properties of mineral fillers and their effect on asphalt mixtures
performance. Better understanding of the roll of fine material (passing the No. 200
sieve) on the performance of asphalt mixtures is an important new frontier in paving
material research.
(3) Mixture Evaluation--This group includes papers relating to the need for the design
and evaluation of asphalt-aggregate mixtures. New and improved laboratory and
field methods and concepts for evaluation and control of mixtures are described.
(4) Fatigue, Modeling, and Theoretical--The last group of papers concerns the use of
aggregate and mixture properties to develop systems of predicting or modeling the
long-term performance of asphalt pavements in terms of fatigue, fracture mechanics,
and prediction of long-term rutting performance.
While many of the papers might have been placed in several groups, it is hoped that this
organization will help the reader understand and use the technology presented and to help
Committee D-4 in developing the new standards and tests needed to advance materials
technology in the highway community.
Importance of Mixture Performance

In the long run, it is important to develop test methods and specifications for constituent
materials and for combinations of materials that are related to pavement performance. For
example, with respect to grading and particle shape characteristics of aggregates, rather than
place rigid limits on aggregate grading, blend percentages, or particle shape properties, the
objective should be to link tests and specifications to meaningful performance criteria. The
measured properties of individual materials would than serve as guides in the materials
selection and mix design processes and not an end in themselves. The key is developing
evaluation procedures which will give an accurate indication of the long-term performance
of a mixture when produced, placed, and compacted properly. An important corollary is
the employment of reliable procedures for verification of mix performance as it is produced
through the hot mix plant, and the use of quick quality control tests which will verify that
the desired mixture is being produced in a consistent manner.
Future Symposia
Committee D-4 is planning several future symposia on topics related to the production
and evaluation of Hot Mix Asphalt. In December of 1992, a symposium on A Critical Look
at the Use of Waste Materials in Hot Mix Asphalt will be held in Miami, FL. Plans for
other future symposia include Physical Properties of Asphalt Cement Binders in 1993. Also,
there are plans to have a symposium on asphalt-aggregate mixture design in 1994.
ASPHALT MIXTURE PERFORMANCE ix
Acknowledgements
I wish to thank the other members of the Committee D-4 Papers and Symposia Subcom-
mittee (D04.92) who all helped in the review of abstracts and papers for the 1991 Symposium
on Effects of Aggregates and Mineral Fillers on Asphalt Mixture Performance. They were:
John Hardin, Bill Maupin, Fred Waller, and Ken Wardlaw. Also, I would like to thank
John Hardin for his service as Vice-Chairman of the Symposium and his help at the session
in handling details. I am also very appreciative of the very professional and friendly help
which was received from the ASTM symposium and publications staff. Their help, partic-
ularly in the advanced planning of the symposium and in handling all the paperwork in the
peer review process, was invaluable. Without their efforts, it would have been very difficult
to keep the production of the symposium and STP volume on target and on schedule.
Richard C. Meininger
National Ready Mixed Concrete Association
Silver Spring, MD; symposiumchairman
Aggregates
CopyrightbyASTMInt'l(allrightsreserved);ThuDec2207:29:10EST2011
Downloaded/printedby
IndianInstituteofTechnologyKharagpurpursuanttoLicenseAgreement.Nofurtherreproductionsauthorized.
Nouman A. Ali, 1 Joseph S.S. Chan,2 Thomas Papagiannakis, 3 Eric G. Theriault, 4 and
Arthur T. Bergan5
THE USE OF STEEL SLAG IN ASPHALTIC CONCRETE
REFERENCE: Ali, N.A., Chart, J.S.S, Papagiannakis, T., Theriault, E.G., and Bergan, A.T.,
"The Use of the Steel Slag in Asphaltic Concrete", Effects of Aggregates and Mineral
Fillers on Asphalt Mixture Performance, ASTM STP 1147, Richard C. Meininger, editor,
American Society for Testing and Materials, Philadelphia, 1992.
ABSTRACT: Advanced testing and evaluation were performed on laboratory prepared

Marshall specimens from eight asphalt concrete mixes. Five of these mixes contain steel slag
aggregates. These were characterized and results were compared to the remaining three
conventional asphalt concrete mixes commonly used in the province of Nova Scotia, Canada.
Advanced testing included indirect tensile strength, resilient modulus, creep and permanent
deformation, moisture damage, and fatigue. Laboratory results showed that asphalt mixtures
containing steel slag aggregates exhibited more superior characteristics than those of
conventional mixes. Pavement performance of three idealized pavement cross-sections were
modelled using VESYS and with material properties measured. Mixes which contained 100%
steel slag aggregate had shown better resistance to rutting and to cold temperature cracking.
KEYWORDS: steel slag, mechanical properties, pavement performance
One of the most prominent and severe surface distresses on major highways
throughout Canada is instability rutting. Highway agencies are actively searching for new
products/procedures which could provide a rut resistant asphaltic concrete surface and still
maintain the other desirable characteristics, i.e. durability, flexibility, etc.
1Assistant Professor, Department of Civil Engineering, Technical University of Nova

Scotia, Halifax, N.S., Canada, B3J 2X4.
2Research Engineer, Transportation Centre, College of Engineering, University of
Saskatchewan, Saskatoon, Sk., Canada S7N 0W0.
3Assistant Professor, Faculty of Engineering, Memorial University of Newfoundland,
St. John's, Newfoundland, Canada, A1B 3X5.
4Consulting Engineer, Jacques Whifford and Associates Ltd., 3 Spectacle Lake Dr.,
Dartmouth, N.S., Canada, B3B 1W8.
SProfessor, Department of Civil Engineering, University of Saskatchewan, Saskatoon,
Sk., Canada, S7N 0W0.
3
Copyright9 1992 by ASTM International
www.astm.org
4 ASPHALTMIXTURE PERFORMANCE
In the meantime, as natural supplies of high quality coarse and f'me aggregates used in
highways have become less abundant, the highway engineer is faced with the challenge of
finding altemative aggregates to meet the growing requirements for these materials. Both
environmental and economic factors also contribute to the growing need for the use of
reclaimed materials in bituminous pavements.
One of these reclaimed materials, steel slag, has not been used extensively in
pavements even though it has shown promising results in pavement research as an altemative
source of aggregate. Steel slag has also been proven particularly effective in providing a
highly stable, rut resistant surface.
The objectives of this study are:

1. to design an asphalt concrete mix using steel slag as the only aggregate;
2. to explore possibilities of improving mechanical properties of asphalt mixes using steel
slag instead of natural or manufactured (fractured) aggregate; and
3. to evaluate the effect of using steel slag on mitigating pavement distress and improving
performance of pavement sections.
A P P L I C A T I O N O F STEEL SLAG IN ASPHALTIC CONCRETE
Steel slag is the coarse portion of the residues produced by refining iron to steel in
open-hearth, basic oxygen, and electric arc furnaces. Steel slag is liquid cooled from its
molten state and is a high-density, glass-like material. Steel slag asphalt concrete mixes have
been used in countries such as the United States, England, Japan, and Canada, particularly in
areas with concentration of iron and steel production.
Experience gained in southem Ontario since the early 1970's has shown that the use
of steel slag in asphalt concrete provides the advantageous features of the steel slag which
results in high quality mix. The main advantages of using steel slag in the asphalt mixes are:
- Steel slag is 100% crushed and angular with gradation that usually requires no blending;
- Steel slag asphalt mixes have very high stabilities, satisfactory flows and excellent
stripping resistance;
- Steel slag used as surface course has good wear and skid resistance;
- Steel slag mixes have good heat retention and compactibility.
The main disadvantages of steel slag asphalt mixes am:
- When in contact with moisture, steel slag has a potentially expansive nature of up to 10%
attributable to the hydration of the free CaO and magnesium hydrates;
- Steel slag asphalt mixes have high unit weight which increases the cost of transportation;
- The angularity, surface texture of the steel slag, and its higher absorptive characteristics
increase the asphalt cement content required;
-Steel slag varies in its chemical composition from plant to plant and even within the same
plant.
ALl ET AL. ON THE USE OF STEEL SLAG 5
MATERIALS
The materials used in the laboratory evaluation were: (1) steel slags from both the
electric arc (EA) and the open hearth (OH) furnaces provided by the Sysco plant in Sydney,
Nova Scotia; (2) aggregates that meet the Nova Scotia Depamnent of Transportation and
Communications (NSDOT&C) specifications for Type B asphalt special mix; and (3) ASTM
designated 150-200 penetration grade asphalt cement.
Eight coarse aggregate - fine aggregate combinations were selected to produce mixes
with a wide range of gradation and proportion of steel slag aggregates:
Mix A & B: 100% EA steel slag aggregate. The difference between Mixes A and B is in
the amount of the asphalt cement used. The asphalt cement content of Mix B
is 6.5% and Mix A is at the optimum content (7%).
Mix C: 47% quarry coarse aggregate and 53% EA slag as fine aggregate.
Mix D: 40% EA slag as coarse aggregate, 40% quarry fine aggregate, and 20%
washed quarry fine aggregate.
Mix E: 98% OH slag and 2% filler.
Mix F: 42% quarry coarse aggregate, 32% quarry fine aggregate, and 26% washed
quarry fine aggregate. This mix is considered to be the premium asphalt
concrete mixture now used in the province of Nova Scotia.
Mix G: 40% crushed gravel as coarse aggregate, 30% crushed gravel as fine aggregate,
and 22% natural sand. This is an actual design mix from the NSDOT&C.
Mix H: 35% crushed gravel coarse, 10% quarry coarse aggregate, 35% crushed fine
gravel, and 20% natural blend sand. This is an actual design mix from the
NSDOT&C.
Four of the eight mixes contained electric arc (EA) slag and one contained slag
produced by the open hearth (OH) furnaces. OH slag is no longer in production and is
included for comparison purposes only. Table 1 shows the physical properties of both steel
slags and aggregates used in this study.
Marshall mix design procedures were conducted on all eight mixes using the 75 blows
criteria. A summary of mix design for the eight mixes can be found in Table 2. It should be
noted that asphalt mixes containing steel slag required relatively higher asphalt contents. Mix
A, which contained 100% EC slag, required 16% to 33% more asphalt cement than those
mixes with no slag.
ADVANCED TESTING AND RESULTS
An MTS closed loop electrohydraulic testing system was used in the testing program.
This system is operated by the Department of Civil Engineering, University of Saskatchewan,
and can apply a haversine load pulse over a range of frequencies for a range of load
magnitudes and durations. The system was interfaced with two microcomputers for load
control and data acquisition purposes.
An IBM XT compatible micro-computer was used to provide signals to the MTS

system for load application allowing various load functions to be applied. High-speed data
collection was achieved using a second IBM AT compatible micro-computer.
O)
TABLE 1 -- Aggregate Properties

GO
"13
T
COARSE AGGREGATES I'--

--q
EA Slag OH Slag Conrads L&F Pit Folly Mtn. 6 Mile Br
X
.-I
Petrographic No. 128 112 154 104 142 C
"n
L.A. Abrasion (%) 21.7 29.8 20.1 m
0.8 2.6 4.94 13

Soundness (MgSO4,%) m
~n
Specific Gravity (ASTM C127): 0
Apparent 3.317 2.560 2.733 2.724 2.817 2.719
Bulk 3.044 3.284 2.690 2.592 2.754 2.651 z
c~
Bulk (SSD) 3.128 3.361 2.705 2.646 2.776 2.676 m
Absorption 2.750 2.350 0.590 1.870 0.820 0.940
FINE AGGREGATES
EA Slag OH Slag Conrads Conrads(w) L&F Pit Gays River 6 Mile Br Keddy's
Soundness (MgSO4,%) 10.7 18.7 12.6
Specific Gravity (ASTM C128):

Apparent 3.420 3.440 2.734 2.739 2.735 2.713 2.736
Bulk 2.903 2.942 2.639 2.639 2.603 2.607 2.594 2.623
Bulk (SSD) 3.054 3.086 2.675 2.675 2.651 2.638 2.664
Absorption 5.120 4.920 1.377 1.380 1.860 1.130 1.690 1.570

(w) - washed aggregates
TABLE 2 -- Characteristics of Asphalt Concrete Mixes
Mix Designation A B C D E F G H
Aggregates (source - percentage):
1. EASIag(100%) EA Slag(100%) Conrads(47%) EA Slag(40%) OH Slag(98%) Conrads(42%) L&F Pit(48%) 6 Mile Br(35%)
2. EA Slag(53%) Conrads(40%) Filler(2%) Cortrads(32%) L&F Pit(30%) Folly Mtn(10%)
3. Conrads(w)(20%)* Conrads(w)(26%) GaysRiver(22%) 6 Mile Br(35%)
4. Keddy's(20%)
Asphalt Cement (%) 7.0 6.5 6.1 6.5 5.75 5.3 5.25 6.0
Max. Theo. Density 2.904 2.927 2.628 2.648 2.945 2.516 2.456 2.695
Bulk Rel. Density 2.781 2.776 2.522 2.514 2.827 2.417 2.354 2.355
Air Voids (%) 4.2 5.2 4.0 5.0 4.0 3.9 4.2 3.9
Stability (kN) 13.4 12.0 14.0 12.6 19.7 8.0 8.2 7.0 m
Flow (0.25ram) 13 13 15 17 10 11 9 10 "~
VMA (%) 12.9 >
12.6 15.5 14.9 14.1 14.0 14.2 16.0 .r--
O
Gradation (pm): Z
28000 (1.06 in.) 100.0 .--4

100.0 100.0 100.0 100.0 100.0 100.0 100.0 -r
m
20000 (3/4 in.) 97.0 97.0 97.0 95.0 95.3 96.6 99.3 99.9
14000 (0.53 in.) 89.4 c
89.4 84.9 83.0 83.6 85.3 82.0 92.9 6o
m
10000 (3/8 in.) 73.0 73.0 73.0 68.0 71.8 70.7 68.5 75.9 0
5000 (No. 4)) 53.0 53.0 53.0 52.6 53.0 50.5 52.9 52.3 -n
2500 (No. 8) 37.0 37.0 37.0 32.0 39.2 35.4 45.1 35.3
1250 (No. 16) 28.0 m
28.0 28.0 20.6 26.4 25.0 35.2 22.5 m
630 (No. 30) 23.0 23.0 23.0 14.4 18.6 17.7 21.4 15.1 r'-
315 (No. 50) 17.0 17.0 23.0 10.3 12.6 12.5 11.4 9.2
160 (No. 100) 11.0 11.0 17.0 8.0 8.2 8.9 6.6 6.3
80 (No. 200) 6.0 6.0 6.0 6.1 5.6 6.3 4.2 4.7
(w) - washed aggregates
-.4
An environmental chamber, capable of providing air temperatures between -40~ and

+50~ was used to control test temperatures during the testing of the asphalt concrete
specimens.
Resilient Modulus Test
The diametral resilient modulus test method developed by Schmidt W and detailed in
the ASTM Test Method for Indirect Tension Test for Resilient Modulus of Bituminous
Mixtures (D-4123) was used in this study.
Repeated haversine loading was used in all resilient modulus testing in order to avoid
impact loading to specimens. Thte~ levels of temperature were tested; 0~ 200C, and 44~
Two load frequencies were used: 1 Hz (0.1 second loading and 0.9 second unloading) and
0.33 Hz (0.25 second loading and 2.75 second unloading). Results of resilient modulus testing
are summarized in Table 3.
TABLE 3 -- Diametrical Resilient Modulus (MPa)
Temperature, ~
Mix 0 20 44
Designation F1 F2 F1 F2 F1 F2
A 22000 19500 4200 3250 1050 930

B 9100 7000 1000 950 ......
C 12700 9500 1400 1200 900 750
D 14000 11900 2800 2100 1000 780
E 21500 19000 4000 3150 850 750
F 18250 15400 2700 2100 1100 600
G 19000 16600 2200 1700 ......
H 18000 16150 2000 1600 ......
F1 - 1 Hz; F2 - 0.33 Hz
It should be noted that Mixes B, G, and H were too soft to be tested at 44~ and thus
no resilient modulus data could be generated for these mixes at this temperature. The data in
Table 3 indicates that at the low temperature (0~ Mixes A and E had the highest M~ values.
Both of these mixes contained almost 100% steel slag and were mixed at the optimum asphalt
content. On the other hand, at high service temperature (44~ all mixes that were tested
have no significant difference in their Mr values.
It is significant that the results at 44~ which approximated the temperature of the
asphalt concrete pavement in Nova Scotia on a warm sunny day, were in the order of 5-10%
of the values at 0~ This clearly illustrated the dramatic effect of temperature has on the
stiffness of asphalt concrete and why instability rutting occurs during summer months.
Creep and Permanent Deformation Test
Indirect tensile loading was used to determine the effect of steel slag on the
viscoelastic behaviour of paving mixtures. This behaviour is usually measured by the creep
and the permanent deformation parameters. Creep and permanent deformation tests were
conducted in accordance with procedures outlined in VESYS user manual 212.]. The
fundamental objective of this test series was to obtain parameters which would be used in
predicting rutting performance of a pavement.
Specimens were tested at three temperatures (0~ 20~ and 44~ and the permanent
deformation characteristics can be found in Figures 1, 2, 3. As expected, permanent
deformation increased in an exponential fashion with loading time for all temperatures tested.
Figures 1, 2, and 3 also showed that mixes containing steel slag (except Mix B) exhibited
lower permanent deformation than those containing no slag. Therefore it could be inferred
that the steel slag asphalt mixes have less rutting potential.
Data from the 1000 seconds loading creep test were used to generate creep
characteristic curves at each test temperature (Figures 4, 5, and 6). As expected, the creep
moduli decreased with increased loading time and/or temperature. Mixes containing steel slag
and quarried aggregates exhibited higher creep modulus than did the conventional and special
mixes. Neither conventional nor Mix B, which was under-asphalted, survived the testing at
44~
Low Temperature Characteristics
The low-temperature tensile fracture strengths of briquette specimens were determined

using the indirect tensile test. The testing procedure used was described by Anderson &
Hahn [3_]. The diametral load was applied at a vertical deformation rate of 1.5 mm per minute
and at test temperatures of 0~ -10~ -20~ and -30~ These test temperatures resulted in
failure occurring between 60 to 160 seconds.
Paving mixes with low tensile strength have a tendency to develop low temperature
cracking in the field, especially when used as surface courses. Results of indirect tensile
strength tests are shown in Figure 7. Tensile strengths decreased with temperature,
particularly for temperatures higher than -20~ The results indicated that when subjected to
thermally induced stresses, the mixes containing steel slag will perform as well as or better
than mixes containing conventional aggregates.
Moisture Damage Test
Moisture sensitivity and the effects of stripping are closely related. Stripping is
primarily caused by water reducing the cohesive strength of the asphalt cement or a failure of
the adhesive bond between aggregate and asphalt cement.
Moisture damage evaluation can be accomplished using various methods, the most
common of which are summarized in a recent publication by the National Research Council in
the United States [4]. That report recommended that the procedure described in NCHRP 246
to be used for evaluating long-term moisture damage. This method uses either the ratio of the
resilient modulus or the indirect tensile strength of wet (moisture conditioned) and of dry
(unconditioned) samples as indicators of moisture susceptibility.
10 ASPHALT MIXTURE PERFORMANCE
-1
1~1
~- 1 0 - 2
~
~
r
C
D
F
4J
(/3
_+_,,
i0_3 -
EL 10 -4
10 -~ i I IIIIII I I I IIIIII I I L ttl~l= I I I IIIIIl l
0.1 1 10 100 1000

Load Duration (sec)
FIGURE 1 -- Permanent Strains at 0 Degree Celsius
10 -~I
.c_ I0 -2
0
L_
co
4J
10 -3.
(-
8L_
(1)
O_ 1 0 -~-
~.,,.,,~ F
4~G
10 -5 I I I I IIII I I I I I IIII I I I I I II~'I I I I I I Iill I
O. 1 10 I O0 1000
Load Duration (sec)
FIGURE 2 -- Permanent Strains at 20 Degrees Celsius
10 -~
c 10 -2
,-4,J
cq,) I 0 3
(-
0
E
%_
q.)
O_ 10 -4, ,,~,,,~t -,hr
~,~.,.,~ D
===== E
r ~ F
10 -s I I I I Jill ] I I I i lit I I I I I I III I I ~ I I I Ill 1
0.1 I 10 100 1000

Lood Durotion (sec)
FIGURE 3 -- P e r m a n e n t Strains at 44 Degrees Celsius
10 ~
cuo~o B
,& "" "" ,~" "~ C
0
- ~ D
10 4
Y
10 3
10 2
C3
10 I I I I Illl[ I l I I IIll I I I I I Illl I I I I I Elll I
0.1 1 10 100 1000

Looding Time (sec)
FIGURE 4 -- Creep M o d u l u s at 0 Degree Celsius
Indian Institute of Technology Kharagpur pursuant to License Agreement. No further reproductions authori
I04~
"q' D D ~ ~ B
.,,-.~ ,2,~r~ C
A
C~ ~..,.,,.,,. D
Q_
~-.~ 10 3
03
-O
O
:
o_10 2
(]3
(D
(D
10
0.1 1 10 100 1000
Loading Time (sec)
FIGURE 5 -- Creep Modulus at 20 Degree Celsius
10 ~ aa.".~--,~ C
~ ~ ~'-- ~--'~ E
10 2
0
CL
(D
(D
k_
(D
10 I I I I IIII i I I I I IIII1 I I I I IIII I 1 I I I illl I
0.1 1 10 100 1000

Loading Time (sec)
FIGURE 6 -- Creep Modulus at 44 Degree Celsius
6000 -
oo0oo B
O
4000 -- AA~A~
TTTTT G
C
2000
a 0 IIIIIIIIl[lllllllllllllllllllllllllllllllltllllll I
-40 -30 -20 -I0 0 I0
Temperature (Celsius)
FIGURE 7 -- Indirect Tensile Strengths at Various Temperatures
All eight mixes were conditioned as prescribed and the diametral resilient modulus
were performed before and after conditioning. Tests were performed with a load frequency of
1 Hz and at a temperature of 20~ The ratio of M, values for conditioned and unconditioned
specimens were determined. Table 4 summarized the result of the moisture damage tesl.
Previous research had indicated that moisture damage or stripping can occur to asphalt
concrete pavements when the ratio of dry to conditioned test specimen is below 0.70 to 0.75
[4__1.
TABLE 4 -- Results of Moisture Damage Testing
Mix Resifient Modulus (MPa)

Designation Dry Condition Wet Condition Ratio
A 4000 3400 0.85

B 1035 630 0.61
C 1400 1100 0.78
D 2700 2160 0.80
E 4030 3450 0.85
F 2720 1870 0.68
G 2200 1300 0.59
H 2050 1480 0.72
Frequency = 1 Hz; Temperature = 20~
Results in Table 4 indicated that mixes containing steel slag were not prone to
stripping. A surprising find was that mixes without steel slag all showed a propensity to
stripping as their ratios were all below 0.75.
Also noteworthy in the results was that the ratios of Mix A and Mix B, identical
mixes except for the different asphalt cement content, were dramatically different. The ratio
dropped from 0.85 for Mix A to 0.61 for Mix B. This reestablished that mix parameters, such
as asphalt cement content and air voids, will have significant effects on stripping potential of a
asphalt paving mix.
Fatigue Testing
Only three of the eight mixes (Mix A, Mix D, and Mix F) were considered for fatigue
testing. The fatigue characteristic of the mixes were measured using diametral indirect tensile
test in controlled stress mode. All tests were performed at a temperature of 21~ and load
frequency of 1 Hz with 0.1 second loading and 0.9 second unloading. Indirect tensile stresses
in the range of 40 to 170 kPa were utilized.
Fatigue analysis required the determination of induced tensile strain in the paving
mixture and relating the tensile strain to the allowable number of load applications by using
the following equation:
(1)
g
where
N, = the number of load applications to failure
E = the initial tensile strain
K , K2 = are the material constants which can be determined through regression
Since strain increased continually throughout the duration of the controlled stress test,
the initial strain reported herein was the indirect tensile strain obtained at mid-beight of
specimens after 200 load applications.
The fatigue characteristic curves for Mix A , Mix D, and Mix F are presented in
Figure 8. The fatigue parameters, K t and K 2, are summarized in Table 8. The values of K~
and K2 can be used as an indicator of how a "modifier" affects the fatigue mechanism of a
paving mixture. The flatter the slope of the fatigue curve, the larger the value of I~. If two
materials have the same K~ value, then a large value of K2 would indicate a potential for
longer fatigue life. On the other hand, when the fatigue curves are parallel (that is K s is
constant), a lower K1 value represents a lower fatigue life. Two intersecting fatigue curves
indicate that the magnitude of initial induced strain will determine which material will have a
longer fatigue life.
Results of the fatigue tests shown in Figure 8 indicated that the use of steel slag as
aggregate ( Mix A and Mix D) significantly increased the value of K~ when compared to
conventional paving mix (Mix F). The use of steel slag only increased the value of K 2
moderately.
1000
9--
8--
7-- ***** Mix F

6-- a_aga_a Mix D
~o A,-,,-,,-,,~ Mix A
I 5--
I,I
(D 4--
v
A
3--
~C
cf)
C-
O '~'~ D
100 i , i , i i, I , , ~ i , i ,, I I I I I I I I[[
10 ~ 10 ~ 10 s 10 6
NO. of Cycles to Foilure
FIGURE 8 -- Fatigue Characteristic Curves
PERFORMANCE ANALYSIS
To be considered as an altemative to aggregates now being utilized, steel slag must be

cost effective. Economic decisions should not be based on construction costs alone but rather
on life cycle costs, which encompass the performance and predicted serviceability index of the
pavement with time. The VESYS performance model was used to predict the performance of
three representative pavement sections. Performance can be expressed in terms of rutting,
cracking, roughness, and present serviceability index (PSI). Full details of the VESYS model
are described in reference [2_].
Three idealized pavement sections were modelled. Each pavement section has a 150
mm asphalt concrete layer over a 375 mm base course layer. The difference between the
three pavement sections is in the type of asphalt concrete mix used in the first layer. Two
sections utilized steel slag mixes (Mix A and Mix D) while the third used crushed aggregate
mix (Mix F). The mechanical properties of the base course and subgrade soil required by the
VESYS model were determined by a previous study [5_]. The mechanical properties of the
asphalt concrete layers (Mixes A, D, and 17) are summarized in Table 5. An analysis period of
20 years and an average traffic of the 130 ESAL per day were used in the analysis. A
summary of the VESYS model results is shown in Table 6.
Rut depth, a measure of permanent deformation in the wheel path, is a function of

permanent deformation parameters, stiffness of the materials, and traffic volume. As shown in
Table 6, the pavement constructed with 100% steel slag (Mix A) exhibited lowest rut depth.
TABLE 5 -- Input Parameters Required by VESYS Model
Seasons
Winter Spring Summer Fall
Resilient Modulus (MPa):

Mix A 20,700 3,800 990 3,800
Mix D 13,110 2,450 880 2,450
Mix F 16,800 2,400 950 2,400
Permanent Deformation Parameters:

p ct la ~t p Gt la
Mix A 0.023 0.720 0.038 0.508 0.208 0.535 0.038 0.508
Mix D 0.056 0.758 0.123 0.533 0.822 0.592 0.123 0.533
Mix F 0.095 0.740 0.127 0.558 0.613 0.591 0.127 0.558
Fatigue Coefficient:
K1 K2
Mix A 10.23 X 10"14 5.00
Mix D 3.58 X 10-13 4.75
Mix F 3.01 X I0 -~~ 3.98
TABLE 6 -- Performance Modelling Results
Time (years)
5 10 15 20
Rut Depth (mm):

Mix A 5.6 10.2 13.5 16.0 18.1
Mix D 13.2 23.5 30.0 34.7 38.5
Mix F 10.0 16.2 20.2 24.6 28.0
Fatigue Cracking Index:

Mix A 0.37 1.87 3.74 5.61 7.50
Mix D 1.48 7.44 14.50 22.32 29.70
Mix F 0.78 3.55 6.30 9.45 13.40
Present Serviceability Index:

Mix A 4.23 3.73 3.35 2.98 2.72
Mix D 3.81 2.94 2.16 1.48 0.87
Mix F 4.10 3.36 2.85 2.51 1.93
The fatigue cracking index (a dimensionless parameter), is a function of fatigue

parameters (K 1 and K2), traffic loading and layer thickness. It provides an indication of the
amount of fatigue cracking over the service life of the pavement. The cracking index value of
between 1.0 and 1.5 indicates light; between 2.5 to 3.5, moderate; and, between 2.5 to 3.5,
severe surface cracking. Results in Table 6 indicated that pavement constructed with Mix A
had the lowest cracking index value and will experience severe cracking in approximately 10
years of service while Mixes D and F will have severe cracking in approximately 5 years of
service.
The PSI provides an indication of rideability of the pavement structure. Mix A

pavement will have a PSI value of 2.72 after 20 years in service while Mix D pavement will
reach a PSI value of less than 2.5 in 10 years. In other words, pavement sections built with
100% steel slag will provide a considerably longer service life than that of the other sections.
It remains to be seen whether this improvment in performance can be measured in the field.
SUMMARY AND CONCLUSIONS
Based on results of advanced testing and modelling of steel slag aggregate, the
following conclusions can be drawn:
1. An aggregate conforming to accepted asphalt concrete standards can be produced.

2. Asphalt concrete mixes containing steel slag aggregates exhibited superior qualities
compared to those of conventional asphalt concrete mixes widely used in the province of
Nova Scotia.
3. Steel slag aggregate, when properly designed into asphalt concrete, can enhance the
resistant properties to moisture damage or stripping.
4. Steel slag aggregate also increased the fatigue life of asphalt concrete mixes.
5. Steel slag aggregate in asphalt concrete mixes generally indicated an improvement in
permanent strain and creep characteristics, indirect tensile strengths, resilient modulus, and
fatigue life.
6. The combination of the above make steel slag a viable alternative aggregate.
REFERENCES
[1] Schmidt, R.J., "A Practical Method for Measuring the Resilient Modulus of Asphalt
Treated Mixes," Highway Research Record, No. 404, Highway Research Board,
Washington, D.C., 1972, pp. 22-32.
[2] "Predictive Design Procedures, VESYS User Manual," Report No. FHWA-RD-77-154,
Federal Highway Administration, Washington, D.C., 1978.
[3] Anderson, K.O., and Hahn, W.P., "Design and Evaluation of Asphalt Concete with
Respect to Thermal Cracking," Proceedings of the Association of Asphalt Paving
Technologists, Vol. 37, 1968, pp. 1-31.
[4] Small, K.D., "Evaluation of Procedures Used to Predict Moisture Damage in Asphalt
Mixtures," Report No. FHWA-RD-86-000, Federal Highway Administration, Washington,
D.C., 1986.
[5] Chart, J.S.S., Taylor, B., All, N., Bergan, A.T., and Stamatinos, G., "Investigation of
Resilient Modulus Behaviour as Part of Flexible Pavement Design in Saskatchewan,"
Proceedings of Canadian Technical Asphalt Association, Vol. 34, November 1989,
pp. 194-205.
Fouad M. Bayomy I
DEVELOPMENT AND ANALYSIS OF C E M E N T - C O A T E D AGGREGATES FOR ASPHALT

MIXTURES
REFERENCE: Bayomy, F.M., "Development and A n a l y s i s of C e m e n t - C o a t e d

A g g r e g a t e s for A s p h a l t Mixtures," Effects of A q q r e q a t e s and Mineral
Fillers on Asphalt Mixture Performance, ASTM STP 1147, R i c h a r d C.
Meininger, editor, American Society for Testing and Materials,
Philadelphia, 1992.
ABSTRACT: Aggregate quality has a central role in the p e r f o r m a n c e of hot

mix asphalt concrete (HMAC) mixtures. The a v a i l a b i l i t y of good q u a l i t y
aggregates is of concern in many countries including the U.S.A. One of
the solutions to the aggregate depletion p r o b l e m is to u p g r a d e and
improve the properties of the available abundance of marginal
aggregates. This paper introduces a new aggregate u p g r a d i n g technique
for the use in HMAC mixtures. The proposed technique is referred to as
"Cement Coating Technique" (CCT). The CCT addresses the surface texture
of aggregate particles and improves the bond between the p a r t i c l e
surface and the asphalt binder. The concept of the CCT is based on
shielding the surface of the aggregate particles by a hydrated Portland
cement film. The new cement coated aggregates can then be used in
asphalt mixtures following the usual mix procedures. Asphalt concrete
mixes made with cement coated aggregates are referred to as "CEMPHALT"
mixtures. Cement content, water cement ratio and hydration time were
found to be the three main parameters influencing the coating process.
The o p t i m u m values of these parameters were determined. Performance
e v a l u a t i o n of CEMPHALT mixtures indicated significant improvements
achieved in permanent deformation, fatigue and m o i s t u r e damage
resistance.
KEYWORDS: aggregates, asphalt, cement, upgrading, coating, pavement,

rutting, fatigue, moisture damage, low quality aggregates.
The issue of pavement performance has always evolved when dealing

with pavement design and maintenance strategies. The solution to this
continuing question can usually be determined t h r o u g h a good
u n d e r s t a n d i n g of the processes and mechanisms of different modes of
pavement distress. In the process of improving the structural design of
asphaltic pavements, fracture and distortion, which are l o a d - a s s o c i a t e d
distress modes, have been considered as the principal modes of pavement
distress under the prevailing traffic and e n v i r o n m e n t a l conditions.
Disintegration, stripping and ravelling were e m p h a s i z e d as a m e c h a n i s m
associated with material rather than structural design c o n s i d e r a t i o n s
IAssociate Professor, Department of Civil Engineering, University

of Idaho, Moscow, ID 83843.
19
Copyright by
Copyright 9 ASTM
1992 Int'l
(all rights
by ASTM
reserved); Thu Dec 22www.astm.org
International
07:29:10 EST 2011
[i]. With respect to mix design aspects, an improvement of mix strength

and resistance to deformation would be gained by improving adhesion of
the asphalt binder to aggregate surface and by increasing the internal
friction of the aggregate matrix. The adhesion and internal friction are
by large affected by the particles surface texture. In the asphalt-
aggregate system, the interaction b e t w e e n asphalt b i n d e r and aggregate
surface has a central role on the overall p e r f o r m a n c e of the mixture.
The objective of this research was to develop an aggregate

treatment method that would increase the surface roughness of the
aggregate particles and improve the adhesion to asphalt binders. The
anticipated benefits would be an improvement in the mix stability and in
the resistance to deformation and moisture damage.
FACTORS INFLUENCING AGGREGATE-BINDER ADHESION
Early investigations on the mix design o p t i m i z a t i o n for adhesion

have m o s t l y dealt with the rheology of b i n d e r - a g g r e g a t e interaction and
theories of adhesion between bituminous binders and aggregate surfaces
were developed [2]. A recent SHRP research at Auburn U n i v e r s i t y [3] on
the m e c h a n i s m of adhesion of asphalt binder and aggregate surface is in
progress. This research is expected to present better u n d e r s t a n d i n g of
the bond c h a r a c t e r i s t i c s of a s p h a l t - a g g r e g a t e system specially w h e n
polymer m o d i f i e d asphalt is used. Early research on the stripping
p h e n o m e n o n has revealed several theories of adhesion m e c h a n i s m s b e t w e e n
asphalt binder and aggregate surface [4 through 12]. Jiminez [13]
e m p h a s i z e d the seriousness of the stripping p h e n o m e n o n using examples of
complete failure of a pavement within two weeks after opening to traffic
in one case, and a failure by loss of surface fines over a period of two
years in another case in Arizona. Fromm [i0], L o t t m a n [ii], and Ford
[12], have summarized the factors that affect the stripping p h e n o m e n o n
to be: water, traffic, cool temperatures, low asphalt content, low
asphalt viscosity, high air voids, cleanliness of fine aggregates,
coating (dust) on aggregates, composition of aggregates, and surface
texture. Rice [14] has investigated the r e l a t i o n s h i p b e t w e e n aggregate
and binder in the presence of water. He attributed the stripping
phenomenon mostly to the hydrophilic and hydrophobic surface
c h a r a c t e r i s t i c s of aggregates.
The available methods of improving adhesion b e t w e e n aggregate

surface and bitumen film lie in two categories: either by t r e a t i n g
aggregate surfaces or by reducing the surface t e n s i o n of the bitumen
binder by suitable a d h e s i o n - i m p r o v i n g agents. M a j i d z a d e h [2] summarized
the available agents and additives that were used in bituminous
mixtures. Cady [15], in his research on upgrading low-quality
aggregates, identified the aggregate problems that are amenable to
upgrading in bituminous mixtures to be stripping, d e g r a d a t i o n and
absorption. The approach he recommended was through treatment of
aggregates rather than bitumen. Coating and impregnation techniques
were recommended. Ishai [6] concluded that adjusting filler content in
the mix might improve the adhesion and he suggested that the use of
portland cement as a filler would improve adhesion. It has also been
noted that 1 to 2% portland cement was used in B r i t i s h p r a c t i c e to
reduce stripping [16]. Swanson [17] used the hydrated lime as an anti-
stripping additive in asphaltic mixtures. G a d a l l a h [18] used p o r t l a n d
cement to improve the characteristics of asphalt emulsion t r e a t e d mixes;
his results showed increase in stability with addition of cement.
The cement coating technique (CCT), presented in this paper, was

o r i g i n a l l y developed and implemented in Kuwait [19 and 20]. The
technique is b e l i e v e d to present a solution to the stripping p r o b l e m as
well as other pavement distress problems that are m a i n l y a t t r i b u t e d to
aggregate characteristics.
BAYOMY ON CEMENT-COATED AGGREGATES 21
METHODOLOGY AND CONCEPT
Most of the concepts presented in the literature, that describes

the m e c h a n i s m of adhesion between aggregate surface and bituminous
binder, indicate that a porous and rough-textured p a r t i c l e surface
develops good mechanical interlock that w o u l d certainly promote adhesion
and retard stripping [4 through 17, 21 and 22]. This p r o m o t i o n of
adhesion is not difficult to understand, from an e n g i n e e r i n g viewpoint,
nor is there a question that such roughness leads to a sizeable increase
in the particle internal friction which, in turn, m o d i f i e s the shear
strength and increases the load-bearing capacity of the asphalt-
aggregate system.
Pursuing this intuitive engineering approach, the author developed

the concept of using portland cement to create a r o u g h - t e x t u r e d surface
on the aggregate particles; that is, if it is possible to shield the
aggregate particles in a portland cement film and let this film adhere
to the aggregate surface by allowing appropriate time for hydration, the
hydrated cement film would create a rough textured surface on the
aggregate particles. The new cement-coated particles can be used in
bituminous concrete mixtures as usual. This concept is presented in
Figure i.
FIG. 1 -- Cement Coating Concept
It is believed that pre-coating would remedy the stripping p r o b l e m and

reduce the distress potential in the pavement due to an increase in the
particle internal friction and increased bond strength of the binder to
the aggregate surface as well. In accordance with this concept, the
author used "Portland cement" as the treatment material and "coating" as
the treatment method in order to derive the final procedures for the
cement coating technique (CCT).
Coating Criterion: The aggregate particles are to be coated by a

hydrated cement film which is thick enough to p e r m a n e n t l y shield the
aggregate particle surface but not so thick as to cause sticking among
particles and form concrete lumps.
To reach an objective d e t e r m i n a t i o n of the cement coating process,

following the a b o v e - m e n t i o n e d criterion, three parameters need to be
d e t e r m i n e d at their optimum:
i. percent of cement to be added for each type and size of

aggregate.
2. Water content needed for cement hydration; and
3. M i n i m u m hydration time needed to achieve permanent bond of
the cement paste coating film to the p a r t i c l e surface.
The basic scheme employed in the development of the coating process

was to incorporate the coating criterion defined above with a mix design
m e t h o d (for example, the Marshall method, ASTM D1559) to optimize the
p a r a m e t e r s of the above mentioned coating process.
DEVELOPMENT OF THE COATING PROCESS
Development Staqes
A l t h o u g h the cement coating process appeared easy from the

conceptual viewpoint, a very lengthy e x p e r i m e n t a l p r o g r a m and various
trials were required before it was possible to effect it. The
procedural d i f f i c u l t i e s stemmed from the different factors related to
the process, some of which are: aggregate size distribution, type of
aggregate particles, natural or crushed stones, fines content, air
voids, compaction,.., etc. These factors and others, like m i x i n g
t e m p e r a t u r e and type of compaction, had a significant effect on the
o p t i m i z a t i o n parameters. In addition, the question of how the process
was going to be implemented in field practice was one of the major
d i f f i c u l t i e s encountered throughout the development stages. Figure 2
outlines the m a j o r factors considered in the experiment design of the
development stages of the C C T process. Marshall stability test (ASTM D
1559) and i m m e r s i o n / c o m p r e s s i o n test (ASTM D 1075) were the basis for
parameters' o p t i m i z a t i o n while the coating criterion for efficient
coating and n o n - s t i c k i n g particles was observed in all stages. Modulus
of resilience (ASTM D 4123), loss of stability and split tensile
strength tests were performed for mixtures evaluation. The laboratory
development of the CCT procedures were achieved in four stages. The
factors investigated and the conclusions achieved in each stage are
summarized in Figure 3. The detailed testing p r o g r a m of these stages is
explained by the author in Chapter 3 of Reference [19]. Results were
categorized into two groups; group (a) involved the visual tests (Table
i) and group (b) involved the mechanical properties d e t e r m i n e d for each
mix in each stage. The visual tests involved observing two indicators in
the coated particles; a) efficiency of coating and u n i f o r m i t y of the
coating film, and b) presence of sticking particles.
The size of the data base of the mechanical properties is too large to
be presented in this paper. However, results achieved in the final
development stage are presented here as part of the e v a l u a t i o n of the
d e v e l o p e d technique. Analysis of both visual and m e c h a n i c a l test results
revealed the following conclusions :
a. O p t i m u m W a t e r C o n t e n t : It was found that w a t e r / c e m e n t ratio of 0.2

was the optimum. Water required for aggregate absorption to reach
s a t u r a t e d - s u r f a c e - d r y (SSD) condition should be added. The total
amount of water in the mix should be equal to that d e t e r m i n e d for
cement hydration (0.2 w/c) plus that d e t e r m i n e d for aggregate SSD
condition. It is expected that additional water might be needed in
the field to compensate for losses due to e v a p o r a t i o n and other
field related factors. In case of excess water in the aggregate
stock pile, adjustment should be made for the total water content.
,._[Aggregate 1 Cement Hydration
IType & Size/ Addition Time Compaction Optimizatir
(Days) Tests
,__ Coarse
1/2 - 1" I
(12.7- I Visual
25.4 mm)
Evaluation
130
0
__/1/4
C~ - 1/2" 1 ~ Samples' -<
/ (6.35- Requirements 0
12.7 mm) Z
t Agg. Evaluation (no1) 192 0
m
Coarse Marshall S t a b i l i t y (n-3) 576 m
-- ~ 1/4" Pi z
Imm./Comp. (n-6) -i ID .-I
<6.35mm 1152 c5
0
Split Tensile (n-3) 576
m
Natural Modlus of Resilience (n-3) 576 EJ
: / >
"- Sand ',_
. . . . . . . . . . . . . . . ,. . . . . . . . . . . . . . L . . . . . . . . . . . . . . . ,. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . j . . . . . . . . . . . . . . . . . . . . . . . . ,
-11
1"11
G3
@ rT1
r
FIG. 2 - - Investigated Parameters in the Development of the Cement Coating Technique
r~
GO
CopyrightbyASTMIntl'(allrightsreserved);ThuDec2207:29:10EST2011
Downloaded/printedby
IndianInstituteofTechnologyKharagpurpursuanttoLicenseAgreement.Nofurtherreproductionsauthorized.
DEVELOPMENT STAGE FACTORS INVESTIGATED ACHIEVED RESULTS

:::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
::::::::::::::::::: ::::::::::::::::::::::::::::::::::::::
::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::::
iiiiiiiiiiiiiiiiliiiiiiiiiiiii
iiiiiiiiiiiiiiiiiiiii,iiiiiii
ii 1 Sieve Analysis;
Coating all Aggregates] Marshall Stability; Air
Including Filler ] Voids; Bulk Density
Gradation, Density,
I
and Air Voids
were not accepted
ii!i!!ii!i!i iiiiiiiiiiiiiiiiii!iii!il Fines content; Water/Cement

Solving Density & ratio; and Compaction,
Gradation Problems Type and Effort
4% cement with 0.2 W/C
I
ratio; add filler after
hydration; Increase
compaction.
iiiiiiiiiiiiiiiiiiiiiiiiii
iiiiiiiiiiiiii!iiiiiiiiiiil
iii
Individual Coating & [ _ _ Coating Coarse or Fine Agg.;
Condition of Cement ] Cement, Powder or Morter
Addition J
Both Coarse and Fine Aggrg
are to be coated; Cement
added either powder or
morter, W/C = 0.2 to 0.3
iiiililililililililiiii~iii~iiiiiiiiiiiiiiiililililil
1 Cement Content for each
size of aggregate; coating
in bulk media; Final [ efficiency and sticking
Optimization. ] potential.
Cement Content Formula; [

Optimum W / C Ratio; ]
Minimum Hydration Time;[
iiiiiiiiiii!iiiiiiiiiiiiiiiiiiiiiii;iiiiiiiiiii!ii!i;iii!iiiiiiiiiii!iiiiiiii;iiiiiiiiiii Coating Individual Sizes
I in bulk media.
Aggregate Gradation, Bond Breaking
Abrasion; Soundness
Mix Design Stability; Mr; ITS;

Moisture Damage
Performance Stripping; Fatigue;

Permanent Deformation
FIG. 3 -- DevelopmentStages and Evaluation o f C C T Process
TABLE 1 -- isual T e s t R e s u l t s for C o a r s e and F i n e C o a t e d A g g r e g a t e s
Type Cem. W/C Hydr, Visual E v a l u a - Type Cem. W/C Hydr Visual E v a l u a - Type Cem. W/C H y d r . I Visual E v a l u a - Type C e m . W / C ]Hydr.i Visual E v a l u a -
and C o n ' t Ratio T i m e tion o f C o a t i n g and C o n ' t R a t i o T i m e ! tion o f C o a t i n g and C o n ' t Ratio T i m e tion o f C o a t i n g and C o n ' t Ratio T i m e tion o f C o a t i n g
Size % (Days and Sticking Size % (Day.' and Sticking Size % (Days and Sticking Size % (Day~, and Sticking
I[::!!:!!!:i!iii!ili~ 2 Poor =i!i!i:!!:!!:::: :':'= ' 2 Poor |:i:i:!:i:i:~:!:i:!:;:!:!:~| I 2 Poor
. ..:i~ii!i~i~i!iii!i!i!~!i!,
i 0.15 42 Poor
Poor ::::::::::::::::::::::::: 0.15 4 Poor ======================
===::::::::::::::::::::::::::: 0.15 4 Poor O. 15 4 Poor
!i!i!i:!i:!i:!i::!i
7 Poor )i::i)::~i~)if:i::::i!i 7 Poor =.:.:.:.:.:.:.:.:.::.:.:.:4 , 7 Poor =:'::::ii::iiii::ii: ' 7 Poor
:::::::::::::::::::::::::::
Poor ~i~,~,~i~i~,:i~:ili~,~i 2 Poor ==='================= ' 2 Poor .... ::!iiii::!iiiiiiiiiiiili!~
:::::::::::::::::::::: , 2 Poor
1.0% 0,20 4 Poor :':'::::: 3.0% 0.20 4 Poor = !i!i!i!i~i!!:~!:
!i!::: 5.0% 0.20 4 Poor i:@::i::iii!::iii~!)
5,5% 0.20 4 Poor
:::::::::::::::::::::::
7 Poor ii!:=ili::iiiii::=:i==i=:iil 7 Poor ==:=:=================:= ' 7 Poor
2 Poor ::::::::::::::::::::::::::::: 2 Poor, T,S.
2 P.... TS iiiii!!iiiiiii!iiiii!i 2 Poor, Ts
0.30 4 Poor ====================== 0.30 4 Poor, T.S. _ ....................... 0.30 4 Poor, T.S, .... iiiiiiii
7 Poor :i:;:!:!:!:!:!:!:!:~:!
!
7 Poor, T.S. I';i!i;i!ilil)::)iiii::ii: 7 Poor, T.S. 7 Poor, T.S.
2 Poor i ] 2 Poor :::::::::::::::::::::::: 2 Pair ii::ii::!i!i::!i
==== ===================
=::::::::::::::::::::: ============================ 2 Good
0.15 4 Poor ~iii:~ii'ii:i::ii i!ii:iiii! 0.15 4 Poor =:+:.:+:+:,:: :
.:,:.::+:::.::+:a 0.15 4 Pair ii!iiiiliiii!iii!i!!i
.:+:+:,>:.:+:, o.15 4 G~
7 Poor ::::::::::::::::::::::::::::::: [3O
7 Poor ========================= 7 Pair 7 Good
2 Fair =
l+:.:+:+:
=:=========:+:.: =====
=====:.:~ 2 Good 2 Good
1.5% 0.20 42 Pair
Fair iiiiiiiiiiiiiiiiii
+:.:.:.:::: 3.5% 0.20 O
4 Fair =:.:,:.:,:.:.:.:.:.:.:
+:+:+:+::: :.:.j 5.5% 0.20 4 Good 6.0% 0.20 4 Good
7 Fair ::;::::::::::):::::
.......... 7 Pair .:,:,:.::.::::: 7 Good -<
i~;iiiiiii!iiiiiilil
:+x.::+>::: 7 Good
2 Pair, T.S. i:iii!iii!!i!!iiiiii!i:i!::i::::
::: 2 Fair, T.S. ========================= 2 Good, T.S. ~iii!i:!i!i:!!!ii!!
.:.:+:.:.:+:.:.: 2 Good, T.S. O
Coarse 0.30 4 Fair, T.S. Coarse 0.30 4 Fair, T.S. Coarse 0,30 4 Good, T.S. 0.30 4 Good, T.S. Z
Aggr. 7 Fair, T.S. Aggr. 7 Fair, T,S. Sand. 7 Good, T.S. Fine 7 Good, T,S. O
3/4" 2 Fair 3/8" 2 Fair 3/16" 2 Good Natural 2 Very Good m
(19.1 0.15 4 Fair (9.53 0.15 4 Fair (4.76 0,15 4 Good Sand O. 15 4 Very Good I'n
turn) 7 Fair ram) 7 Fair mm) 7 Good 7 Very Good Z
2 Good :::::==:: ::::::::::::::::::::::::::::::: 2 Very
6
02o 4 40% o 4 Very Good 6.0%
::::::::::::::::::::::::::::::::: 0.20 4 Excellent 6.5% 0.20 4 Very Oood O
7 Good i!~::!i!i:!!ill
:'i!i!i!!i::~ii!i!i
!i!i 7 Very Good -'i:!:i:!:!:!:
:i:!::!:i:i:1 7 Excellent 7 Very Good B
2 V. Goo,~, T.S. t~ii!:~!:!ii=;!i::==t Excellent, Sticky e:::ii!!i i)::~i~))):i 2 W Good, T.S. m
0.30 4 V. Good, T.S, 0.30 t2
4 v. Good, T.S. .)::I!:.::::::::::::~I~L 0.30 Excellent, Sticky :!::::i:!:i:l:i:::::?:::i:i:i:::::i:::::i:::i!i 0.30 4 V, Good, T.S.
7 V. Good, T.S. >
7 V. Good, T.S, ":?!::'~!!::::~:i" 7 V. Good, T.S.
2 Very oood =========================================
2 Exoe,lon,,Stiekyo
!iiiii{i!!iiiiiiiiiii 2
ii!iii?i~iii!i:!ii~!
0.15 4 Very Good :::::::::::::>:::: 0,15 4 Good -:~:::::~::::~. 0,15 20
iiiiii;i,ii,t o.154 oood r'n
7 Very Good 7 oood -::ilili::iliiiiiiilili::i~ 47 Good
tiiii.i:.:.:.:.:.:.:.:.:.: 7 Oood G3
2 Excellent i!iiii!iii!ii::!i
!i!!:i:!: 2 Excellent =,:.:,::.:.:.:.:.:.:.:.:
:~
2
4
v~
Very Good
iiiiiiiiiii!iiii!i!iiiiiili
7.0 % 0.20
24 Good m
2,5% 0.20 4 Excellent i!ii!ii!iiiii!!!:ii:
iiiii:
!iiiiiii4.5% 0.20 4 Excellent ii::!iiiiiiiiliiiii::iiii
6.5%: 0.20
G9
7 Excellent E:i:i:i:i:i:i:i:!:!:!:!:!:!l
72 Excellent,Excellentstickyiiiii!!U!iiii!iiii!!iiiiii!iiiiiiiii
2 Excellent, Sticky ='!i!i:!~!i!i:ii::::"
72 V. Good, Sticky [iii
i!i!iiii!
]
72 Good, Sticky
0.30 4 Ex0cl~ent, Sticky ~i~i~:i~!~i:~;~i::!::~:: 0.30 4 Excellent, Sticky i!i!iii!iii!i!ii!!iiiii! 0.30 4 V. Good, Sticky 0.30 4 Good, Sticky
7 Excellent. Sticky iiiiiiiiiiiiiiiiii!iiiiii 7 Excellent, Sticky [i!::::ii!::!::::;ili::iiiill 7 V. Good, Sticky 7 Good, Sticky
Note: T.S. = Tends to Stick O'l
b, H y d r a t i o n Time: In t h e l a b o r a t o r y , t w o d a y s w e r e f o u n d to be
m i n i m u m to a l l o w for c e m e n t h y d r a t i o n and to i n s u r e p e r m a n e n t
a d h e s i o n of t h e c e m e n t f i l m on p a r t i c l e s u r f a c e s .
C~ C e m e n t C o n t e n t : E x t e n s i v e test r e s u l t s o b t a i n e d p r o v e d t h a t a m o u n t
of c e m e n t n e e d e d is a f u n c t i o n of a g g r e g a t e size. A rational
a p p r o a c h w a s a d o p t e d to d e t e r m i n e c e m e n t c o n t e n t for e a c h a g g r e g a t e
size by a s s u m i n g a h o m o g e n o u s a g g r e g a t e c o m p o s i t i o n of s p h e r i c a l
p a r t i c l e s w i t h d i a m e t e r (d), and c o a t e d by a t h i n c e m e n t film of
c o n s t a n t t h i c k n e s s (t) (Figure 4). T h e n t h e w e i g h t of c e m e n t
r e q u i r e d to coat (n) p a r t i c l e s in a u n i t v o l u m e is:
n particles in Particle diameter after coating

a unit volume d+2t
--\ /
_ _
Cement Film
Thickness (t)
L 1 _-
FIG. 4 -- Idealization of C e m e n t Content Computation
Wc = nyc~d2t (i)
Weight of d r y a g g r e g a t e particles in a u n i t volume:
Wag = nYd~d3/6 (2)
Then a theoretical cement content, C O, can be e s t i m a t e d by:
C~ = (nYc~d2t) / (nYd~d3/6) (3)
where: u n i t w e i g h t of p o r t l a n d cement,
Yd = a p p a r e n t u n i t w e i g h t of d r y a g g r e g a t e particles,
and d = a g g r e g a t e p a r t i c l e size
Then: C O = kt/d, where k = 6Yc/y d (4)
If a c o n s t a n t v a l u e of t h i c k n e s s (t) is assumed, then a cement

c o n t e n t (C) c a n be d e t e r m i n e d by:
C = E/d (5)
where: K = kt = (6yc/Yd)t
It w a s i m p o s s i b l e to d e t e r m i n e c e m e n t f i l m t h i c k n e s s , t h e r e f o r e ,
a r b i t r a r i l y c h o s e n v a l u e s of c e m e n t c o n t e n t w e r e u s e d in p r e p a r i n g
cement-aggregate m i x e s of d i f f e r e n t sizes of c o a r s e a g g r e g a t e s .
These mixes were visually evaluated following the coating

c r i t e r i o n . T h e v i s u a l t e s t r e s u l t s a r e p r e s e n t e d in T a b l e i. T h e
c e m e n t c o n t e n t v a l u e s for t h e g o o d c o a t i n g w i t h no s t i c k i n g
c o n d i t i o n w e r e c o r r e l a t e d to t h e a g g r e g a t e s i z e (d) . A l i n e a r
f i t t i n g b e t w e e n (C) and Log(d) w a s p e r f o r m e d in t h e form:
C = C O + S Log(d) (6)
where C = % c e m e n t of d r y a g g r e g a t e w e i g h t ,
d = a g g r e g a t e s i z e in mm,
C O = i n t e r c e p t {% c e m e n t n e e d e d to c o a t 1 mm aggregate
size, (d = 1 mm)}, a n d
s = s l o p e of t h e r e g r e s s i o n l i n e
Figure 5 shows the relationship between the aggregate size and

c e m e n t c o n t e n t . T h e r e g r e s s i o n c o n s t a n t s w e r e f o u n d to be: C O =
1 0 . 4 2 3 and s = - 6 . 3 3 7 5 w i t h r e g r e s s i o n c o e f f i c i e n t of
d e t e r m i n a t i o n (r 2) = 0 . 9 3 2 4 6 .
Cement Content, %
7 ...... : 0--~:~ 0 : ...........;...........
6".. ;
6 . . . . . . . . :o.: ~-. ................ .............
..... ..o "'..

5 Regression Line ............ ......
4 ~-:;;~ Lower Boundary . "i;..~"-.

r . . . . . ~ ........
--eli. Upper Boundary "'"'o'"'-
3
1 i t i i i t i i i
1 3 5 10 20
Aggregate Size (d}, mm
FIG. 5 -- Relationship of C e m e n t Content and Aggregate Size
T h e a b o v e e q u a t i o n w a s f o u n d v a l i d in t h e r a n g e (4.1 m m ~ d S 2 5 . 4
mm). F o r s a n d s w h e r e (d < 4.1 mm), a v a l u e of 6% to 7% c e m e n t
c o n t e n t was f o u n d to be a d e q u a t e . Any additional cement would
r e s u l t in f o r m a t i o n of l u m p s a n d p a r t i c l e s t i c k i n g .
Recommended Procedure
i. D e t e r m i n e t h e w a t e r c o n t e n t (w,) of t h e v a r i o u s a g g r e g a t e f r a c t i o n s
t h a t n e e d to be c o m b i n e d to p r o d u c e t h e r e q u i r e d g r a d a t i o n .
2. D e t e r m i n e w a t e r n e e d e d for a b s o r p t i o n ( s a t u r a t e d s u r f a c e d r y
c o n d i t i o n ) , w a-
3. F o r e a c h f r a c t i o n of c o a r s e a g g r e g a t e , d e t e r m i n e t h e n e e d e d c e m e n t
c o n t e n t (in p e r c e n t of d r y w e i g h t ) . C e m e n t w e i g h t = C x A g g r e g a t e
weight.
4. C a l c u l a t e w a t e r c o n t e n t n e e d e d to be a d d e d to t h e c e m e n t - a g g r e g a t e
m i x as follows:
W e i g h t of w a t e r = 0.2 x C e m e n t w e i g h t + (w a - Wa) x A g g r e g a t e
weight.
5. M i x a g g r e g a t e s w i t h c e m e n t (dry) for about o n e - t h i r d of t h e t o t a l
m i x i n g t i m e and t h e n a d d water; c o n t i n u e m i x i n g u n t i l a h o m o g e n e o u s
m i x is a c h i e v e d (mixing t i m e is d e p e n d e n t on m i x e r t y p e a n d w e i g h t
in e a c h batch).
6. K e e p the m i x e d a g g r e g a t e s for at least t w o d a y s in a h u m i d i t y -
temperature controlled environment.
7. O v e r a l l p e r c e n t a g e of c e m e n t c a n be c a l c u l a t e d b a s e d on t h e p e r c e n t
of c o m b i n a t i o n s for e a c h fraction.
8. F i n e c r u s h e d and n a t u r a l sands can be c o a t e d in t h e same w a y b u t
using a cement content ranging between 6 to 7%.
9. T h e c o a t e d a g g r e g a t e f r a c t i o n s are to be c o n s i d e r e d as n e w
a g g r e g a t e s . B l e n d i n g and m i x i n g p r o c e d u r e s for m i x d e s i g n is to be
p e r f o r m e d as usual.
EVALUATION
Aqqreqate Evaluation
Conceptually, c e m e n t c o a t e d a g g r e g a t e s are c o n s i d e r e d n e w
a g g r e g a t e s , as if t h e y w e r e b r o u g h t f r o m a n e w source. Therefore,
r o u t i n e t e s t s for a g g r e g a t e e v a l u a t i o n t h a t are u s u a l l y u s e d in p a v e m e n t
c o n s t r u c t i o n are p e r f o r m e d . M a j o r t e s t s w i l l include: sieve analysis,
a b r a s i o n and soundness. R e s u l t s of t h e s i e v e a n a l y s i s i n d i c a t e d that,
t h e a g g r e g a t e b l e n d b e c a m e c o a r s e r and m o r e m i n e r a l f i l l e r w o u l d be
r e q u i r e d to s a t i s f y t h e % fines (% p a s s i n g # 2 0 0 sieve) in t h e g r a d a t i o n .
A b r a s i o n t e s t s u s i n g t h e Los A n g e l e s a b r a s i o n test (ASTM C 131) for
coarse aggregates indicated better abrasion resistance (Table 2).
H o w e v e r , d a t a in T a b l e 2 r e p r e s e n t o n l y o n e s a m p l e f r o m e a c h a g g r e g a t e .
On a s t a t i s t i c a l b a s i s t h e r e m a y be no d i f f e r e n c e in t h e a b r a s i o n
r e s i s t a n c e as m e a s u r e d by t h e Los A n g e l e s a b r a s i o n test. D u r a b i l i t y of
t h e c o a t e d a g g r e g a t e s w a s d e t e r m i n e d by m e a n s of t h e s o u n d n e s s t e s t
(ASTM C 88). R e s u l t s (Table 3) s h o w e d t h a t c o a t i n g c o a r s e a g g r e g a t e s
improved, significantly, its r e s i s t a n c e to c h e m i c a l d e g r a d a t i o n . F i n e
a g g r e g a t e s i n c l u d i n g n a t u r a l sand d i d not s h o w s i g n i f i c a n t d i f f e r e n c e .
TABLE 2 -- Los A n g e l e s Abrasion Test Results (ASTM C 131)
Test Aggregate Size and % Loss due

Condition Type Grading* to Abrasion
Cement Coated 3/4 " (19.1 ram), (B)* 15.60
Dry 3/8 " (9.52 ram), (C)* 17.44
Aggregate Uncoated 3/4 " (19.1 ram), (B)* 16.71
3/8 " (9.52 ram), (C)* 19.62
Cement Coated 3/4 " (19.1 ram), (B)* 15.00
Saturated** 3/8 " (9.52 ram), (C)* 17.57
Aggregate Uncoated 3/4 " (19.1 ram), (B)* Not Tested
3/8 " (9.52 mm), (C)* Not Tested
* Grading in accordance to (ASTM C 131) specification.

** Aggregate was immersed in 60 *C water for 24 hours, then dried for the test.
TABLE 3 -- Soundness Test Results (ASTM C 88)
Aggregate Aggregate % Loss due to Chemical Degradation in

Size Type Magnesium Sulfate Sodium Sulfate
3/4 " Uncoated 13.00 3.90
(19.1 ram) Cement Coated 0.50 1.50
3/8 " Uncoated 5.50 2.90
(9.52 mm) Cement Coated 0.60 0.30
3/16 " Uncoated 13.50 8.60
(4.76 ram) Cement Coated 15.00 6.60
Fine Natural Uncoated 6.50 10.40
Sand Cement Coated 6.30 27 10
Mixture Evaluation
To evaluate the effect of aggregate coating on the basic mix design

properties, a m i x t u r e conforming with Type III surface course, Kuwait
s p e c i f i c a t i o n s [23] was selected. The mix c o n t a i n e d a c o m b i n a t i o n of 20%
A g g r e g a t e 3/4-inch, 39% A g g r e g a t e 3/8-inch, 10% limestone screenings,
24% natural sand and 7% mineral filler. The o p t i m u m asphalt content,
d e t e r m i n e d by the Marshall m e t h o d of mix design, was 5.1% of total mix
weight for the m i x with cement coated a g g r e g a t e (CEMPHALT) and it was
5.0% for the mix with uncoated aggregate (Control) mix. The p r o p e r t i e s
e v a l u a t e d were stability and moisture damage at v a r i o u s ages. For the
m o i s t u r e damage, a third mix was considered where a control mix w i t h 1%
hydrated lime (HL) added in replacement of 1% mineral filler.
Results of Marshall stability and immersion t e n s i l e strength ratio

at d i f f e r e n t ages are shown in Figures 6 and 7 respectively. The results
showed an increase in the stability with age for all mixes, coated
(CEMPHALT) and u n c o a t e d (Control). This may have r e s u l t e d from b i t u m e n
h a r d e n i n g with time. However, it w a s o b s e r v e d that t h e r e was an increase
in the stability of CEMPHALT mix ranging from 25 to 50% more than that
of the control mixes d e p e n d i n g upon the original mix design of the
control mix, with or without hydrated lime. The immersion t e n s i l e
strength results showed, consistently, a gain in the immersed t e n s i l e
strength w h e r e the ratio of the wet strength to the d r y strength was
always higher than 100% for the CEMPHALT mix. This w o u l d be a t t r i b u t e d
to the continuous gain in mix strength due to the c o n t i n u a t i o n of the
h y d r a t i o n of cement film under water. This in turn indicates b e t t e r
r e s i s t a n c e to stripping based on the improvement in the immersion
tensile strength ratio achieved.
PERFORMANCE EVALUATION
P e r f o r m a n c e of the d e v e l o p e d C E M P H A L T m i x under t r a f f i c loads was

e v a l u a t e d in water damage by immersed wheel t r a c k i n g test, fatigue and
p e r m a n e n t deformation. A separate paper is being p r e p a r e d on the results
of the p e r f o r m a n c e e v a l u a t i o n program. A brief summary is p r e s e n t e d in
this p a p e r to clarify the effect of the cement c o a t i n g on the m a j o r
pavement distresses u s u a l l y e n c o u n t e r e d in asphalt mixes.
a. W a t e r Damage: The immersed wheel t r a c k i n g test was p e r f o r m e d in

accordance with the T R R L procedures [24]. A c c o r d i n g to the test
specification, if the mix does not fail w i t h i n 24 hours, it w o u l d be
accepted and considered safe for m o i s t u r e damage under t r a f f i c loads.
Results showed that CEMPHALT samples did not fail for a period of 48
h o u r s w h e r e t h e test; w a s t e r m i n a t e d . On t h e o t h e r h a n d s a m p l e s of t h e
c o n t r o l m i x f a i l e d w i t h i n 18 to 20 hours. T h e s e r e s u l t s a g a i n c o n f i r m e d
t h e c o n c l u s i o n s a c h i e v e d by t h e i m m e r s i o n c o m p r e s s i o n and t e n s i l e
strength tests; that the asphalt mix made with cement coated aggregates
w o u l d r e s i s t s t r i p p i n g a n d m i n i m i z e w a t e r d a m a g e e f f e c t on t h e s e m i x e s .
Marshall Stability, Kgs

2500
2000
1500
A
z
1000
A
500
Q CEMPHALT • CONTROL (with 1% HL) -•- CONTROL
I I I I I I II I ~ I I I I I rl I I I I I I I I
10 100 1000
AGE (DAYS)
FIG. 6 -- E f f e c t of A g i n g on M a r s h a l l Stability for C E M P H A L T and C o n t r o l

Mixes.
ITSR RATIO, %
140
120<
100
80
60
9 CEMPHALT x C O N T R O L ( w i t h 1% HL) ....
40 1 I I I ~ I ~II I I i I I I III I I f I I I II
I 10 100 1000
AGE (DAYS)
FIG. 7 -- M o i s t u r e Damage and A g i n g Effect on C E M P H A L T and Control
Mixes
BAYOMY ON CEMENT-COATED AGGREGATES. 31
b. Fatigue: Samples from the ~wo d e s i g n a t e d m i x e s were t e s t e d in

fatigue u s i n g the indirect tensile diametral r e p e a t e d load test. The
stress - number of cycles to failure r e l a t i o n s h i p s w e r e d e v e l o p e d for
the two m i x e s at two different t e m p e r a t u r e s (28 and 50 ~ t h e y are
p r e s e n t e d in Fig. 8. The fatigue life of the C E M P H A L T mix was
s i g n i f i c a n t l y larger than that of the control mix. This is a t t r i b u t e d to
the increase in mix stiffness w h i c h is b e l i e v e d due to the increase in
cohesion and friction in the mix matrix.
5OO
22 ?
Q.
-" ?"!. . . . "" 9 1 (2) 28 C
2O0
u) -L " ---,. -~i = "'''" lmmm
&-- 100
! ,i : " : " ~ - , ~ " .o~ " (2)50 C
~ le-.. o
~) 50
cO
c-
~)
t--
20
0
~ 10 ~:(:i)C~MPHALT::mi•
c-
5
0.1 0.3 1 3 10 30 100 3OO
Number of Cycles to Failure (Nf) x 1000
FIG. 8 -- Fatigue Test Results for C E M P H A L T and Control Mixes
c. R u t t i n g : Permanent d e f o r m a t i o n test (Rutting) was p e r f o r m e d on

cylindrical samples. The test p r o c e d u r e was similar to that of the
dynamic modulus test (ASTM D 3497) except that the test was c o n t i n u e d to
failure. Permanent strain was recorded with number of cycles. Fig. 9.
shows a typical result obtained at one selected stress level and one
selected temperature. Results at other t e m p e r a t u r e s and under different
stress levels showed the same behavior of C E M P H A L T mix with respect to
the control mix. For all tests performed, the o b s e r v e d p e r m a n e n t strain
of C E M P H A L T mix was not only less than that of the control but also the
failure of the control samples was m u c h earlier than those of the
C E M P H A L T mix. This w o u l d indicate that the CCT p r o c e s s has r e s u l t e d in
s i g n i f i c a n t increase in m i x resistance to p e r m a n e n t deformation.
Obviously, it was due to the increase in adhesion and internal friction
created by the coating process.
CONCLUSIONS
The d e v e l o p m e n t of the CCT process has been established. For o p t i m u m

coating, a cement content can be d e t e r m i n e d by m e a n s of e q u a t i o n (6).
The w a t e r content required was found to be equal to the amount needed
for aggregate absorption (SSD condition) plus the amount needed for
cement hydration. A 0.2 w a t e r / c e m e n t ratio was found to be optimum. Two
days were found to be m i n i m u m for cement hydration that will result in
cement film adhesion enough to prevent film breaking during mix
handling.
The laboratory evaluation of asphalt mixes made with cement coated

aggregates (CEMPHALT) indicated significant improvement in mix
stability, strength and resistance to moisture damage, fatigue and
permanent deformation.
0.1 (1) 5o c
--4)---
Y
E (2) 5o c
E 0.08 ::::::::::::::::::::::::::::::::::::::::::::: o
E A (1) 4o c
- - -4&- - -
E
._ 0.O6 ~ : 12) 40 C
m !
(1) 30 O
---i-- ~
9" - ' 0.04 . . . . . .
/
#
(21 30 C
|
$ []
0.02 t , m ~ ~1~. 9
Q e
0 ~ m
O.Ol O. 1 1 10 100 1,000
Number of C y c l e s x 1000
FIG. 9 -- Permanent Deformation Test Results for CEMPHALT and Control

Mixes
EPILOGUE
The cement coating technique became part of the general

s p e c i f i c a t i o n of road c o n s t r u c t i o n in Kuwait by 1984. All asphalt
pavements for expressways built in Kuwait since then were constructed
with cement coated aggregates as mandated by the specifications.
REFERENCES
[1] Monismith, C.L. and Finn, F.N., "Flexible Pavement Design: State-
of-the-Art," Journal of T r a n s p o r t a t i o n Enqineerinq, Vol. 103, No.
TEl, American Society of Civil Engineers, New York, NY, J a n u a r y
1977.
[2] Majidzadeh, K. and Brovold, F.N., "State of the Art: Effect of

Water on Bitumen Aggregate Mixtures," Special Report No. 98,
H i g h w a y Research Board, Washington, D.C., 1968.
[3] Jean, Y.W., and Curtis, C.W. "A Literature Review of the
Absorption of Asphalt Functionalities on Aggregate Surfaces,"
Report No. SHRP A/IR-90-OI4, Strategic Highway Research Program,
National Research Council, Washington, D.C., 1990.
[4] Douglas, J.F., "Adhesion Between Binders and Aggregates," Journal

of Institution of Civil Engineers, No. 3, England, January 1947,
p. 292.
[5] Gzemski, F.C., "Factors Affecting Adhesion of Asphalt to Stone,"

Proceedings of the Association of Asphalt Pavement Technologists,
St. Paul, MN, Vol. 17, 1948, p. 74.
[6] Ishai, I. and Craus, J., "Effect of the Filler on Aggregate-

Bitumen Adhesion Properties in Bituminous Mixtures," Proceedings
of the Association of Asphalt Pavement Technologists, St. Paul,
MN, Vol. 46, 1977.
[7} Ensley, E.K. and Sholtz, H.A., "An Investigation of Asphalt-

Aggregate Adhesion by Measurement of Heats of Immersion," Highway
Research Record, No. 340, Highway Research Board, Washington,
D.C.,1970. pp. 38-44
[8] Lottman, R.P. and Johnson, D.L., "Pressure Induced Stripping in

Asphaltic Concrete," Highway Research Record, No. 340, Highway
Research Board, Washington, D.C.,1970. pp. 13-28
[9] Lottman, R.P., "Predicting Moisture-Induced Damage to Asphaltic

Concrete," Report No. 192, National Cooperative Highway Research
Program, Transportation Research Board, Washington, D.C. 1978.
[10] Fromm, J.J., "The Mechanisms of Asphalt Stripping from Aggregate

Surfaces," Proceedings of the Association of Asphalt Pavement
Technologists, St. Paul, MN, Vol. 43, 1974, pp. 191-219.
[11] Lottman, Robert P., Chen, R.P., Kumar, K.S. and Wolf, L.W., "A
Laboratory Test System for Prediction of Asphalt Concrete Moisture
Damage," Transportation Research Record, No. 515, Transportation
Research Board, Washington, D.C., 1974, pp. 18-26.
[12] Ford, M.C., Manke, P.G. and O'Bannon, C.E., "Quantitative

Evaluation of Stripping by the Surface Reaction Test,"
Transportation Research Record, No. 515, Transportation Research
Board, Washington, D.C., 1974, pp. 40-51.
[13] Jiminez, R.A., "Testing for Debonding of Asphalt from Aggregates,"

Transportation Research Record, No. 515, Transportation Research
[14] Rice, James M., "Relationship of Aggregate Characteristics to the

Effect of Water on Bituminous Mixtures," American Society of
Testing and Materials, ASTM STP No. 240, 1958, pp. 17-34.
[15] Cady, P.D., Blankenhorn, P.R. and Kline, D.E., "Upgrading of Low-
Quality Aggregates for PCC and Bituminous Pavements," Report No.
207, National Cooperative Highway Research Program, Transportation
Research Board, Washington, D.C., 1979.
[16] Jacobs, F.A., "Properties of Rolled Asphalt and Asphaltic Concrete

at Different States of Compaction," Transport and Road Research
Laboratory, TRRL Supplementary Report No. 288, Crowthorne,
Berkshire, England, 1977.
[17] Swanson, E.G., "Use of Hydrated Lime in Asphalt Paving Mixtures,"
Presentation at the Annual Conference of the National Lime

Association, Las Vegas, Nevada, April 1963.
[18] Gadallah, A., Wood, L., and Yoder, E., "Effect of Portland Cement
on Certain Characteristics of Asphalt-Emulsion-Treated Mixtures,"
!ransportation Research Record, No. 712, Transportation Research
[19] Bayomy, F. "Development and Analysis of a Cement Coating

Technique: An Approach Toward Distress Minimization and Failure
Delay in Flexible Pavements." Ph.D. Dissertation, The Ohio State
University, Columbus, OH, 1982.
[20] Bayomy, F.M.S. and H.R. Guirguis, "An Upgrading Technique for AC
Pavement Materials," Proceedinqs of the International Conference
of Roads and Development, Volume 2, Paris, France, May, 1984.
[21] Kalcheff, I.V., "Research on Bituminous Concrete Properties with

large-size Aggregates of Different Particle Shape," Special Report
No. 109, Highway Research Board, Washington, D.C., 1968.
[22] Kalcheff, I.V. and D.G. Tunnicliff, "Effects of Crushed Stone

Aggregate Size and Shape on Properties of Asphalt Concrete,"
Proceedinqs of the Association of Asphalt Pavement Technoloqists,
St. Paul, MN, Vol. 51, 1982.
[23] "Specifications for Road Materials and Construction," Ministry of

Public Works, Roads and Drainage Department, Kuwait, 1980.
[24] "Bituminous Materials in Road Construction," Department of

Scientific and Industrial Research, Transportation and Road
Research Laboratory, England, 1962.
Gregory P. Bouchard
EFFECTS OF AGGREGATE ABSORPTION AND

CRUSH PERCENTAGE ON BITUMINOUS CONCRETE
REFERENCE: Bouchard, G.P., "Effects of Aggregate

A b s o r p t i o n and Crush P e r c e n t a g e on B i t u m i n o u s
Concrete", E f f e c t s of A g g r e g a t e s and M i n e r a l
Fillers on Asphalt Mixture Performance, A S T M STP
1147, Richard C. Meininger, editor, American
Society for Testing and Materials, Philadelphia,
1992.
ABSTRACT: Recently, the push to improve

bituminous concrete mixtures has r e s u l t e d in
higher fractions of crushed aggregate. Since
coarse aggregate is becoming more difficult to ex-
tract as gravel sources are depleted, the cost of
supplying crushed aggregate is increasing. Adding
too much crushed aggregate can produce two
m e c h a n i s m s for problem mixes. The first arises
from the crushing and washing of the manufactured
fines portion of the aggregate. Hammermill, im-
pact or cone crushers have a t e n d e n c y to p r o d u c e
g r a d a t i o n s for fine aggregates below the m a x i m u m
d e n s i t y line when washed. This can lower the
voids in mineral aggregate which is important to
resist rutting.
A theory is advanced that aggregate water ab-

sorption can be used to predict the b o n d i n g sur-
face volume available to improve results of heavy
traffic mixes. Let the bonding surface volume be
defined as the volume around each particle opened
by pores. The second m e c h a n i s m can be d e s c r i b e d
as the p r o p e n s i t y for shearing of m a s t i c to ag-
gregate bonds over time when aggregates w i t h low
water absorption v a l u e s are used. Marshall
stability tests for a mix using a g g r e g a t e s w i t h
generally high values of water absorption are com-
pared to tests for a mix using aggregates with low
values of water absorption. The c o n c l u s i o n sup-
ports the premise that shearing under load in-
creases as a v e r a g e water absorption values
decrease.
G r e g o r y P. Bouchard, P. Eng. is the a s s i s t a n t

m a n a g e r of gravel operations for Borland C o n s t r u c t i o n
(1989) Limited of 751 Lagimodiere Boulevard, Winnipeg,
Manitoba, Canada R2J OT8.
35
Copyright by ASTM Int'l (all rights reserved); Thuwww.astm.org
Copyright 9 1992 by ASTM International
Dec 22 07:29:10 EST 2011
KEYWORDS: asphalt cement, arterial mixes,

b i t u m i n o u s concrete, bonding surface volume, sur-
face shearing, stability, voids in m i n e r a l ag-
gregate.
The i m p o r t a n c e of a h i g h p r o p o r t i o n of c r u s h e d
p a r t i c l e s in a b i t u m i n o u s m i x t u r e b e c a m e e v i d e n t long
ago. The tendency for mixes to flow during hot summer
days and heavy traffic was noticeably increased when a
high p r o p o r t i o n of rounded glacial sand and stone were
used. Major service improvements were achieved with
the advent of high production stone crushers. The in-
c r e a s e of c r u s h e d p a r t i c l e p e r c e n t a g e , higher stone
c o n t e n t s and the r e d u c t i o n of the a s p h a l t c e m e n t p e r -
c e n t a g e to as low as 4.7% by total mix were some of the
r e a s o n s why. Mixes placed during the 1960's and early
seventies on a r t e r i a l routes in W i n n i p e g , Manitoba
e m p l o y i n g this latter design have shown r e m a r k a b l e per-
formance.
The g e n e r a l increase of r u t t i n g t h r o u g h o u t N o r t h
A m e r i c a recently, has p r e s s e d t e c h n i c a l a n a l y s t s to
find solutions quickly to the problem. During the late
1970's, a general mix design change took place in some
locations. A higher Voids in M i n e r a l Aggregate
(V.M.A.) to a c c o m m o d a t e a h i g h e r a s p h a l t c e m e n t p e r -
c e n t a g e was set as the goal. Q u i t e often, these
designs produced inferior mixes which p e r f o r m e d p o o r l y
o v e r the long term. One p o s s i b l e e x p l a n a t i o n is, to
a c h i e v e a higher V.M.A. percentage without the use of a
p r o p e r l y g r a d e d blend sand (clean b e a c h t y p e sand), a
h i g h e r p r o p o r t i o n of well g r a d e d sand was g e n e r a l l y
used. This decreases larger inter-particle contact and
d e v i a t e s the final blend farther from the more stable,
m a x i m u m d e n s i t y line. Typically, it is b e t t e r to
d e v i a t e the finer portion of the mix away from the max-
imum d e n s i t y line and not the entire blend. To ensure
that the mix is coarse enough for a 16 mm m a x i m u m sized
mix, at least 38% on average should be retained on the
5.0 mm sieve. Furthermore, it is s u g g e s t e d t h a t this
c o a r s e r p o r t i o n has a high crush count to a s s i s t f i r m
p a r t i c l e interlock. For the sand portion, care must be
t a k e n to o p t i m i z e the s e l e c t i o n of m a t e r i a l . Because
the cost of p r o d u c i n g c r u s h e d sand to a p r o p e r g r a d a -
tion is expensive, the economics of p r o d u c t i o n has to
play an important role.
BOUCHARD ON BITUMINOUS CONCRETE 37
The reason m a n u f a c t u r i n g 100% crushed b i t u m i n o u s

fines is costly could be because crushers have a ten-
dency to produce gradations for fine a g g r e g a t e s below
the maximum density line after washing. Since the fine
a g g r e g a t e g r a d a t i o n is mostly r e s p o n s i b l e for c o n s i s -
tent air void control, crusher chip gradation may have
to be altered drastically to produce voids. This can
be done by removing retained 2 mm p a r t i c l e s but the
amount n e c e s s a r y is high. This can increase the cost
of producing bituminous fines tenfold which is another
indication as to why safe allowable percentages of gla-
cial sands should be determined. It is also because of
the diminishing rock availability in existing pits that
the cost to provide high level crush count mixes can be
prohibitively high for some locations. However, inter-
mediate levels of glacial sand used in conjunction with
crushed aggregate with a higher absorption availability
has yet to be investigated fully. Early evidence indi-
cates that this blend could be the economical solution
to long term rut resistance.
In Manitoba, most of the sand used for bituminous

c o n c r e t e is glacial sand which has been r o u n d e d to a
v a r y i n g degree by the water action that d e p o s i t e d it.
This not only reduces the available surface absorption
which restricts bonding, but also allows the individual
sand granules to act as ball bearings. Particles then
have a t e n d e n c y to roll around adjacent p a r t i c l e s as
o p p o s e d to jutting up against them.(1) Subsequently,
rounded glacial sands should be used carefully for high
t r a f f i c volume routes. It is far from clear however,
that it should be omitted all together, since they do
reduce the cost and promote a higher V.M.A. if the
gradation is properly chosen.
The generally higher water absorption of quarried

d o l o m i t i c limestone in Manitoba may help increase the
b o n d i n g potential of the mastic to the stone. It has
b e e n found that k e e p i n g the Los A n g e l e s A b r a s i o n
s p e c i f i c a t i o n below 35% ensures that the h a r d n e s s of
the a g g r e g a t e is sound e n o u g h to r e s i s t f r a c t u r e
(Manitoba Department of Highways s p e c i f i c a t i o n for a
h a n d p i c k e d sample of aggregate used for c o n c r e t e and
b i t u m i n o u s concrete). The angularity of the q u a r r i e d
particles also contributes strength to the mix.(2)
W h i l e the hot mix is c u r i n g after c o m p a c t i o n , the

d o l o m i t i c limestone aggregate draws in the w a r m e d mix-
ture of a s p h a l t cement and m i n u s 80 um m a t e r i a l
(Mastic). Upon cooling, the mastic becomes p e r m a n e n t l y
i n t e r l o c k e d m o r e d e e p l y w i t h i n the s u r f a c e r e d u c i n g
the s h e a r i n g p o t e n t i a l greatly. This can be c o m p a r e d
to c o a t i n g a pail full of m a r b l e s w i t h hot m a s t i c and
then somehow test the resulting stability of the mold.
The r e s u l t w o u l d indicate a w e a k m i x t u r e as b o n d s be-
tween the mastic molecules and the marble surface would
shear q u i c k l y under load. If the same pail were filled
with mastic and c r u s h e d dolomitic limestone, the
r e s u l t i n g stability will be far greater.
Comparison of Mixes
Introduction
M a n y b i t u m i n o u s p a v e m e n t s in and a r o u n d the c i t y
of W i n n i p e g have s u f f e r e d i n s t a b i l i t y r u t t i n g b e t w e e n
one-half inch and one inch in depth. During 1988, city
o f f i c i a l s i n c r e a s e d the crush count s p e c i f i c a t i o n to
60% for retained 5 mm particles for arterial p a v e m e n t s
h a v i n g a m a x i m u m top size of 16 mm (approx. 5/8").
This in turn m a d e it more s e n s i b l e for c o n t r a c t o r s to
c h a n g e t h e i r stone s u p p l y to q u a r r i e d l i m e s t o n e , m o s t
of w h i c h was dolomitic. Dolomitic limestone is made up
of at least 60% m a g n e s i u m carbonate with the r e m a i n d e r
consisting primarily of c a l c i u m carbonate. The
s p e c i f i c a t i o n also r e q u i r e d that p a v e m e n t s w o u l d o n l y
be a c c e p t a b l e if, a m i n i m u m V.M.A. p e r c e n t a g e of four-
teen is achieved, the minimum retained i0,000 um (I0 mm
or approx. 3/8") p e r c e n t a g e is f i f t e e n (85% p a s s i n g
maximum), and the minimum Marshall stability is 7000 N
or 1586 ibs. Careful mix design changes and plant con-
trol p r o c e d u r e s then took place to meet this s p e c i f i c a -
tion. The g r e a t m a j o r i t y of p a v e m e n t s p l a c e d s i n c e
this change has been overlays over concrete so any rut-
t i n g w i l l be e x c l u s i v e l y w i t h i n the b i t u m i n o u s p a v e -
ment. To date, m a n y rut m e a s u r e m e n t s h a v e b e e n m a d e
and the r e s u l t s are low c o m p a r e d to r e s u l t s for p a v e -
m e n t s p l a c e d b e f o r e 1988 after a three y e a r s e r v i c e
life. For the new mix d e s i g n p a v e m e n t s , which were
p l a c e d by a v a r i e t y of c o n t r a c t o r s , the rut m e a s u r e -
m e n t s w e r e as low as 0 mm b e t w e e n i n t e r s e c t i o n s and
o n l y as h i g h as i0 mm at some i n t e r s e c t i o n l o c a t i o n s .
M a n y i n t e r s e c t i o n s however, have s u f f e r e d l i t t l e ap-
p a r e n t r u t t i n g even t h o u g h h i g h e r axle load v e h i c l e s
have t r a v e l l e d the routes.
Mix Number One: Low W a t e r A b s o r p t i o n
The g l a c i a t i o n h i s t o r y of M a n i t o b a is i n t e r e s t i n g
from the p e r s p e c t i v e that two ice sheets c o l l i d e d here
subsequently mixing limestone bedrock fragments from
the n o r t h west with g r a n i t i c bedrock f r a g m e n t s from the
n o r t h e a s t e r l y C a n a d i a n shield. All sand a n d g r a v e l
d e p o s i t s h a v e v a r y i n g p e r c e n t a g e s of l i m e s t o n e and
g r a n i t e d e p e n d i n g upon location. G r a n u l a r pits to the
n o r t h and to the east of Lake W i n n i p e g and south to the
U.S. b o r d e r w i t h i n i00 km or so of the O n t a r i o border,
h a v e a high p e r c e n t a g e of g r a n i t e or trap rock (70% or
greater). To the south and to the west of t h e s e loca-
tions, g r a n u l a r pits are made up of b e t w e e n 50% and 80%
limestone. The low a b s o r p t i o n mix a n a l y z e d was p l a c e d
on P r o v i n c i a l R o a d 307 l o c a t e d in the s o u t h e a s t e r l y
corner of the province. A l t h o u g h the exact p r o p o r t i o n
of granite particles was not ascertained, the
p r e d o m i n a n t a g g r e g a t e was by far granite. The m a x i m u m
top s i z e of the m i x was 19 mm and the c r u s h c o u n t for
p a r t i c l e s r e t a i n e d on the #4 sieve (4.76 mm) was a m i n -
imum of 50%. Listed below are the pertinent
properties.
Mix Number One
% % % % % % Marshall
Passing Passing Passing A.C. Air VMA Stability
9.5 mm 4.76 mm 75 um Total Voids N. ibs
M e a n (x)
75.6 62.3 4.40 5.44 4.20 16.20 4907 1112
S t a n d a r d D e v i a t i o n (S)
4.6 4.8 0.33 0.43 0.51 0.49 768 174
Sample Size = 20.
T h i s m i x is v e r y d i f f i c u l t to improve. The
s t a b i l i t y is low, but to i n c r e a s e it, the m i n u s 75 um
p r o p o r t i o n m u s t go up. This will d e c r e a s e the air
voids to as low as 2.5 or 3.0% at the same a s p h a l t ce-
m e n t percentage. This level of air voids has p r o v e d to
be d a n g e r o u s l y low for a newly p l a c e d p a v e m e n t as fur-
ther d e c l i n e in voids ensue as t r a f f i c c o m p a c t s the mix
to refusal. The asphalt cement p e r c e n t a g e c o u l d now be
l o w e r e d to add extra void space necessary, however, at
5.44% A.C. by total mix, the asphalt p e r c e n t a g e is al-
r e a d y low. It could be lowered another 0.2 or 0.3% but
this w o u l d not a c c o m m o d a t e enough minus 80 um m a t e r i a l
to raise the M a r s h a l l stability substantially. The ad-
d i t i o n of a stability e n h a n c i n g m a t e r i a l w h i c h does not
e f f e c t mix voids w o u l d be appropriate.
This mix design is indicative of the p r o b l e m s en-

c o u n t e r e d when selecting the proper material q u a l i t i e s
n e e d e d to p r o d u c e a rut r e s i s t a n t pavement. For
bituminous pavements, special attention s h o u l d be
f o c u s e d on p r o v i d i n g a p r o p o r t i o n of the a g g r e g a t e
h a v i n g the potential to draw cooling mastic deep w i t h i n
surface pores. This reduces the potential of a g g r e g a t e
s u r f a c e s h e a r i n g under load w h i c h f u r t h e r c o m p a c t s a
b i t u m i n o u s m i x t u r e and r e d u c e s air voids. Low v a l u e s
o b t a i n e d from the Marshall stability can indicate that
the p r o p e n s i t y for this m e c h a n i s m to o c c u r i n c r e a s e s
w i t h axle loadinqs and increased tire pressures.
Mix Number Two: Hiqh Water A b s o r p t i o n
The aggregate composition of this mix is as fol-

lows:
16.0
- mm (approx. 5/8") dolomitic limestone coarse
i0.0
- mm (approx. 3/8") limestone m a n u f a c t u r e d fines
clean
- well graded granular fines
clean
- poorly graded granular V.M.A. blend sand
Material Description
The m a t e r i a l c o m i n g from the g r a n u l a r pit has an

a p p r o x i m a t e makeup of 65% limestone and 35% g r a n i t e and
is l o c a t e d on the Pine Ridge land form near B i r d ' s
Hill, Manitoba. For the g r a n u l a r fines and a d d i t i v e
(clean b e a c h type sand) the a b s o r p t i o n v a l u e s v a r i e d
b e t w e e n 0.8% and 1.5%. The individual sand and p e b b l e
p a r t i c l e s from the granular pit are v i s i b l y r o u n d e d in
a p p e a r a n c e i n d i c a t i n g e x t e n s i v e d e p o s i t o r y w a t e r ac-
tion.
The limestone material is blasted at a Stonewall,

M a n i t o b a quarry. W a t e r a b s o r p t i o n v a l u e s for this
m a t e r i a l varied between 1.2% and 2.0%.
The g r a n u l a r V.M.A. blend sand g r a d a t i o n s w e r e

h o m o g e n o u s but the bituminous fines g r a d a t i o n did vary.
It is the c a r e f u l selection of b a n k m a t e r i a l that
r e d u c e s g r a d a t i o n fluctuation. Final g r a d a t i o n fluc-
t u a t i o n u l t i m a t e l y determines the V.M.A. and air voids
v a r i a n c e at a set a s p h a l t c e m e n t p e r c e n t a g e . If the
v a r i a n c e is too large, p r o b l e m m i x e s w i l l be e n c o u n -
tered. By i n c o r p o r a t i n g four p r o d u c t s into the mix,
b o t h the s t a b i l i t y and V.M.A. can be e i t h e r r a i s e d or
lowered as needed since the bad effects e n c o u n t e r e d by
i n c r e a s i n g one, can be c o u n t e r a c t e d by i n c r e a s i n g the
other. This has p r o v e d to be an e f f e c t i v e tool for
final plant mix adjustment to meet target p r o p e r t i e s on
the road.
80UCHARD ON BITUMINOUS CONCRETE 41
Listed below are the pertinent properties for mix

n u m b e r two.
Mix Number Two
% % % % % % Marshall
Passing Passing Passing A.C. Air VMA Stability
i0 mm 5.0 mm 80 um Total Voids N. ibs
M e a n (x)
81.60 63.00 4.50 5.48 4.10 14.80 9222 2039
Standard D e v i a t i o n (S)
3.25 3.39 0.59 0.17 0.88 0.64 1533 347
Sample Size = 78.
In c o m p a r i s o n to mix n u m b e r one, m i x n u m b e r two

has s u b s t a n t i a l l y h i g h e r s t a b i l i t i e s w h i l e the o t h e r
p r o p e r t i e s are k e p t w i t h i n s p e c i f i c a t i o n . T h r e e cru-
cial control items u s e f u l for r e p e a t e d l y meeting
s p e c i f i c a t i o n were:
(i) K e e p i n g the final p a s s i n g minus 80 um material

consistently at 4.5%.
(2) P r o p e r s e l e c t i o n and a d j u s t i n g of the V.M.A. in-

d u c i n g b l e n d sand p e r c e n t a g e such t h a t a c o a r s e
stable mix can be placed while ensuring ample void
space for asphalt cement and air voids.
(3) Ensure that proper compaction efforts are e x e r t e d

and that rolling patterns maximize results.(3)
Statistical Games
It has b e c o m e a p p a r e n t a f t e r h u n d r e d s of s i e v e
g r a d i n g a n a l y s i s that the v a r i a n c e of i n d i v i d u a l sieve
size g r a d i n g is r e l a t e d d i r e c t l y to the v a r i a n c e of the
V.M.A. property. This r e l a t i o n s h i p can be u s e d to bet-
ter p r e d i c t the a c c e p t a n c e of the best mix grading, and
the best asphalt content such that tender mixes
p r o d u c e d by o v e r - a s p h a l t i n g and fine g r a d i n g d e s i g n s
are avoided. It can also be used for q u i c k c a l c u l a t i o n
and c o m b i n a t i o n results on the basis of a small n u m b e r
of cold feed samples.
Let us d e f i n e the D e v i a t i o n E n v e l o p e as the a c t u a l

r a n g e t h a t e a c h sieve v a r i e s over a l a r g e n u m b e r of
test results.(4) Conceivably, w h e n the r e c o m b i n e d
g r a d a t i o n of a g g r e g a t e s in an a s p h a l t p l a n t f a l l s to
the b o t t o m of the d e v i a t i o n envelope, the mix is c l o s e r
to the m a x i m u m density line and thus will have a h i g h e r
d e n s i t y and a lower V.M.A. Conversely, the V.M.A. w i l l
be g r e a t e s t when the r e c o m b i n e d g r a d a t i o n is at the top
of the envelope. If this was to e x p l a i n d i f f e r e n t M a r -
s h a l l d e n s i t i e s o b t a i n e d t h r o u g h n o r m a l t e s t i n g , one
can always expect a subsequent V.M.A. deviation.
After testing approximately eighty asphalt

samples, cold feed crusher and s c r e e n e r g r a d a t i o n s and
g r a d a t i o n s from the hot bins of a b a t c h p l a n t w e r e
c o r r e l a t e d to the v a r i a t i o n of V.M.A. R e s u l t s from the
analysis indicated that under tight crusher and
s c r e e n e r control, one c o u l d e x p e c t a 2% v a r i a t i o n of
V.M.A. w h e n a s p h a l t s a m p l e s w e r e t e s t e d u n d e r s t r i c t
laboratory guidelines. Thus, a m i n i m u m r e q u i r e d 14%
V . M . A . for a 3/4" m a x i m u m s i z e d mix, w h i c h is the A s -
p h a l t I n s t i t u t e ' s r e c o m m e n d e d minimum, w o u l d r e q u i r e an
a v e r a g e of 15% V.M.A. on mix design and p l a n t m i x for-
m u l a e target test results. This would p r o v e to me that
w i t h i n the e x p e c t e d 2% V.M.A. v a r i a t i o n , the m i x w i l l
always be above the 14% minimum.
If m i x d e s i g n V.M.A. r e s u l t s fall b e t w e e n 14.0%

and 14.5%, it is an o b v i o u s i n d i c a t i o n t h a t a b o r d e r -
line case exists and an increase of the V.M.A. i n d u c i n g
b l e n d sand, or the addition of one to the m i x d e s i g n is
warranted. If mix d e s i g n V.M.A. r e s u l t s fall b e t w e e n
14.5% and 15.5% and subsequent plant mix V.M.A. r e s u l t s
do the same, it could be c o n c l u d e d that the mix is O.K.
a n d no c h a n g e s of m a t e r i a l s are n e e d e d u n l e s s the
s t a b i l i t y r e s u l t s fail.
It should be added at this point however, that to keep

air v o i d p e r c e n t a g e s to w i t h i n s p e c i f i c a t i o n , V.M.A.
values between 15.5% and 18.5% must insist upon poten-
tially dangerously high a s p h a l t contents. Inter-
g r a n u l a r f r i c t i o n a l forces m u s t be r e d u c e d u n d e r s u c h
circumstances since unabsorbed free asphalt oil
m o l e c u l e s provide lubricating. The result of this is a
t e n d e r mix p r o n e to early wheel p a t h d e f o r m a t i o n e v e n
t h o u g h V.M.A. and air void r e q u i r e m e n t s are h e l d to
specification.
Some s p e c i f i c a t i o n writing authorities tie p e r c e n t

p a s s i n g the 4.76 mm sieve to a v a r y i n g m i n i m u m a l l o w -
able V.M.A. requirement. For a finer mix, these mini-
m u m V.M.A. v a l u e s are high c o m p a r e d to the A s p h a l t
I n s t i t u t e ' s r e c o m m e n d e d values. For example, a mix
d e s i g n w i t h 63% p a s s i n g the 4.76 mm sieve, a m i n i m u m
r e q u i r e d V.M.A. of 15.8% was specified. U s i n g the
p r i n c i p l e of a 2% expected variation of V.M.A., to set
an a b s o l u t e V.M.A. m i n i m u m to 15.8% a v e r a g e v a l u e s of
p r o d u c t i o n should be expected to be 16.8%. Since it is
d i f f i c u l t to a s c e r t a i n at what p o i n t a t e n d e r m i x oc-
curs, this high V.M.A. result, with enough a s p h a l t oil
to b r i n g air voids to w i t h i n s p e c i f i c a t i o n , has at
least the p o t e n t i a l for an i n f e r i o r m i x to be laid.
F u r t h e r m o r e , if a plant is set at an a v e r a g e of 16.8%
V.M.A. one can expect a portion of test results to show
plus 17.5% V.M.A. Using 4% air voids with 17.5% V.M.A.
p r o d u c e s a percent voids filled value of 77.1%. It is
this c r i t e r i o n that may d e s e r v e a t t e n t i o n . At w h a t
point are the interparticle frictional forces w i t h i n a
b i t u m i n o u s m i x t u r e r e d u c e d to d a n g e r o u s l e v e l s by the
increase of asphalt oil contents.
Summary and Conclusions
S t r e n g t h a r i s i n g from the b o n d i n g of m a s t i c to
deep pores of a more highly absorptive a g g r e g a t e should
t h e o r e t i c a l l y be greater than bonds to a g g r e g a t e s with
a lower absorption. The r e c e n t t e n d e n c y of some
s p e c i f i c a t i o n w r i t i n g authorities towards 100% m a n u f a c -
t u r e d fines from q u a r r i e d g r a n i t e gives a m e c h a n i s m
f r o m w h i c h the p r e m i s e of this p a p e r c o u l d p r o d u c e an
a s p h a l t i c concrete prone to movement from bond shearing
u n d e r h e a v y loads. This w o u l d t h e n lead to r u t t i n g
even t h o u g h p r e s u m a b l y the best aggregates had been ac-
quired.
A great deal of evidence exists to show that

h i g h e r voids in mineral aggregate reduce a p a v e m e n t ' s
s u s c e p t i b i l i t y to rutting under heavy loads. This
comes from many case studies of pavements in place for
a p e r i o d of five years or longer. If the g r a d i n g of
m a n u f a c t u r e d fines reduces V.M.A. s u b s t a n t i a l l y when
the product tends to the coarse side of the d e v i a t i o n
envelope, or closer to the maximum density line, a
second m e c h a n i s m for mix difficulty emerges. When
v o i d s in any portion of a pavement are low, the pore
asphalt will tend to lubricate the mix and pushing will
occur. Air voids must be plentiful to allow m i c r o -
scopic d e p r e s s i o n s from each load to rebound to their
original position.
Finally, close attention to quality control must

also be exercised. Proper analysis of a g g r e g a t e s at
the crusher, and plant mix calibrations are n e e d e d to
control arterial street pavements. If care is not ex-
ercised inferior pavements can be placed with r e d u c e d
lifespans d e t e r i o r a t i n g the reputation of a s p h a l t i c
concrete from being a first rate economical pavement,
towards a status of uselessness under heavy traffic.
REFERENCES
(1) Brown, E.R. and McRae, J.L. and Crawley, A.B.

"Effects of Aggregate on Performance of Bituminous
Concrete," Implications of Aggregates A.S.T.M. STP
1016, Scheuders, H.G. and Marek, C.R., Ed.,
American S o c i e t y for T e s t i n g and M a t e r i a l s ,
Philadelphia, 1989, pp.34-62
(2) Emery, J., "Asphalt Paving Rutting E x p e r i e n c e in

Canada," Proceedings, Canadian Technical A s p h a l t
Association, 1990
(3) Gervais, F. and Abd E1 Halim, A.O. "Rutting of As-

phalt Overlays," Proceedings, C a n a d i a n T e c h n i c a l
Asphalt Association, 1990
(4) Besterfield, D.H. "Quality Control," P r e n t i c e -

Hall, Inc., Englewood Cliffs, New Jersey 07632
Stephen A. Cross and E. Ray Brown1
SELECTION OF AGGREGATE PROPERTIES TO MINIMIZE RUTTING OF

HEAVY DUTY PAVEMENTS
REFERENCE: Cross, S.A. and Brown, E.R., "Selection of Aggregate Properties

to Minimize Rutting of Heavy Duty Pavements," Effects of Aaareaates and
Mineral Fillers on Asphalt Mixture Performance, ASTM STP 1147, Richard C.
Meininger, Ed,, American Society for Testing and Materials, Philadelphia, PA,
1992.
ABSTRACT: Samples of pavements from 14 states across the United States

were obtained to evaluate the aggregate properties that affect rutting. Twelve
pavements had provided excellent service for five or more years and 30
pavements had experienced premature rutting in less than five years of service.
The aggregates from cores obtained at 0.3 meter (1 foot) intervals across each
pavement were tested for gradation and maximum aggregate size, fractured
face count, and the National Aggregate Association Flow Test. The cores were
also tested for in-place void content and asphalt cement content. The
remaining cores were reheated and recompacted using the Gyratory Testing
Machine (GTM).
The data show that aggregate properties have little effect on rutting when the
voids are low. When the voids are above 2.5%, mixes with higher fractured
face counts and more angular fine aggregate are more resistant to premature
rutting.
KEYWORDS: pavement rutting, aggregate properties, asphalt mixture
1Mr. Cross and Dr. Brown are, respectively, Senior Research Associate
and Director, the National Center for Asphalt Technology at Auburn University,
AL 36849.
45
Copyright@ 1992byby ASTM International
Downloaded/printed
www.astm.org
INTRODUCTION
Backclround
In recent years several states have experienced an increase in the

amount and severity of permanent deformation in all or a portion of the layers
of hot mix asphalt (HMA) pavements. This increase in permanent deformation,
or rutting, has been attributed to the increase in truck tire pressures, axle loads,
and volume of traffic. Recent studies [1-3] have shown that truck tire inflation
pressures have increased substantially and are now averaging around 690 kPa
(100 psi). Hudson et al. [4] have shown truck tire pressures to be as high as
830 kPa (120 psi).
Concern for rutting and high truck tire pressures led to a National
Symposium on the subject in 1987 [5]. The conclusions drawn from this
symposium were that the higher truck tire pressures and increased truck
weights have led to an increase in rutting. The participants also believed that
with more attention to the selection of materials, design and construction rutting
could be minimized.
Obiective
The objectives of this study are to identify the aggregate properties that
affect rutting and to provide information necessary to identify aggregate
properties that produce mixtures that will perform satisfactorily, and to identify
those that are susceptible to rutting under today's traffic Ioadings.
Scope
This study was undertaken in 1987 to evaluate pavements from across

the United States encompassing various climatic regions, containing aggregates
of differing origins and angularity, encompassing different specifying agencies
and construction practices, and containing a large sample size to make the
results national in scope.
Forty-two pavements were sampled from fourteen states (Figure 1)

across the United States. Rut depth measurements were made across the
pavement to quantify the amount of rutting occurring at each site. Traffic
counts were also obtained at each site. A detailed laboratory testing program
was performed on samples of the asphalt mixture from rutted and good
performing pavements. The data were analyzed to determine the aggregate
properties that are necessary for construction of rut resistant asphalt
pavements.
CROSS AND BROWN ON RUTTING OF HEAVY DUTY PAVEMENTS 47
FIGURE - 1 States participating in rutting study.
PLAN OF STUDY
Field testing consisted of obtaining 15.24-cm (6 inch) diameter cores and

rut depth measurements at 0.3-meter (1 foot) intervals across the traffic lane of
each site. Observations of the pavement condition and any unusual conditions
were also noted. Rut depth measurements were obtained using a 3.65-m (12
foot) elevated straightedge to establish a horizontal reference line. The
distance from the straightedge to the pavement surface was then recorded to
the nearest 1.6-ram (1/16 inch) over the proposed core locations.
Tests were conducted in the laboratory to characterize the aggregate

and mixture properties. The 15.24-cm (6 inch) diameter cores were first
measured to determine the layer thickness of each core and then sawed into
layers and the bulk specific gravity (Test Method for Bulk Specific Gravity of
Compacted Bituminous Mixtures Using Saturated Surface-Dry Specimens,
ASTM D2726) determined for each layer. Two cores were then selected and
the theoretical maximum specific gravity (Test Method for Theoretical Maximum
Specific Gravity and Density of Bituminous Paving Mixtures, ASTM D2041)
determined. The two cores were then extracted to determine the asphalt
cement content (Test Method for Quantitative Extraction of Bituminous Paving
Mixtures, ASTM D2172), and the gradation of the mineral aggregate (Standard
Method for Sieve Analysis of Fine and Coarse Aggregate, ASTM C136 and
Test Method for Material Finer than No. 200 Sieve in Mineral Aggregates by
Washing, ASTM C117). From the gradation analysis the maximum nominal
aggregate size and the maximum nominal aggregate size divided by the
mixture layer thickness were determined. The fineness modulus was calculated
from the gradation analysis in accordance with ASTM C136, and the surface
area of the aggregate determined using the surface area factors provided in
MS-2 [6]. The dust to asphalt cement ratio, by mass, was determined from the
gradation analysis and the extracted asphalt cement content.
The extracted aggregate was further examined and the percent of

aggregate with 2 or more crushed faces for the coarse aggregate (retained on
the No. 4 sieve) and the fine aggregate (passing No. 4 and retained on No. 30
sieves) were determined by visual inspection. To determine the angularity of
the fine aggregate (passing No. 4 sieve) the National Aggregate Associations
Uncompacted Voids Test Method A [7] was performed. The uncompacted void
content and the time in seconds for the aggregate to fill the container were
reported.
The remainder of the 15.24-cm (6 inch) diameter cores were heated,

broken up and combined by layers and recompacted utilizing the Gyratory
Testing Machine (GTM) in accordance with Standard Test Method for
Compaction and Shear Properties of Bituminous Mixtures by Means of the U.S.
Corps of Engineers Gyratory Testing Machine (ASTM D3387). The GTM
compactive effort was 300 revolutions, 827 kPa (120 psi) ram pressure, and 1
degree gyration angle. Triplicate samples were made and tested for unit
weight, voids total mix (VTM), and voids in the mineral aggregate (VMA). The
GTM parameters of gyratory elasto-plastic index (GEPI), and final test head
roller pressure (RP) were also recorded. The shear stress to produce 1 degree
angle was calculated from the roller pressure in accordance with ASTM D3387.
State Supplied Data
All of the sites investigated were selected by the various participating

state DOTs. The primary selection criteria given to each state was one good
performing (non rutting) pavement at least 5 years of age and two rutted
pavements less than 5 years old. Additional selection criteda included no
maintenance work except routine maintenance performed on the site prior to
sampling and that the pavements had to be located on routes with heavy traffic.
The traffic data requested from each site were the total applied
equivalent 80kN (18 Kip) single axle loads (ESALs) over the life of the
pavement or the last overlay whichever was applicable, and the percent trucks
in the average annual daily traffic (AADT). The traffic data are shown in
Table 1.
TABLE 1 -- S u m m a r y of t r a f f i c d at a , p a v e m e n t age,
rut depth and average layer thickness.
AGE AT TOTAL AVG. MAXIMUM

MIX TIME OF 80kN TRUCKS, LAYER RUT
SITE TYPE SAMPLING ESALs % AADT THICK. DEPTH
(yrs) (millions) (cm) (mm)
1 SURFACE 8.80 13.605 50 6.18 38.1

2 OGFC 5.75 2.048 20 2.03 22.8
3 OGFC 5.50 3.118 22 1.76 9.5
4 SURFACE 13.10 2.736 12 3.00 6.4
5 OGFC 5.40 5.030 41 1.95 15.9
6 SURFACE 4.90 4.090 34 3.62 14.6
7 SURFACE 4.90 1.698 23 3.99 8.7
8 SURFACE 12.00 11.338 34 3.18 10.2
9 SURFACE 2.10 2.167 i0 *** 25.4
i0 SURFACE 6.20 2.947 21 2.02 3.2
Ii SURFACE 1.90 0.713 16 2.79 14.0
12 SURFACE 0.80 0.417 5 4.37 36.8
13 SURFACE 0.90 2.943 25 4.05 42.1
14 SURFACE 0.60 1.459 15 6.35 37.6
15 SURFACE 1.90 0.912 28 3.78 2.4
16 SURFACE 1.75 0.870 29 3.59 13.9
17 SURFACE 2.70 1.654 24 3.06 11.7
18 OGFC 5.00 1.473 21 2.18 5.1
19 SURFACE 3.30 0.385 3 3.88 9.9
20 OGFC 2.10 0.334 19 2.14 8.0
21 SURFACE 3.70 0.524 6 3.56 34.8
22 SURFACE 6.00 4.400 50 5.18 12.7
23 SURFACE 4.00 3.300 40 3.61 14.9
24 SURFACE 13.00 5.300 9 3.24 8.0
25 SURFACE ... . . . . . . 6.64 13.7
26 sURFAcE 600 0.584 22 238 137
27 SURFACE 18.00 2.174 9 4.95 8.5
28 SURFACE 1.30 0.688 20 4.10 7.6
29 SURFACE 15.50 1.379 i0 3.26 13.0
3O SURFACE 6.70 2.821 17 2.92 8.3
31 SURFACE i0.00 6.835 24 3.62 27.4
32 SURFACE 2.75 1.498 12 5.04 24.9
33 SURFACE 2.75 1.102 ii 2.98 17.8
34 SURFACE 2.75 1.246 8 3.30 24.9
35 SURFACE 2.75 0.970 8 3.15 16.1
36 SURFACE 14.00 6.200 20 3.87 6.4
37 MAINT. i0.00 4.800 23 2.32 31.8
38 SURFACE ii.00 4.600 28 4.30 15.1
39 SURFACE 13.50 5.200 42 2.66 10.3
4O SURFACE 5.60 1.867 26 3.14 25.4
41 SURFACE 4.40 0.500 14 3.77 14.4
42 SURFACE 2.10 1.486 44 3.87 0.9
... Data not available.

*** Sample from millings.
Indian Institute of Technology Kharagpur pursuant to License Agreement. No further reproductions authoriz
SUMMARY OF TEST RESULTS
General Observations
The majority of the test sites were located on level tangents of 4 lane
divided highways. Thirty of the test sites were located on Interstate highways.
Four of the sites (Sites 1, 19, 21 and 25) were located near intersections and
the pavements were subjected to static Ioadings, and breaking and acceleration
forces of tires. Open graded friction courses (OGFC) were found at the surface
at sites 2, 3, 5, 18 and 20. The lack of sufficient thickness of the OGFC
prevented complete testing of this mixture for these five sites. Two sites (sites
25 and 37) were sampled and tested but not included in the analysis. Site 25
was located in an area of lane widening and the pavement was not of
consistent cross-section across the width of the pavement. Site 37 had
experienced extensive maintenance treatments and no test information was
available on the upper 7 to 10-cm (3 to 4 inches) of the pavement. All of the
pavements carried heavy truck traffic.
The primary mode of distress in the pavements evaluated was rutting.

Some moisture damage was apparent in several layers of cores from several
sites. Moisture damage was the primary mode of distress at two sites (sites 3
and 18) and was evident in four other sites (sites 2,25,27, and 38). The
moisture damaged sites were included in the analysis of the data.
Rut Depth Measurements
Relative elevations obtained at core locations were used to establish the

pavement cross section. From this data the maximum rut depth at the surface
could be determined by measuring the vertical distance between a straight line
connecting high points on opposite sides of the rut and the low point on the
pavement surface near the middle of the rut. The average thickness of each
pavement layer was determined from core measurements. A summary of the
rut depth and layer thickness measurements are shown in Table 1.
Data From Pavement Cores
Approximately eleven cores were obtained at 0.3-m (1-foot) intervals

across the sampled lane. The cores were examined for unusual conditions, the
differing mixtures identified and the layer thicknesses determined. The various
mixtures identified and the mixture layer thicknesses are shown in Table 1.
The results of the extraction and gradation analysis are shown in Table 2.
The average voids total mix (VTM), standard deviation, and 20th
percentile VTM were determined for each layer at each site. The 20th
percentile VTM would be the void content with 20% of the data below this value
and 80% above. Previous work at NCAT [8, 9] has shown the 20th percentile
CROSS AND BROWN ON RUTTING OF HEAVY DUTYPAVEMENTS 51
TABLE 2 -- A g g r e g a t e g r a d a t i o n , a s p h a l t cement
content, and 20th percentile VTM.
PERCENT PASSING ASPHALT 20th

CEMENT PCT'L
SITE 19.0 12.5 9.5 #4 #8 #16 #30 #50 #100 #200 CONTENT VTM
(mm) Cram) (mm) (%) (%)
1 i00 96 88 68 56 47 39 29 18 11.2 5.7 0.4

2 ... ... ... ... ... ... ... ... ... ...
3
5.6 3.1
5 . . . o . , , o . . . . o . . . . , . . o o o o o o
6 {66 99 91 61 43 34 2v 15 10 so
7 i00 99 93 74 60 49 38 24 Ii 6.7 5.3 2.2
8 I00 99 91 61 45 35 26 15 9 7.2 4.5 2.1
9 i00 i00 98 74 55 41 29 19 ii 6.8 7.1 ***
I0 i00 i00 98 75 59 49 35 19 8 4.1 6.8 5.1
ii i00 99 91 69 53 41 28 14 8 6.1 6.3 2.7
12 i00 i00 97 73 54 43 36 26 16 10.2 6.5 1.3
13 i00 I00 97 67 49 32 20 13 9 6.1 6.2 3.5
14 I00 97 91 69 52 42 35 23 14 I0.i 5.2 1.5
15 I00 i00 97 57 33 24 18 13 9 7.3 6.3 5.5
16 i00 I00 98 61 37 26 18 13 I0 7.4 6.6 3.5
17 i00 i00 98 58 36 25 18 12 8 6.2 6.2 4.0
18 i00 i00 I00 76 19 ii 9 7 6 4.7 4.3 14.6
19 i00 93 79 59 43 33 27 20 13 8.2 5.7 0.9
20 i00 i00 i00 44 20 14 12 i0 8 5.9 5.6 12.3
21 i00 99 85 59 47 36 23 12 7 4.3 6.2 2.2
22 i00 98 87 56 39 30 23 16 ii 7.4 5.3 1.5
23 i00 96 83 55 38 29 25 20 14 9.9 5.0 1.8
24 i00 I00 i00 99 78 54 39 29 22 16.3 6.3 1.4
26 i00 i00 92 54 25 16 13 i0 8 6.9 8.1 3.2
27 I00 88 76 54 34 23 17 13 i0 8.0 4.8 8.2
28 100 100 96 69 48 36 28 15 9 7.4 5.6 3.5
29 i00 97 89 61 45 37 30 18 13 8.8 4.8 2.6
30 i00 99 99 84 65 50 39 20 ii 7.7 7.0 1.3
31 i00 99 89 62 43 31 23 16 12 9.4 5.3 i.i
32 I00 96 84 68 54 41 29 16 8 5.9 5.4 3.2
33 i00 95 83 66 53 41 29 17 i0 8.5 5.2 2.3
34 i00 93 81 62 46 34 25 15 7 5.0 5.0 1.4
35 i00 95 83 65 50 37 27 15 8 5.3 5.3 1.6
36 99 92 79 51 41 33 28 25 21 9.5 5.8 1.4
38 I00 92 77 51 39 32 28 22 15 7.7 5.5 0.8
39 i00 i00 94 67 55 49 42 27 13 8.4 6.3 4.6
40 i00 88 71 53 49 45 40 21 i0 7.2 5.9 1.4
41 i00 99 92 76 60 45 34 24 16 9.1 7.8 0.9
42 I00 i00 98 80 55 27 19 14 ii 6.6 5.7 5.1
... N o t e n o u g h m a t e r i a l t o t e s t .
*** S a m p l e f r o m m i l l i n g s .
Indian Institute of Technology Kharagpur pursuant to License Agreement. No further reproductions authorized
VTM to be a reasonable void content to use to predict rutting, and a good

approximation of the wheel path void content. The 20th peroentile voids are
shown in Table 2.
The gradation analysis was used to determine the maximum nominal

aggregate size and the maximum nominal aggregate size divided by the
mixture layer thickness. The fineness modulus, the surface area of the
aggregate and the dust to asphalt cement ratio determined from the gradation
and the extracted asphalt content are shown in Table 3. The percent of
aggregate having 2 or more crushed faces for the coarse and fine aggregate,
and the National Aggregate Associations Uncompacted Void content and flow
time are reported in Table 4.
Recompacted Properties
The GTM recompacted properties of unit weight, V'I'M, VMA, and the
GTM parameters of GEPI, and final test head roller pressure (RP) were
recorded for the GTM samples. The data are shown in Table 5.
ANALYSIS OF DATA
The analysis of the data was limited to the top pavement layer. Previous
work at NCAT [8, 9] and work by Parker and Brown [10] has shown that most
rutting is confined to the top 7 to 10-cm (3 to 4 inches) of the pavement. The
general data analysis procedure consisted of performing correlation analysis to
determine if the independent variables were significantly correlated to the
dependent variable rate of rutting. A significant correlation was defined as one
which produces a correlation coefficient "R" significantly different from zero at a
confidence level of 90%. This corresponds to an alpha value from the Students
t-test of 0.10. If significant correlations were found, the relationship was further
investigated using regression analysis. No correlation between the dependent
and independent variables was assumed if the alpha value from correlation
analysis was greater than 0.10.
Evaluation of Rut Depth Data
Each of the 42 pavements were selected by the states as either good or

rutted based on a subjective analysis of the rut depth and the age of the
pavement. Figure 2 shows the 30 sites with the lowest rate of rutting and the
subjective rating given each pavement by the various participating states. The
separation between good and rutted pavements appears to occur between
0.0051 and 0.0064-mm (0.00020 and 0.00025 inch) per square root ESAL.
Only one of the good pavements had a rate of rutting greater than 0.0058-mm
(0.00023 inch) per square root ESAL and two of the rutted pavements had a
rate lower than 0.0058-mm (0.00023 inch) per square root ESAL. Site 3 was
TABLE -- A g g r e g a t e and mix properties from

extraction and gradation analysis.
MAXIMUM MAXIMUM
NOMINAL AGGREGATE DUST TO ASPHALT
AGGREGATE SIZE / ASPHALT SURFACE FILM
SITE SIZE LAYER CEMENT AREA THICKNESS
(cm) THICKNESS RATIO (m 2 /kg) (microns)
1 1.27 0.2056 1.96 10.23 5.58

2 . . . . . . . . . . . . . . .
3 . . . . . . . . . . . . . . .
4 1.27 0.4231 1.00 6.56 8.96

5 . . . . . . . . . . . . . . .
6 1.27 0.3504 1.66 6.62 7.29

7 1.27 0.3183 1.26 7.60 7.06
8 1.27 0.4000 1.78 6.29 6.63
9 1.27 *** 0.96 6.98 10.63
I0 0.95 0.4714 0.60 6.03 11.79
ii 1.27 0.4560 0.96 6.05 10.88
12 1.27 0.2905 1.55 9.35 7.31
13 1.27 0.3132 0.99 5.59 11.48
14 1.91 0.3000 1.96 8.82 5.93
15 1.27 0.3357 1.15 5.75 11.41
16 1.27 0.3534 1.12 5.91 11.57
17 1.27 0.4151 1.00 5.21 12.33
18 0.95 0.4371 1.09 3.63 11.91
19 1.27 0.3272 1.44 7.38 8.04
20 0.47 0.2224 1.05 4.34 13.44
21 1.27 0.3563 0.69 4.88 13.16
22 1.27 0.2454 1.41 6.51 8.10
23 1.27 0.3520 1.98 7.92 6.44
24 0.47 0.1467 2.59 12.85 5.14
26 1.27 0.5333 0.85 4.92 17.29
27 1.27 0.2564 1.65 6.04 8.00
28 0.95 0.2324 1.33 6.53 8.63
29 1.27 0.3894 1.84 7.61 6.37
3O 1.27 0.4357 i.i0 7.89 9.20
31 1.27 0.3509 1.78 7.26 7.43
32 1.27 0.2521 1.09 6.16 8.59
33 1.27 0.4260 1.64 7.30 7.20
34 1.27 0.3843 1.01 5.36 9.07
35 1.27 0.4037 1.00 5.65 9.42
36 1.91 0.4925 1.62 9.05 6.78
38 1.27 0.2952 1.41 7.57 7.48
39 1.27 0.4776 1.33 8.67 7.32
4O 1.91 0.6061 1.22 7.42 8.07
41 1.27 0.3371 1.17 8.91 9.22
42 0.95 0.2459 1.16 6.02 9.66
... Not enough material to test.

*** Sample from millings.
54 ASPHA~MIXTURE PERFORMANCE
TABLE 4 -- Summary of t e s t s on extracted aggregates.
COARSE FINE
AGGREGATE AGGREGATE
2 OR MORE 2 OR MORE NAA NAA
CRUSHED CRUSHED UNCOMP. FLOW
SITE FACES FACES VOIDS TIME
(%) (%) (%) (s.c.)
1 58.85 80.45 42.35 ii.00

2 ... ...
3 95.60 99.70 44 70 12.09

4 i00.00 i00.00 45 25 11.80
5 ... ,..
6 95.55 87.50 41,35 li i5

7 99.80 I00.00 41,45 10.84
8 I00.00 94.35 3970 10.62
9 82.35 95.75 42,90 11.37
i0 98.75 98.60 44,10 12.40
ii 57.25 63.50 39,85 ii. 15
12 73.50 83.70 4370 12.15
13 68.60 98.55 44,75 12.40
14 95.35 94.35 43,45 11.74
15 i00.00 99.30 4205 12.30
16 98.65 99.45 42.45 11.48
17 i00.00 99.75 41.80 11.03
18 . . . . . .
19 77.80 98.30 43140 li ;3
20 99.20 i00.00
21 83.05 95.20 41 ;5 10.84
22 81.95 96.50 43.45 11.14
23 100.00 i00.00 42.35 11.73
24 i00.00 i00.00 47.10 12.32
26 91.30 97.70
27 98.60 100.00 44~75 12.72
28 67.95 73.00 41.15 10.93
29 52.35 74.70 42.15 12.10
30 45.25 44.05 42.10 11.98
31 67.10 80.30 43.70 11.20
32 56.25 83.15 36.15 10.20
33 39.35 42.85 39.15 10.17
34 34.35 42.50 39.70 10.96
35 33.90 36.50 39.65 10.76
36 24.55 34.10 44.55 13.28
38 0.00 0.00 46.40 12.50
39 98.40 98.10 42.60 +
40 81.00 58.50 43.40 +
41 98.10 56.70 43.70 +
42 i00.00 I00.00 45.30 +
... N o t e n o u g h m a t e r i a l to test.
+ Data not available.
TABLE 5 -- Summary of G T M recompacted mix properties.
BULK VOIDS VOIDS ROLLER

SPECIFIC TOTAL MINERAL PRESSURE
SITE GRAVITY MIX AGGREGATE GEPI 300 REV.
(%) (%) (kPa)
1 2.421 0.4 13.4 2.10 6.89

2 . . . ~ ~ ~ . . . o . .
3
4 2.361 2.9 15.8 1.20 112.63
5 9 . . 9 o . . o . . . . o . .
6 2.378 2.9 13.8 1.25 98.82

7 2.388 2.1 14.2 1.30 91.15
8 2.422 2.3 13.1 1.22 85.01
9 2.414 0.6 17.2 1.28 63.57
I0 2.245 5.7 20.6 1.15 142.72
ii 2. 374 2.4 17.1 i. 25 142.72
12 2.351 1.0 16.0 1.27 26.89
13 2.423 1.9 16.5 1.25 71.48
14 2.502 2.1 14.5 1.20 79.06
15 2. 3 3 9 2.7 17.1 i. 13 94.92
16 2. 3 6 9 I. 5 16.7 i. 17 80.67
17 2. 333 2.1 16.1 i. 15 90.67
18 2. 1 8 0 I0.5 19.5 i. I0 138.93
19 2.428 i.i 14.7 1.06 54.24
20 2. 199 9.6 21.7 I. 18 133.41
21 2.432 0.7 15.4 1.20 61.36
22 2.508 1.3 13.9 1.15 90.32
23 2.456 1.7 13.5 1.18 98.82
24 2.571 1.8 17.7 1.15 56.77
26 2.263 3.3 21.0 1.08 90.32
27 2.279 9.5 20.0 1.03 136.29
28 2.336 2.3 14.7 1.20 82.51
29 2.347 3.0 13.8 1.20 103.42
3O 2.304 1.9 17.4 1.20 93.08
31 2.359 3.5 15.5 1.20 86.60
32 2. 4 3 6 3.1 15.3 i. 27 75.08
33 2.459 1.8 14.0 1.25 68.19
34 2.418 i.i 12.3 1.22 72.81
35 2.401 2.0 14.1 1.21 66.67
36 2.365 1.0 15.3 1.28 41.37
38 2. 3 1 9 3.2 15 9 0 1.18 54 9 40
39 2.248 5.5 18.9 1.2 121.83
4O 2.367 1.6 14.9 1.35 71.22
41 2. 3 7 9 1.4 19 9 4 1 9 35 60 9 54
42 2.415 2.6 15.9 1.05 80.46
Not enough material to test.
one of the rutted pavements with a rate of rutting less than 0.0058-mm
(0.00023 inch) per square root ESAL, however, the primary mode of distress at
this site was raveling, caused by stripping, and not rutting. Therefore
0.0058-mm (0.00023 inch) of rut depth per square root ESAL was identified as
delineating between good and rutted pavements. This rate of rutting agrees
closely with the rate of rutting identified by Kandhal et al. [8] and Parker and
Brown [10] in field studies of rutted pavements. A rate of rutting of 0.0058-mm
(0.00023 inch) per square root ESAL was utilized to delineate between good
performing and rutted pavements in subsequent analysis.
Mix Parameters from In-Place Data
There are always differences in mix properties measured during mix

design and measured during plant production. The aggregate properties and
gradation obtained from cores and mix properties determined from recompacted
material were used to estimate the properties of the mix "as-placed." The
estimated mixture and aggregate properties can then be utilized to determine
the properties that affect rutting.
Aqqre.qate Properties--When the air voids are low rutting is likely to occur
regardless of the other mix properties. Therefore, to analyze the effect of
aggregate properties, only mixes with in-place voids above 2.5% were
evaluated. When the air voids are below 2.5% the aggregate properties will
have little effect on rutting. The results of the correlation analysis between the
aggregate properties and gradation with rate of rutting, for all of the data and
for the data with in-place voids above 2.5%, are shown in Tables 6 - 9.
None of the aggregate properties alone were correlated with rate of

rutting when all of the data were analyzed. The correlations improved when
only those mixes with in-place voids over 2.5% were analyzed. Figure 3 shows
the relationship between the percent of coarse aggregate with 2 or more
crushed faces and rate of rutting when the in-place voids are greater than
2,5%, The correlation has an R-square of 0.42 and shows as the percent
crushed faces decreases the rate of rutting increases.
The angularity of the fine aggregate was determined from the NAA
Uncompacted Voids Test. The higher the uncompacted void content the more
angular the aggregate. The relationship between the uncompacted voids and
rate of rutting when the in-place voids are greater than 2.5% is shown in
Figure 4. The correlation has an R-square of 0.22. From Figure 4 it appears
that site 13 may be an outlier. The NAA void content for site 13 was more than
two standard errors from the mean and it is believed that construction traffic
was allowed on the mat while the mat was still hot. Treating this data point as
an outlier, the correlation has an R-square of 0.67. The correlation shows that
as the angularity of the fine aggregate decreases the rate of rutting increases.
CROSS AND BROWN ON RUI-rlNG OF HEAVY DUTY PAVEMENTS 57
SITE NUMBER
FIGURE - 2 Subjective performance rating vs rate of rutting.
TABLE 6 -- Summary of correlation coefficients for aggregate properties.
PARAMETER R Alpha n
Coarse Aggregate Crushed Faces -0.099 0.57 35
(Plus No. 4)
Fine Aggregate Crushed Faces -0.020 0.91 35
(No. 4 - No. 3O)
NAA Uncompacted Voids -0.136 0.44 34
Maximum Nominal Aggregate Size 0.234 0.18 35
Maximum Aggregate Size/Layer Thickness -0.031 0.86 34
Dust/Asphalt Cement Ratio -0.171 0.33 35
Surface Area -0.046 0.79 35
Asphalt Cement Film Thickness 0.180 0.30 35
Fineness Modulus -0.036 0.83 35
1-Alptm = proba0ility H not equal to zero.
n = Number of observations.
TABLE 7 -- Summary of correlation coefficients for aggregate properties

with 20th percentile in-place voids greater than 2.5%.
PARAMETER R Alpha n
Coarse Aggregate Crushed Faces -0.644 0.01 15
(Plus No. 4)
Fine Aggregate Crushed Faces -0.318 0.25 15
(No. 4 - No. 30)
NAA Uncompacted Voids -0.469 0.09 14
Maximum Nominal Aggregate Size 0.429 0.11 15
Maximum Aggregate Size/Layer Thickness 0.003 0.99 15
Dust/Asphalt Cement Ratio -0.152 0.59 15
Surface Area -0.312 0.26 15
Asphalt Cement Film Thickness 0.313 0.26 15
Fineness Modulus 0.248 0.37 15
TABLE 8 -- Summary of correlation coefficients for aggregate gradation.
PARAMETER R Alpha n
Percent Passing 19.0-mm Sieve -0.096 0.58 35
Percent Passing 12.5-mm Sieve 0.033 0.85 35
Percent Passing 9.5-ram Sieve -0.065 0.71 35
Percent Passing No. 4 Sieve -0.009 0.96 35
Percent Passing No. 8 Sieve 0.032 0.85 35
CROSS AND BROWN ON RU-I-I-INGOF HEAVY DUTY PAVEMENTS 59
TABLE 9 -- Summary of correlation coefficients for aggregate gradation

with 20th percentile in-place voids greater than 2.5%.
PARAMETER R Alpha n
Percent Passing 19.0-mm Sieve N/A N/A 15
Percent Passing 12.5-mm Sieve 0.023 0.94 15
Percent Passing 9.5-mm Sieve -0.135 0,63 15
0.03
Y -- 0 . 0 3 1 3 8 - 0.00025(X)
0.025 R2 = 0.42 9
0.02
o
0.015
8
0.01
0.005
0 i i r i I I1~
4O 50 60 70 80 90 100
COARSE AGGREGATE C R U S H E D F A C E S (%)
FIGURE - 3 Coarse aggregate (Plus No. 4) Percent 2 or more crushed

faces vs. rate of rutting for mixes with 20th percentile in-place
voids greater than 2.5 %.
0.03
Y = 0.093 - O.002(X)
0.025 R2 = 0.67
0.02
0
Of
0.015
B
S
0.01
0.005
0 I I I I I I I I I I I
35 36 37 38 39 40 41 42 43 44 45 46 47
N A A U N C O M P A C r E D VOIDS (%)
FIGURE - 4 NAA Uncompacted Voids vs Rate of Rutting for mixes with

20th percentile in-place voids greater than 2.5%.
The gradation of the aggregate was determined from the sieve analysis
and used to evaluate the surface area, the fineness modulus and the maximum
nominal aggregate size. As shown in Tables 6 - 9, none of these properties
alone had a good correlation with rate of rutting when all of the data were
analyzed. The correlations improved slightly when only the mixtures with in-
place voids over 2.5% were analyzed. When the voids were above 2.5%, the
best correlation between gradation and rate of rutting was with the percent
passing the No. 100 sieve. The correlation has an R-square of 0.17 which is
too low to be significant. No correlation was found between maximum nominal
aggregate size (R2 = 0.18), surface area (FI= = 0.10) or fineness modulus
(R2 = 0.05) and rate of rutting.
Much attention has been focused on the maximum aggregate size as a

function of the mixture layer thickness. The relationship between the maximum
nominal aggregate size divided by the mixture layer thickness for mixes with
voids greater than 2.5% is shown in Figure 5. From the plot it is evident that
no correlation exists between maximum aggregate size divided by mixture layer
thickness and rate of rutting. None of the pavements evaluated contained
aggregates with a nominal size greater than 1/2 the mixture layer thickness.
No other single aggregate property had a correlation with rate of rutting.
0.03
0.023
<
0.02
O
~ 0.015
2
O.Ol
0.~5
0 lit t t t t t
0.2 0.25 0,3 0.35 0.4 0.45 0.5
MAX. AGGREGATE SIZE / LAYER. THICKNESS
FIGURE - 5 Maximum nominal aggregate size divided by mixture layer

thickness vs rate of rutting for mixes with 20th percentile in-
place voids greater than 2.5%.
Tables 10 and 11 show the means and standard deviations of the

aggregate properties and gradation evaluated by performance for mixes with in-
place voids greater than 2.5%. The data clearly show that mixes with higher
percentages of crushed coarse and fine aggregate have lower rates of rutting.
High fine aggregate voids and flow time, as determined in the NAA method,
also results in a lower rate of rutting. From the above data it is clear that
angular crushed aggregates must be used to resist premature rutting.
Gyratory Testin.q Machine Parameters--The summary of the test results

from the GTM samples are shown in Table 5. Table 12 shows the results of
the correlation analysis performed on the GTM parameters. The analysis
showed a significant correlation between shear stress to produce 1 degree
gyration angle and rate of rutting. The relationship between the GTM shear
stress and rate of rutting is shown in Figure 6. The correlation has an R-square
of 0.17. Sites 1 and 36 are more than two standard errors from the mean and
appear to be outliers (Site 1 does not show on the plot due to scale). Site 1 is
at a signalized intersection and the pavement is subjected to static loads. The
surface layer at this site is also a mixture of two or more mixes and the
pavement was so distorted that the mixes could not be definitely identified. Site
36 is located at a high elevation, over 1500-m, (5000 ft) and contained 9.5%
passing the No. 200 sieve. Much of this minus No. 200 material was uncoated.
Treating sites 1 and 36 as outliers the correlation has an R-square of 0.49.
There was no correlation found between VMA (R= = 0.01), or GEPI (Rz = 0.01)
and rate of rutting.
62 ASPHALT MIXTUREPERFORMANCE
TABLE 10 -- Summary of aggregate properties with 20th percentile

in-place voids greater than 2.5%, by performance.
PARAMETER PERFORMANCE MEAN STANDARD

DEVIATION
Coarse Aggregate Crushed Good 98.82 1.46
Faces (%), (Plus No. 4) Rutted 76.43 19.76
Fine Aggregate Crushed Good 99.46 22.78
Faces (%), (No. 4- No. 30) Rutted 86.37 13.59
NAA Uncompacted Voids Good 44.11 1.29
(%) Rutted 41.21 2.47
Maximum Nominal Good 1.04 0.28
Aggregate Size (cm) Rutted 1.25 0.10
Maximum Aggregate Good 0.36 0.11
Size/Layer Thickness Rutted 0.37 0.10
Dust/Asphalt Cement Good 1.13 0.30
Ratio (%) Rutted 1.20 0.34
Surface Area (m2/kg) Good 5.88 1.51
Rutted 6.07 0.81
Asphalt Cement Film Good 9.52 1.64
Thickness (microns) Rutted 10.49 3.09
Fineness Modulus Good 4.23 0.55
Rutted 4.22 0.28
CROSS AND BROWN ON RU-I-rlNG OF HEAVY DUTY PAVEMENTS 63
TABLE 11 -- Summary of aggregate gradation properties with 20th

percentile in-place voids greater than 2.5%, by
performance.

DEVIATION
Percent Passing Good 100.0 N/A
19.0-mm Sieve Rutted 100.0 N/A
Percent Passing Good 98.3 4.2
12.5-mm Sieve Rutted 98.9 1.5
9.5-mm Sieve Rutted 94.5 4.6
No. 4 Sieve Rutted 63.1 5.1
TABLE 12 -- Summary of correlation coefficients for GTM recompacted

mix properties.
PARAMETER R Alpha n
Voids Total Mix -0.362 0.03 35
Voids in the Mineral Aggregate -0.084 0.63 35
Gyratory Elasto-Plastic Index 0.094 0.59 35
Shear Stress at 1 degree angle -0.433 0.01 35
The means and standard deviations for the GTM parameters based on
performance are summarized in Table 13. The results show that pavements
with higher VMA, and higher shear stress to produce 1 degree gyration angle,
performed better. The results for GEPI were inconclusive.
Rutting Model
Analysis of the above information indicates that there is no single

aggregate property alone that can predict the rate of rutting with any high
degree of confidence. The literature reports [8-10] that low voids cause rutting
and it has been shown that aggregate properties, when the in-place voids of the
mix dropped below 2.5%, have little effect on rate of rutting. Therefore, to
determine the effect of aggregate properties on rate of rutting, it was necessary
to evaluate mixes with higher air voids. The aggregate properties selected for
evaluation were percent 2 or more crushed faces for the coarse and fine
aggregate, and the NAA uncompacted voids. The gradation parameters
evaluated were the percent passing the 9.5-ram, No. 4 and No. 200 sieves.
The parameters of maximum nominal aggregate size and the maximum nominal
aggregate size divided by the mixture layer thickness were also included.
The best model for predicting rate of rutting from the above parameters,
when the in-place voids were greater than 2.5%, were percent 2 or more
crushed faces for coarse aggregate (plus No. 4) and NAA Uncompacted voids.
The model is shown in Figure 7. Site 13 is more than two standard errors from
the mean and appears to be an outlier. Utilizing site 13 as an outlier for
reasons previously described, the correlation has an R-square of 0.77 and
shows that as the angularity of the aggregate decreases, the rate of rutting
increases. The correlation has the following form:
P = 0.080038- 0.00008(CF) - 0.00151(NAA) (1)

where:
P = Predicted Rate of Rutting,
(rut depth (mm) / square root ESAL)
CF = 2 or More Crushed Faces Coarse Aggregate, (%)
NAA = NAA Uncompacted Voids, (%)
CONCLUSIONS
Based on the data obtained in this study the following conclusions are
warranted.
1. Rutting is usually caused by low in-place air voids. Certain aggregate

properties are important when the in-place air voids are above
approximately 2.5%.
0.06
\ Y = (5.472/X) - 0.0107
0.05
i 0.114
.~ O.O3
~0.02
Outliers m m
0 I 9 u I 9 gt 9 iII
0 100 200 30O 4OO
GTMSHEAKSTRENGTH(kPa)
FIGURE - 6 GTM shear stress to produce 1 degree angle vs rate of rutting.
TABLE 13 -- Summary of GTM parameters by performance.

DEVIATION
Voids Total Good 4.68 3.42
Mix (%) Rutted 1.92 0.88
Voids in the Mineral Good 17.74 2.51
Aggregate (%) Rutted 15.32 1.92
Gyratory Elasto-Plastic Good 1.16 0.07
Index Rutted 1.25 0.19
Shear Stress at 1 Good 262.7 76.3
degree angle (kPa) Rutted 213.4 63.8
0.03
,-3
<
P = 0.08004 - 0.00008(CF) - 0.00151(NAA)
o.o~
R 2 -- 0.77
0
0
~Z
C~ 0.02
(.t3
8
~J / / ~ - L i n e of equality
0.015
,.~ 0.01
m
/ m I Out
. . lier'aw
. . . I|
0.0o5
1 I I I f
0 0.005 0.01 0.015 0.02 0.025 0.03
ACTUAL RUT DEPTH (ram) / SQ ROOT ESAL
FIGURE - 7 Actual rate of rutting vs predicted rate of rutting (P) for

percent 2 or more crushed faces (CF) and NAA voids
(NAA) with 20th percentile in-place voids greater than 2.5%.
2. The shear stress to produce 1 degree gyration angle as indicated by

the GTM roller pressure gave the single best correlation with rate of
rutting.
3. If the in-place air voids are above 2.5%, the angularity of the
aggregate as measured by percent 2 or more crushed faces for the
coarse aggregate (plus No. 4) and NAA Uncompacted Voids for the fine
aggregate (Passing No. 4) are correlated to rate of rutting.
4. For the pavements evaluated, the maximum aggregate size divided by

the mixture layer thickness was not correlated to rate of rutting.
5. For the pavements evaluated, the voids in the mineral aggregate was
not correlated to rate of rutting.
6. For the pavements evaluated, the gradation of mix as measured by

the sieve analysis and fineness modulus was not correlated to rate of
rutting.
BIBLIOGRAPHY
[1] Middleton, D.R., F.L. Roberts and T. Chira-Chavala, "Measurements and
Analysis of Truck Tire Pressures on Texas Highways," Transportation
Research Record 1070, Transportation Research Board, Washington,
D.C., 1986, pp. 1-8.
[2] Thompson, M.R., "Analytical Methods for Considering Tire Pressure
Effects in Pavement Design," Proceedings, A Symposium/Workshop on
High Pressure Truck Tires, American Association of State Highway and
Transportation Officials and Federal Highway Administration, Austin,
Texas, February 1987.
[3] Kim, O. and C.A. Bell, "Measurement and Analysis of Truck Tire
Pressures in Oregon," Transportation Research Record 1207,
Transportation Research Board, Washington, D.C., 1988, pp. 100-110.
[4] Hudson, S.W. and S.B. Seeds, "Evaluation of Increased Pavement
Loading and Tire Pressures," Transportation Research Record 1207,
Transportation Research Board, Washington, D.C., 1988, pp. 197-206.
[5] " ," Proceedings, A Symposium/Workshop on High Pressure
Truck Tires, American Association of State Highway and Transportation
Officials and Federal Highway Administration, Austin, Texas, February
1987.
[6] "Mix Design Methods for Asphalt Concrete, Manual Series No. 2," The
Asphalt Institute, College Park, Maryland, March 1979.
[7] Meininger, Richard C. "Proposed Method of Test for Particle Shape and
Texture of Fine Aggregate Using Uncompacted Void Content," The
National Aggregate Association, Silver Springs, Maryland, March 1989.
[8] Kandhal, Prithvi, Stephen A. Cross and E.R. Brown, "Evaluation of
Bituminous Pavements for High Pressure Truck Tires," FHWA Report
No. FHWA-PA-90-008-87-01, Pennsylvania Department of
Transportation, December 1990.
[9] Brown, E.R. and Stephen A. Cross. "Comparison of Laboratory and Field
Density of Asphalt Mixtures," Transportation Research Record 1300,
Transportation Research Board, Washington D.C., January 1991.
[lO] Parker, Frazier and E.R. Brown. "Effects of Aggregate Properties on
Flexible Pavement Rutting in Alabama," Effects of Aq,qre,qate and Mineral
Fillers on Asphalt Performance, ASTM STP 1147, Richard C. Meininger,
Ed., American Society for Testing and Materials, Philadelphia, PA, 1992.
Frazier Parker and E. Ray Brown
EFFECTS OF AGGREGATE PROPERTIES ON FLEXIBLE PAVEMENT RUTTING IN

ALABAMA
REFERENCE: Parker, F. and Brown, E.R., "Effects of Aggregate Properties on Flexible

Pavement Rutting in Alabama," Effects of Ageregates and Mineral Fillers on Asohalt
Mixture Performance, ASTM 8TP 1147, Richard C. Meininger, Ed., American Society for
Testing and Materials, Philadelphia, 1992.
ABSTRACT: Pavement rutting which results in a distorted pavement surface is the

accumulation of permanent deformation in all or a portion of the layers in a pavement
structure. The objective of this study was to examine the effects of aggregate properties
on the rutting of asphalt concrete mixtures. This objective was accomplished by analyzing
rutting data from the Alabama Highway Department's pavement condition database. Field
sites were evaluated and sampled and laboratory tests performed on aggregate from field
samples.
Analyses of the Department's pavement condition database indicate that rutting in

Alabama is increasing, and that this increase is attributable to either increased loading
intensity or increased asphalt concrete rutting susceptibility. The analyses also indicate
that rutting varies geographically and that this variation can be explained by quality of
locally available aggregate. Those areas with crushed stone and angular natural sands are
less susceptible to rutting.
Analyses of data from field test sites indicate that permanent deformation causing rutting
is generally confined to the top 7 to 10 cm (surface and binder courses). There was little
evidence that lower base/subbase courses or subgrade were significant contributors to
rutting. A rate of rutting of 5.0 x 10-4cm/E,J-E-SALor 2.5 x 10-7cm/ESAL delineated good
and poor performing pavements.
Aggregate properties studied include: gradation, coarse aggregate percent fractured

faces, and fine aggregate particle shape and texture. The correlation coefficients for most
of these properties were low, however, there appeared to be definite trends. The poor
correlations show that rutting is a very complicated process that is affected by multiple
factors, and demonstrates the importance ot both aggregate properties and asphalt
content during material selection and mix design.
KEYWORDS: pavement rutting, aggregate properties, asphalt mixture
Drs. Parker and Brown are, respectively, directors of the Highway Research Center and the
National Center for Asphalt Technology at Auburn University, Auburn, AL 36849.
68
Copyright @ 1992byby ASTM International
Downloaded/printed
www.astm.org
PARKER AND BROWN ON FLEXIBLE PAVEMENT RUFFING 69
RUTI-ING OF FLEXIBLE PAVEMENTS
Pavement rutting which results in a distorted pavement surface is the accumulation of

permanent deformation in all or a portion of the layers in a pavement structure. Longitudinal
variability in the magnitude of rutting causes roughness. Water may become trapped in ruts
resulting in reduced skid resistance, increased potential for hydroplaning and spray that
reduces visibility. Progression of rutting can lead to cracking and eventually complete
disintegration.
Flexible pavement rutting is not a new problem. As long as flexible pavements have
been used, rutting has been recognized as a primary distress mechanism and a primary
design consideration. In the past, the consensus was that rutting was generally restricted to
subgrades. In fact, flexible pavement design concepts in the 1986 AASHTO Guide for
Design of Pavement Structures are based on 1) providing sufficient pavement structure
(rutting resistant materials) to reduce stresses in the subgrade to the point where rutting will
not develop, and on 2) providing surface quality and thickness to resist fatigue cracking.
Performance models used in the 1986 AASHTO Guide were developed from the
AASHO Road Test where the nominally 550 kPa tire inflation pressures caused rutting only
in weaker materials such as subgrade soils and possibly subbases. Truck tire inflation
pressures are now significantly higher. The resulting stresses in near surface materials are
also higher, and there is increased awareness that permanent deformation in the high
quality asphalt layers (surface, binder and base) may be a significant contributor to rutting.
This is demonstrated when asphalt concrete overlays of Portland cement concrete
pavements rut.
Repetitive applications of heavy trucks with increasingly high pressure tires drives rut
formation in high quality asphalt layers. The stresses induced in near surface layers by the
high pressure tires may exceed the ability of the materials to resist densification below
critical void levels and subsequent plastic flow.
MODEL FOR RUFFING OF ASPHALT CONCRETE
A model that describes the rutting of asphalt concrete as a two phase process is
illustrated in Figure 1. This model was developed from observation and examination of
pavement structure cross sections and in-place mix properties.
DENSIFiCATIGNI PLASTIC /
DENSIF]CATION
s E3
l-- t--
re S VOIDS
8~4~
TIME TIME
FIGURE 1 -- Model for rut development in asphalt concrete.
In the first phase repeated load applications causes densification from as constructed
void content (8% or less). In properly designed mixes densification stabilizes at about 4%
and rut depth development ceases or decreases to very low rates as illustrated in Figure la.
Densification may stabilize at higher voids but if the voids are much higher than 4%,
problems with durability may develop.
At about 4% voids, the ability to resist permanent deformation in properly designed

mixes is optimum. At this stage it is critical that the aggregate skeletal structure have the
ability to resist further densification, and this is best accomplished with well graded
aggregate with angular rough textured particles.
Asphalt content is also critical as the mix reaches about 4% voids. Excess asphalt will
decrease intergranular contacts, weakening the aggregate skeletal structure and leading to
further densification. Excess asphalt can weaken otherwise very stable aggregate
structures. This emphasizes that aggregate properties and optimum asphalt content are
equally important aspects of the mix design and construction process.
For pavements that experience severe rutting, densification continues and second
phase conditions develop. When voids reach about 2%, the mix becomes very unstable
and plastic flow develops, as illustrated in Figure lb. Rut depth increases rapidly and
upheaval outside wheel paths begins. Carried to extremes, pushing and shoving may
develop causing a dramatic increase in roughness. Dilation may occur as the material shears
and flows plastically from wheelpaths and may cause an apparent increase in voids.
CAUSES OF RUTTING OF ASPHALT CONCRETE
Recent studies [1-5] have shown that truck tire inflation pressures, and therefore
contact pressures, have increased dramatically from the 550 kPa on which design
procedures are commonly based. Average truck tire inflation pressures for radial tires are
now around 700 kPa. This means that a significant portion of the truck tires have inflation
pressures higher than 700 kPa, often in the 900 to 960 kPa range.
The study by Marshek, Chen, Connell and Hudson [6] notes several additional
problems with high tire pressures. A commonly made assumption has been that contact
pressure approximately equals inflation pressure. The study showed that increased tire
pressure produced proportionally smaller gross contact areas. This suggests that the
commonly made assumption of equal pressure becomes increasingly less valid. The study
also showed that contact pressures were not uniform. This suggests that some contact
areas will have pressures greater than that calculated for a uniform pressure.
Extraordinarily high tire pressures mean that asphalt concrete layers which are nearest
to the surface in a pavement structure may not be immune to rutting. Although recent
modifications such as asphalt content selection based on 75 blow Marshall compaction
have increased rutting resistance, material quality provided by existing specification may be
insufficient to meet the demands of today's traffic.
Assuming that truck traffic, loading, and tire pressures are not likely to decrease; the
obvious solution is to increase the resistance of asphalt concrete to permanent
deformation. As with most simple and obvious solutions it must, however, be approached
with caution. Beneficial changes in one property may lead to detrimental changes in
another property. For example, decreasing asphalt content will result in increased rutting
PARKER AND BROWN ON FLEXIBLE PAVEMENT RUTTING 71
resistance, but decreased fatigue resistance. Increased asphalt cement viscosity will result
in a stiffer mix which may be more resistant to rutting, but a mix that is more susceptible to
thermal cracking.
Changes in aggregate properties will likely be most fruitful in improving rut resistance.
Button and Perdomo [7] concluded that the round shape and smooth texture of natural
(uncrushed) aggregate particles was one of the chief mixture deficiencies that contribute to
rutting. Aggregate comprises over 90% of asphalt concrete mixtures and provides the
basic load carrying skeletal structure. With well graded angular aggregate, proper selection
of asphalt properties and asphalt contents can produce mixtures that are not only rut
resistant, but also durable and resistant to fatigue and thermal cracking.
In Alabama, as in other states, high quality crushed aggregates are not available
statewide and economics often dictates the use of locally available natural sands and
gravels. These materials are used extensively in the southern and western part of the state,
and generally have rounded particle shapes which are detrimental to mix stability. Even
when gravel is crushed, the larger particles that result are likely to have no more than two
crushed faces since the maximum particle size of available natural gravel is around 3.8 cm.
In the northern and eastern part of the state, limestone is the predominate crushed
aggregate but cannot be used in surface courses because of polishing. In the past, blast
furnace and steel slags were used extensively for surface course mixes in this region, but as
their availability has diminished, the use of crushed gravel has increased.
STUDY OF RUTTING IN ALABAMA
Part of an overall study of rutting of flexible pavements in Alabama focused on the

influence of aggregate properties in asphalt concrete mixes. Rutting data from the Alabama
Highway Department pavement condition data base was analyzed to 1) assess the nature
and extent of rutting, and 2) to determine if geographical variations in rutting exists that
could be related to geology and thus, variations in locally available aggregate properties. In
addition sampling and testing was conducted at thirteen (13) test sites. Sites were selected
to provide a relatively uniform statewide geographic distribution and materials with a range of
rutting resistance. Laboratory tests were run on aggregate recovered from cores taken from
the 13 test sites.
RUTTING DATA FROM PAVEMENT CONDITION DATABASE
Condition and traffic data for the Alabama state and interstate system are collected
from representative 60 m long test sections in each 1.6 lane km of pavement. Data
collection procedures are described in reference 8. Data collection was initiated by the
Department in 1984 and was repeated in 1986 and 1988. Data records for each 1.6 lane km
contain identifying and descriptive information, quantitative pavement condition information
(including rutting) and estimated traffic data.
Eight rut depth measurements, four in the outer wheel path and four in the inner
wheel path, are taken in a 60 m test section for each 1.6 lane km. Measurements were
made in all lanes, but rut depths in outer lanes were always larger and were used exclusively
in analyses. A 1.2 m long straight edge is placed across the wheel path and the maximum
rut depth measured. For this study these eight measurements were averaged and used to
represent the rutting for each lane mile of pavement. Reference is made in the remainder
of the text to average rut depth.
In the analyses, estimates of the traffic applied to the pavement in terms of total
numbers of 80 kN equivalent single axle loads (ESAL) were required. The annual average
daily traffic (AADT) and percent commercial vehicles used in the computation of ESAL's
were the estimated values for the year in question, i.e. 1984, 86, or 88. Traffic was
assumed split evenly by direction (50/50). The pavement age, in days, was the difference
between the date rated and the date the last surface layer was placed. For older pavements
and pavements with significant traffic growth, use of traffic estimates for the years
measurements were made (1984, 86 and 88) will over estimate total applied traffic.
However, since most of the comparisons were relative, this is not considered a serious
problem that would warrant a more accurate estimation of traffic.
Three parameters were analyzed to determine the nature and extent of rutting.
These were the mean of average rut depths [MRD], the mean of average rut depth
to ESAL ratios [M(RD/ESAL)], and the ratio of mean average rut depths to mean
ESAL's [MRD/MESAL]. The ratios RD/ESAL and MRD/MESAL are indicators of the rate of
rut formation with traffic. Comparisons of the three variables were made between Highway
Department Divisions to determine if geographical differences exist. Data for these
comparisons were grouped according to roadway type (state routes, interstate routes and
combined routes) for 1984, 1986, and 1988 data bases. The data was also combined for
overall comparisons.
Overall Comoarison
Table 1 contains a summary of all applicable data for 1984, 1986, and 1988. Column 2
contains the frequency which is indicative of the number of lane km of pavement. Columns
3, 4 and 5 contain, respectively, MRD, MRD/MESAL, and M(RD/ESAL).
Values from Table 1 are plotted in Figure 2. From this figure the following
observations can be made:
Rut depths are larger on interstate routes than on state routes. This is likely due to
the larger traffic volumes on interstate routes.
~ Rut depth has increased slightly from 1984 to 1988.
The rut depth increase, from 1984 to 1988, is larger for interstate routes (0.0379 cm)
than state routes (0.0091 cm).
Because of the small frequency for interstate routes, the rut depth relationship is
more erratic, i.e., the level of the overlay program can have an observeable influence
on rut depth.
Values of the ratios of means, MRD/MESAL, as shown in (b) are considerably

different than values of means of the ratios, M(RD/ESAL), as shown in (c). This is due
to the large numeric differences between the numerators (average rut depths)
and the denominators (ESAL's), and the wide range of ESAL values. Values shown
for both parameters are multiplied by 107.
PARKER AND BROWN ON FLEXIBLE PAVEMENT RU'I-I'ING 73
TABLE 1 - Summary rut depth data, outer lanes.
Year Frequency MRD(cm) MRD/MESAL (cm x 10-7) M(RD/ESAL)(cm x 10 -7)
State & Interstate
1984 22,683 0.2606 4.874 27.732

1986 22,004 0.2606 5.273 30.437
1988 21,444 0.2716 5.392 31.913
State
1984 21,503 0.2552 6.480 29.017

1986 20,801 0.2558 6.909 32.006
1988 20,260 0.2644 6.957 33.739
Interstate
1984 1180 0.3573 1.138 3.912

1986 1203 0.3436 1.280 2.713
1988 1184 0.3953 1.560 3.780
Frequency - Number of test site. Eight rut depth measurements at each test site, four in inner and
four in outer wheel paths.
MRD - Mean average rut depth.
MRD/MESAL - Ratio of mean average rut depth to mean 80kN equivalent single axle loads.
M(RD/ESAL) - Mean of the ratio of average rut depth to 80kN equivalent single axle loads.
35
o
0.35
z zs
z
0.30
0.25
!~
0.20
~ fo
0.15 I I I
1984 1986 1988
YEAR
a. Mean Rut Depth

I I I
o
84 86 88
YEAR
~ 7.s c. Mean (Rut Depth / ESAL)
u~
o State
3,.0 J - - -
o Interstate
9 Combined
~ 2.5
FIGURE 2 -- Progression of rut development with time.

I i I
0 84 86 88
YEAR
Downloaded/printed
b . Mean Rutby Depth / Mean E S A L
Because of the large intluence ot extreme values ot ESAL's, the ratio of means is
considered a better indicator of rate of rut development than the mean of the ratios.
Both ratios indicate that the rate of rut development is much greater on state routes
than interstate routes. This is likely due to higher quality pavements (including quality
of asphalt bound materials) on the interstate system.
The ratios of means (b) show about the same increase in rate of rut development from
1984 to 1988 for state routes (0.478) and for intersate routes (0.422).
The mean of the ratios (c) show a much larger increase in rate of rut development,
from 1984 to 1988, for state routes (4.722) than for interstate routes (-0.132).
To summarize, all but one of the parameters examined indicated that the problem of
rutting is increasing. For rut depth this could be caused by an increase in pavement rutting
susceptibility, an increase in traffic volume or an increase in loading severity (truck weight
and/or tire pressure). For rate ot rut development, possible causes would be restricted to
pavement rutting susceptibility and loading severity.
ComParison bv Division
Comparisons by highway department divisions were made to determine if geographical

variations in rutting exists and to examine possible reasons for these variations.
Speculation was that geology and, thus, variable quality of locally available aggregates might
be a factor. As shown in Figure 3, Divisions 1,3 and 4 are predominately in the Piedmont
and Appalachian Plateau geologic provinces. Rock deposits in these areas are used for
crushed stone and are the source of sand and gravel materials. Division 2 is divided
between the Appalachian Plateau and the Coastal Plain region.
Divisions 5-9 lie below the Fall Line in the Coastal Plain region. Natural sands and
gravels are the predominate aggregate materials used in this region. The degree of
weathering and, thus, particle size and shape of sand and gravel is influenced by the
distance transported from the source. Particles become rounder and smaller as the
transported distance increases. Implications are that aggregate quality and, therefore, mix
rutting susceptibility should increase with movement southward as the distance from rock
sources in the Piedmont and Appalachian Plateau regions increases.
The influence of particle size and shape for natural sand (fine aggregate) and
uncrushed gravel (coarse aggregate) is straight-forward and well established. However,
when gravel is crushed, the influence of size is not as direct or as well documented. Since
specifications for coarse aggregate 1or surface and binder mixes requires some crushed
particles, natural gravels must be crushed. The problem created by using gravels is that the
degree of particle fracturing is directly related to original particle size. Smaller gravel particles
are less fractured and mixes containing these partially crushed particles are more
susceptible to rutting.
If indeed geology and, thus, geography is a tactor, rutting susceptibility should be

less in Divisions 1-4 than in Divisions 7-9. Divisions 5 and 6 should be intermediate. Figure
4 shows three histograms which illustrate the variability of rutting susceptibility between
divisions. From the three histograms, it can be seen that the mean rut depth and the rate of
rutting, as quantified by the ratio of means and the mean of the ratios, are less for Divisions
1-4 than for Divisions 5-9, The averages of the three parameters for Divisions 1-4 are
/ 7 ,<" ~./<,.~,,t
.r 9 @ TEs,s,T~s
~ , v ~ L. ~ 3 APPALACHIAN P-ATEAUS
~ FALL LIN~:
FIGURE 3 -- Location Map.
0.1875 cm, 3.2664 x 10-7 cm/ESAL and 21.6154 x 10 -7 cm/ESAL compared to 0.3241
cm, 6.9850 x 10-7 cm/ESAL and 34.6862 x 10-7 cm/ESAL for Divisions 5-9.
The data shown in Figure 4 is from the 1984 database for combined state and
interstate routes. However, an analysis of all the data in the 1984, 86 and 88 databases
confirm the trends illustrated in Figure 4. Averages from the databases are shown in Table
2. These averages conlirm that pavements in Division 5-9 are more susceptible to rutting
than those in Divisions 1-4.
The most consistent indicator is the ratio of means. Rut depth and mean of ratios are
more sensitive to pavement age and may fluctuate rather widely in response to major
overlay construction programs. This is likely the cause of the reversal in the relationship
between rut depth and mean of ratios for combined and state route data that occurred
between 1986 and 1988. Contrary to 1984 and 86, 1988 rut depths on state and
combined routes for Divisions 1-4 are about equal to those in Divisions 5-9. In addition, the
means of ratios for 1988 state and combined data become larger in Divisions 1-4. This
reversal in overall trend is thought to be primarily due to a reversal in individual trends for
Divisions 4 and 5 caused by a much larger overlay construction program in Division 5 than in
Division 4.
Despite the exceptions noted above, the analysis of rutting from the data bases
supports the contention that rutting susceptibility is related to geographic location. In
addition, geology and, thus, properties of locally available aggregate provide a logical
explanation for the observed relationship between rutting susceptibility and geographic
location. This phenomenon will be examined further in the analysis of the data from field
test sites.
76 ASPHALT MIXTURE P E R F O R M A N C E
VALUE FREQ~IENCY PERCENT

(OVEP,A ~ ~7n2 ;~414 100
31~Z 24OO 107
m m 2S4~ 1~ 87
0 ~i 77 ;2748 121
m 00~7 2727 120 ml s;,~ 27o6 121
0~27 2~ toe
s,, 742 ~7. lo~
ae
, [ , , ]
oo l~s 2~o ~7~ ~o e~s 7so
tVl[~N R U T D E P T H (~) MFJJ.N (RUT DiE PTH/80 kN E ~ [ . ) (z l o t )
1984 DATA
STATE AND INTERSTATE
OUTER LANES
VAL E FREC~J~NCY PERCENT

(x10~ )
4e74 22.414 100
4~O Z"O0 I07
n 3147 I~ S.7
m
u 2.141 2~C~ ~21
m m m m 4.3718 21~245 12.6
n
~1 ~ 237. lo6
F I G U R E 4 -- Rutting and rate of rutting with
8.6 A H D division, 1984 state and
MEAN RUT DE P T I - ~ AN 8O IoN ESAL (x ~0~) interstate data.
TABLE 2 -- Summary of rutting susceptibility by A H D division.
D~abase Year
Routes Divisions 1984 1986 1988
MRD (crn)
State & Interstate 1-4 0.1874 0.2156 0.2685
5-9 0.3240 0.2950 0.2732
State 1-4 0.1872 0.2176 0.2670
5-9 0.3165 0.2844 0.2566
Interstate 1-4 0.2542 0.2318 0.2819
5-9 0.4736 0.4746 0.4497
MRD/MESAL (cm x I(T 7)
State & Interstate 1 -4 3.266 3.962 5.448
5-9 6.985 6.988 6.299
State 1-4 4.293 5.126 6.591
5 -9 9.215 8.697 7.620
Interstate 1-4 6.086 6.703 9.926
5- 9 16.805 19.578 21.514
M(RD/ESAL) (crn x 1(T7)
State & Interstate 1 -4 21.615 28.125 38.341
5 -9 34.686 32.334 26.579
State 1 -4 22.408 29.555 38.341
5 -9 36.921 34.1 68 27.615
Interstate 1 -4 1.676 0.947 3.439
5 -9 4.026 3.513 3.810
ANALYSIS OF DATA FROM TEST SITES
Thirteen test sites were selected for evaluation and testing. The approximate location
of these sites is shown on Figure 3. Sites were selected to provide a relatively uniform
statewide geographic distribution. Sites were also selected to provide examples in the
Piedmont and Applachian Plateau geologic regions where crushed stone is available, and
in the Coastal Plain geologic region where natural sands and gravels are the predominate
aggregate used in hot mix asphalt. The sites were selected to provide a range of rutting
performance. Five (5) sites were considered by Highway Department personnel to have
provided good rutting performance and eight (8) to have provided poor rutting
performance.
The development of rutting was investigated by cutting a trench and coring through
all asphalt bound layers. All test sites were on four lane facilities and the trenches and cores
were taken across outside lanes only. Two lines of 12 cores each, one line of 10 cm cores
and one line of 10 and 15 cm cores, were cut.
Pavement surface profiles at the trench and at core lines were obtained by measuring
the distance from a leveled 3.6 m long straightedge. Once the trench was cut, similar
measurements were made to layer interfaces to obtain a complete profile for the
asphalt-bound layers. Layer thicknesses were measured from cores and added to surface
profile measurements as a second method for developing complete layer profiles. Cores
were then used to provide material for the laboratory testing.
The profiles of the asphalt bound layers were analyzed to determine where rutting
was developing. When trenches were opened, stringlines were stretched along layer
interfaces to detect depressions in the top of lower layers. These depressions would be
indicative of permanent deformation in the layer itself or lower layers. Nine of the thirteen
pavements were comprised of an original structure plus at least one overlay. Measurements
in the trenches indicated that permanent deformation was primarily confined to near surface
(approximately 10 cm depth) asphalt bound layers. In most pavements this meant that
permanent deformation was limited to surface and binder layers with the interface between
binder and black base layers relatively depression free.
At only Site 9 was there evidence of rutting in base or subbase layers below asphalt
bound layers. At only Site 2 was there evidence that stripping may have contributed to
rutting. At Site 2 several cores in wheel paths disintegrated and could not be completely
recovered. Stripping was confined to the original binder and base layer.
Rut depths were measured at the test sites with a 3.6 m straight edge across the
outer lane of pavement. Rut depths shown in Table 3 are averages of 3 measurements
each in inner and outer wheel paths. Also shown in Table 3 are rut depth measurements
from the 1988 pavement condition data base. These rut depth measurements were made
at approximately the same time as the test site measurements, but were made with a 1.2 m
straight edge and represent averages for entire projects which were several miles long.
This difference in measuring procedure accounts for the different magnitudes of rut depth.
The rut depths from Table 3 are plotted as a histogram in Figure 5. From this
histogram it is apparent that rut depths at sites selected for good rutting performance are
generally less than rut depths at sites selected for poor performance. While rut depth is an
TABLE 3 - Test site ~ting analysis data.
Rut Degth RD/ESAL R D/qESAL

80kN Test" 1988 + Test 1988 Test
Site ESAL Site DB Site DB Site
(cm) (cm) (cm) (a'n) (crn)
1" 6.6x106 1.22 0.97 1.85X10-7 1.47x10-7 47.5x10 ,5

2 2.6X106 1.14 0.76 4.39X10-7 2.92X10-7 70.9x10 "5
3 2.9x106 1.19 1.04 4.11X10-7 3.58x10-7 70.1x10-5
4* 4.3x106 0.76 0.51 1.78x10-7 1.19x10-7 36.8x 10-5
5 2.0x106 2.77 0.89 13.84x10-7 4.44x10-7 195.8x10-5
6 3.7X106 2.13 1.09 5.77X10-7 2.95X10-7 111.0X10-5
7* 3.6x106 0.56 0.33 1.55X10-7 0.91x10-7 29.5x10-5
8 0.5x106 1.42 0.51 28.45x10-7 10.16x10-7 201.2x10-5
9 1.6x106 1.68 0.99 10.46x10-7 6.20x10-7 132.6x10-5
10 2.0x106 1.35 0.71 6.73x10-7 3.56x10-7 95.2x10-5
11" 5.9x106 0.89 0.46 1.50x10-7 0.76xl 0-7 36.6x10 -5
12 0.3x106 0.36 0.05 11.86x10 -7 1.70x10-7 65.0x10-5
13" 1.5x106 0.66 0.41 4.39x10-7 2.72xl 0-7 53.8x10 -5
* Sites selected for good rutting performance.

9 Rut depth measured with 3.6 m straight edge across lane.
+ Rut depth measured with 1.2 m straight edge across wheel path. Mean of average rut depth for
entire design section in which test site located.
2.75
2.50
Rut depth from trench I Rutdepth from 1988AHD ]
2.25 Jand cores at test pavement conditiondata
[ sites base
2.0
~ Selectedfor good performance
~ 1.75
-.]-
1.50 q Selected for poor performance
W
1.25
I--
C)
rr 1.0
0.75
0.50
0.25 Illl
4 7 11 13 2 3 5 6 8 9 10 12
SITE
N
FIGURE 5 - Rut depths at test sites.
indicator of pavement performance, the evaluation must include traffic intensity when
assessing rutting susceptibility.
To study the relationship between traffic and rutting at test sites, traffic was converted
to 80 kN ESAL's applied to the pavement since construction or since the last overlay.
AADT and percent commerical vehicles from the 1986 data base with no growth factors
were used for this purpose. Since pavements were constructed from 1974 to 1985 and
rated in 1988 and 1989, computed 80 kN ESAL's are approximations.
Traffic data for the thirteen test sites is compiled in Table 3. All test sites were on
outer lanes of four lane facilities and the 80 kN ESAL's are estimates for these lanes. Traffic
ranged from 0.3 x 106 ESAL's at Site 12 to 6.6 x 106 ESAL's at Site 1. This represents a 22
fold difference and must be considered when evaluating the influence of traffic on rutting.
The ratio of rut depth to 80 kN ESAL's provides a measure of rate of rutting. Using rut
depths from measurements at the test sites and from the 1988 pavement condition
database, ratios were computed and compiled in Table 3. These ratios are plotted as a
histogram in Figure 6. Except for Site 13, the histogram provides a clear distinction
between the good and poor performing pavement, and suggests a 2.5 x 10 -7 cm/ESAL
rate of rutting criteria for delineating rutting and nonrutting pavement.
Since the proposed model for describing rut development with applied traffic is very
nonlinear, ratios of rut depth and several functions of ESAL's were also examined. The ratio
that was most promising was the ratio of rut depth to the square root of applied 80 kN
ESAL's. These ratios were computed using rut depths measured at the test sites,
tabulated in Table 3, and plotted as a histogram in Figure 7. This histogram provides a clear
distinction between good and poor performing pavements, and suggests a 5.0 x 10 -4
cm/,,/ESAL rate of rutting criteria for delineating rutting and nonrutting pavement. The
parameter, (rut depth)/Eq-ESAL,will be correlated with aggregate properties in the following
section.
ANALYSIS OF LABORATORY DATA
Mix from cores was separated into aggregate and asphalt cement components.
Gradation, fractured face counts on coarse aggregate (plus No. 8) fractions, and particle
shape and texture tests on fine aggregate (minus No. 8) fractions were conducted on
extracted aggregate. Data are tabulated in Table 4.
Uncompacted voids and flow time provide an indication of particle shape and texture
and were measured using apparatus and equipment proposed by the National Aggregates
Association (NAA) [9]. Uncompacted voids of graded samples were measured using
Method A. In addition, 400 gm samples of minus No. 8 material with mix gradations were run
through the apparatus and flow times recorded. These flow times provide an indication of
particle shape and texture, but are influenced by the actual gradation. With graded
samples, as specified in Method A, the effects of different gradations are eliminated.
After completion of laboratory testing a detailed statistical analysis was performed to

determine aggregate properties that are related to rutting. Properties analyzed included
percentages passing the 3/8 inch, No. 50 and No. 200 sieves; fractured face count for
coarse aggregate (plus No. 8) fractions; and uncompacted void content and flowtime for
fine aggregate (minus No. 8) fractions.
30 30
Rut depth from trench t

i I and I Rut depth from 1988 AHD I
cores at test I pavement condition data 25
251 [sites J D&se
o
I Selectedfor good
x performance
oE 2O ~ q Selectedfor poor 20
v
_d performance
<
co
LU
15
Z 15
-i- 10
t- 10
uJ
er 5
1 4 7 11 13 2 3 5 6 8 9 10 12 1 4 7 11 13 2 3 5 6 8 9 10 12
SITE
FIGURE 6 -- Rate of rutting at test sites.
200
R~depth Irom trerlch1

175 land cores at test site.j
I elected for good

o performance
x 150
~ electedfor poor
performance
125[
100F
t- 75
uJ
C3 fl
rr
: Ill 4 7 11 13
;ITE
8
,,
10 12
FIGURE 7 -- (Rut depth) ! -~S--AL at test sites.
TABLE 4 -- Properties of aggregate recovered from asphalt mixtures.
NAA** Flow** Fractures Faces on

Voids Time Plus NO, 8 Material (%!_ pCment Passina Sieve Size (~
Site Layer (%) (Sec) 0 1 2 or more 3/8 cm No. 50 No. 200
1+ 1 44.3 20.6 1.6 0 98.4 94 26 7.6

2 40.5 18.8 27.1 1.4 71.5 70 17 4.9
2 1 41.8 20.5 3.1 3.5 93.4 84 23 6.1

2 41.6 19.2 0.4 0.9 98.7 95 18 5.2
3 40.8 18.7 32.5 3.0 64.5 72 17 3.5
3 1 42.4 20.5 11.6 8.4 80.0 91 21 5.3

2 42.3 19.9 38.1 8.1 53.7 67 20 5.2
4+ 2* 42.0 19.6 2.3 1.5 96.2 90 28 6.2

3 43.2 19.1 39.8 6.8 53.4 68 25 6.7
5 1 41.9 20.1 8.4 10.6 81.0 71 21 7.2

2 41.6 19.2 19.6 0.0 80.4 92 26 7.8
3 41.6 19.5 8.6 0.8 90.6 85 21 5.3
6 1 42.6 21.3 31.0 11.1 57.9 90 18 8.1

2 42.5 20.9 17.1 0.2 82.7 94 12 4.2
3 43.1 21.1 68.9 9.5 21.5 68 10 2.6
7+ 1 43.7 22.6 0.0 0.0 100.0 86 18 7.0

2 43.4 21.4 2.1 0.0 97.9 95 15 4.7
3 44.6 21.1 4.0 0.4 95.5 77 18 7.8
8 1 45.9 23.9 1.9 0.0 98.1 92 24 9.1

2 45.0 21.1 0.0 0.0 100.0 77 22 7.9
9 1 44.4 20.7 30.8 11.4 57.8 91 26 5.7

2 43.3 19.8 74.8 10.6 14.6 93 33 6.6
10 1 41.6 19.5 47.3 7.0 45.6 92 24 8.1

2 42.6 20.7 36.4 0.0 63.6 93 22 10.7
11 + 1 45.2 24.3 10.0 0.0 90.0 97 17 8.1

2 46.1 23.1 0.0 0.0 100.0 81 16 9.6
12 1 40.8 19.7 32.6 6.9 60.5 90 19 5.8

2 45.2 26.7 0.0 0.0 100.0 74 14 7.7
13 + 1 43.6 21.8 41.8 2.1 56.1 94 19 6.0

2 45.0 24.3 0.0 0.0 79.7 63 12 6.7
+ Sites selected for good rutting performance.

* Layer 1 was a friction course.
** Test run on minus No. 8 material.
Aggreoate Gradation
Aggregate gradation definitely affects the rutting resistance of an asphalt mixture, but
is difficult to quantify. Studies have shown that the maximum aggregate size and percent
passing No. 200 sieve are important [10,11]. However, the evaluation of the influence of
individual sieve size particles is difficult. For this study the percent passing 3/8 inch,
percent passing No. 50, and percent passing No. 200 sieves were analyzed to determine
their affect on rutting. As shown in Table 5, the correlation coefficients and coefficients of
determination indicate weak correlations with trends that are occassionaly contrary to
expectations.
TABLE 5 -- Correlation coefficients between rut depth/~SAL and

aggregate properties.
Aggregate Property Correlation Coefficient, R
Passing 3/8 in., % -0.47
Passing No. 50, % 0.17
Passing No. 200, % 0.37
Fractu red Faces, % -0.13
Uncompacted Voids, % 0.15
Flow Time, sec. 0.05
The correlation coefficient between rut depth/,~[ESAL and percent passing the 3/8
inch sieve is -0.47. This indicates that an increase in percent passing the 3/8 inch sieve will
decrease rutting which is opposite the trend observed by others. A plot of the data in
Figure 8 shows that there is considerable scatter in the data. The high correlation
coefficient is basically the result of one data point that has a very low percent passing the
3/8 inch sieve and has very high rutting. It is concluded that a clear trend between rut
depth/~L and percent passing the 3/8 inch sieve is not shown in this study.
The second aggregate size investigated was the percent passing the No. 50 sieve.
The correlation coefficient of 0.17 (Table 5) and the data scatter shown in Figure 9 indicate
very little trend between rut depth/,,FES-AL and percent passing No. 50 sieve. The nature of
the trend is, however, as expected with rate of rutting increasing as percent passing the No.
50 sieve increases.
The correlation coefficient of .37 (Table 5) and the data scatter in Figure 10 indicate a
poor correlation between rut depth/ES,,fESALand percent passing the No. 200 sieve. As with
the No. 50 sieve, the nature of the exhibited trend is as expected.
0.0025-
Y = 0.004643 - 0.000042X 0.0025-
Y = 0.000294 + 0.000027X
R2= 0.22 R2= 0.03
d
0.0020- ._1
CO .< 0.0020-
Iii or) "13
ii UJ
IJ_ 50
O
t- 0 m
0 0.0015- F-
0.0015-
0 0 Z
n- ii 0 C~
13:
O
CO 0 0
O3
T 0.0010- z
t-- "l- 0.0010-
13_ F- O
III z
13._
a UJ "13
r-
t-- O m
x
:::D 0 , 0 0 0 5 - t--
rr 0.0005- r--
cc m
"o
0.0000- m
I I I 0.0000 z
70 80 90 100 12 1L6 20 2'4 2L8
PASSING 3/8 INCH SIEVE (%) P A S S I N G N O . 5 0 S I E V E (%) C
F I G U R E 8 -- Correlation with percent passing 3/8 inch FIGURE 9 -- Correlation with percent passing no. 50
sieve. sieve.
(30
03
CO
0.0025- 0.0025-
Y =-0.000363 + 0.000179X Y = 0.001183 - 0.000004X 60
"U
"I"
R2= 0.14 R 2 = 0.02
r-
-t
--J
.< 0.0020- 9 ..J 0.0020- x
9 < -I
O0 O0 c
LU '-n
LU m
Ii ii
O O 1"13
t- I-- 0.0015-
0 0.0015- 0
O -11
o
z
0
m
0.0010- 0.0010-
000 9 9
a 121
.~ 0.000S- ~.~ 0.0005
n" n"
o
0.0000 , i I i , L 0.0000 I I
4 5 6 7 8 9 10 40 60 80 100
P A S S I N G NO. 200 S I E V E (%) 2 OR MORE FRACTURE F A C E S (%)
F I G U R E 10 -- Correlation with percent passing no. 200 F I G U R E 11 -- Correlation with fractured face count.
sieve.
PARKER AND BROWN ON FLEXIBLE PAVEMENT RUI-FING 85
The two data points (Figures 8-10) with highest rate of rutting (Sites 5 and 8) are not
grouped with the other eleven data points. These sites have unusual characteristics which
explain this nonconformance. Site 5 was the most severely rutted (2.77 cm) site
investigated with significant plastic flow. Site 8 had received little traffic, but had very low in
place voids (0.6~ caused by very high asphalt content (7.8%). Omission of these two data
points only flattens the slopes of the regression lines and in most cases decreases the
correlation coefficients reinforcing the contention that there was no relationship between
gradation and rutting susceptibility. This is an indication of the complexity of rutting and the
multiplicity of factors that influence the process.
Fractured Faces
The fractured face count of aggregate affects a mixes' ability to resist rutting. Some
percentage of fractured aggregate is almost always specified for high volume roads, but
there is very little field data to support or contradict this typespecification. The correlation
coefficient between fractured face count and (rut depth)/~/ESAL for the study was -0.13
(Table 5). This is a very low correlation that shows only a slight trend toward less rutting for
higher fractured face count.
The correlation appears to be much better than this after reviewing Figure 11. The
two mixes with highest rate of rutting also had high fractured face counts (Site 5 - 81.0% and
Site 8 = 98.1%). However, as noted above, these mixes had low in-place voids and Site 5
was the most severely rutted site studied (rut depth = 2.77 cm) with rutting well into plastic
flow. Plastic flow had not started at Site 8, but the mix was characterized by very high
asphalt content. If the data for Site 8 is eliminated, for unrealistically high asphalt content,
the correlation coefficient becomes -0.41 indicating a much stronger trend. If both Site 5
and 8 are omitted, the correlation coefficient becomes -0.72.
The data in Figure 11 shows that all six mixes with fractured face percentages of 80 or
less had a (rut d e p t h ) / , J E ~ greater than .0005. Figure 11 also shows that four out of
seven mixes with a fractured face count greater than 80% had (rut depth)/E,/-ESALless than
.0005, including the mixes for Sites 5 and 8.
Fine Aaareaate ShaDe & Texture
Uncompacted voids from the NAA flow test [9] and flow time from the modified test
measure particle angularity and texture. Higher voids and flow times indicate rougher
textured and more angular particles. The correlation coefficients in Table 5 shows that flow
time from the modified NAA test has very little correlation (r = 0.05) with rutting.
Uncompacted voids from the NAA test provides a better (r = 0.15), but still very poor
correlation.
Figure 12 shows the weak trend for uncompacted voids. The trend is opposite that
expected and indicates an increase in voids will result in an increase in rutting. It appears
that again the data from Sites 5 and 8 distort the regression. If the data point for Site 8 is
omitted, for unrealistically high asphalt content, the correlation coefficient becomes -0.25,
indicating a stronger trend, but more importantly the sign of the correlation coefficient is
reversed indicating, as expected, that rate of rutting decreases as uncompacted voids
increases. If data from both Sites 5 and 8 are omitted, the regression ~ine is flattened and
the correlation coefficient becomes -0.14.
0.0025- 0.0025- o~
Y =-0.001502 + 0.000055X Y = 0.000468 + 0.000020X "o
"-r
R2= 0.02 R 2 = 0.00 I'--
--H
E:
<~ 0 . 0 0 2 0 - x
._J 0.0020-
< 09 c
O0 UJ
m
LLI 1.1_ 'u
U.. O m
O F- -rl
t- 0.0015- O 0.0015 0
O O
O n-
rr z
O o
O o0 m
00
0.0010- 9 "r- 0.0010
"r t-
t-- 13_
13_ UJ
U.J s 9 O(2)
C3 I-
I-- 0.0005- 0.0005
D f'r"
cc
0.0000 F I I I I I 0.0000
40 41 42 43 44 45 46 19 25 2'1 22 23 24 25
NAA UNCOMPACTED VOIDS (%) M O D I F I E D N A A FLOW T I M E (SEC)

FIGURE 12 -- Correlation with uncompacted voids. FIGURE 13 -- Correlation with flow time.
PARKER AND BROWN ON FLEXIBLE PAVEMENT RUI-rlNG 87
Figure 13 illustrates the very weak correlation for flow time. Again, if the data point for
Site 8 is omitted, the correlation coefficient becomes -0.37. This not only represents a
dramatic increase in strength of the correlation, but the change in sign means that the trend
is in the expected direction, i.e., rate of rutting decreases as flow time increases. The
performance of the mix at Site 8 demonstrates the multiplicity of factors that can influence
rutting performance, and the importance of both aggregate properties and asphalt content
during material selection and mix design. Omission of data from Site 5 only slightly flattens
the regression line with essentially no change in the correlation coefficient.
GEOGRAPHIC AND AGGREGATE PROPERTY RELATIONSHIPS
In the section on the analysis of rutting data from the AHD pavement condition
database, it was concluded that pavement rutting susceptibility was related to geographic
area and that variable geology and, thus, variable quality aggregate was the most probable
cause. Specifically it was concluded that pavements in Divisions 5-9 which are located in
the coastal Plain, where natural sands and gravels are used extensively, are more
susceptible to rutting than pavements in Divisions 1-4 which are located predominately in
the Appalachian Plateau and Piedmont geologic regions, where crushed stone is available.
Particle shape and texture are assumed to be indicators of aggregate quality and
several properties of the extracted aggregate from sites in Divisions 1-4 will be compared
with sites in Divisions 5-9. Average values for the properties are contained in Table 6.
TABLE 6 -- Average aggregate properties for divisions 1-4 and divisions 5-9.
Divisions 1-4 Divisions 5-9

Aggregate Property Sites 7,8 & 10-13 Sites 1-6 & 9
Fine Aggregate (-#8)

NAA Voids 44.1% 42.3%
Flow Time 22.3 sec. 20.0 sec.
Coarse Aggregate (+#8)

Two or More Crushed Faces
Surface Mixes 78.6% 77.7%
Binder Mixes 100.0% 52.9%
Combined Mixes 85.2% 70.4%
Coarse Aggregate (+#4)

Two or More Crushed Faces
Surface Mixes 87.4% 82.7%
Binder Mixes 100.0% 52.7%
Combined Mixes 91.3% 73.9%
For the fine aggregate fraction (minus No. 8), uncompacted voids from the NAA test
with graded samples and flow times on samples with mix gradation were compared.
Average uncompacted voids and flow times are larger for the sites in Divisions 1-4 indicating
more angular particles with rougher surface texture. This was the expected response since
both angularity and surface texture are indirectly related to degree of weathering or distance
transported.
Fractured face counts are a measure of particle angularity and were made on coarse
(plus No. 8) aggregate particles. These counts were also be used to compute angularity of
coarser fractions, i.e., plus No. 4 particles. Results on the plus No. 4 fractions are included
because AHD specifications for surface mix aggregate require that 80% of the particles
have two or more crushed faces. Percentages of particles having two or more crushed
faces for the plus No. 8 and plus No. 4 fractions are shown in Table 6. The numbers are
somewhat different, but both sizes indicate similar relationships between aggregate from
sites in Divisions 1-4 and sites in Divisions 5-9.
For surface mixes the percentages of particles with two or more crushed faces is only
slightly larger in Divisions 1-4 than in Divisions 5-9. This is due to the widespread use of
crushed gravel coarse aggregate statewide with only slightly more crushed granite and slag
used in Divisions 1-4.
For binder mixes the percentages of crushed aggregate in Divisions 1-4 are 100%
and approximately twice that in Divisions 5-6. This is due to the widespread use of crushed
limestone in Divisions 1-4. For combined mixes the percentages for both size fractions are
larger in Divisions 1-4 than in Divisions 5-6, indicating more angular course aggregate
particles in Divisions 1-4.
CONCLUSIONS
Analyses of the Department's pavement condition database indicate that rutting in

Alabama has increased slightly from 1984 to 1988, and that this increase is attributible to
either increased loading intensity or increased asphalt concrete rutting susceptibility. The
analyses also indicate that rutting varies geographically and that part of this variation may be
explained by quality of locally available aggregate. Those areas with crushed stone and
angular natural sands are less susceptible to rutting.
Analyses of data from field test sites indicate that permanent deformation causing
rutting is generally confined to the top 7 to 10 cm (surface and binder courses). There was
little evidence that lower base/subbase courses or subgrade were significant contributors to
rutting. At only one site was there evidence that stripping may have contributed to rutting.
There was some evidence that surface treatment layers used in conjunction with thin
overlays may have contributed to rutting susceptibility. A rate of rutting of 5 x 10-4
cm/~ESAL or 2.5 x 10-7 cm/ESAL delineated good and poor performing pavements.
Aggregate properties studied include gradation, coarse aggregate percent fractured

faces, and fine aggregate particle shape and texture. The correlation coefficient for most of
these properties were low, however, there appeared to be a definite trend shown in the
figures. The low correlation coefficients in almost every case were caused by one or two of
the data points being far outside the range of the other data. This demonstrates that rutting
is a very complicated process affected by a number of factors, and that aggregate properties
and asphalt content are important considerations during material selection and mix design.
ACKNOWLEDGEMENTS
The work reported in this article was sponsored and supported by the State of
Alabama Highway Department through the Highway Research Center at Auburn University.
The authors are grateful for this sponsorship and support.
REFERENCES
[1] Middleton, D.R., F.L. Roberts and T. Chira-Chavala, "Measurement and

Analysis of Truck Tire Pressures on Texas Highways," in Transportation Research
Record 1070, Transportation Research Board, Washington, D.C., 1986, pp. 1-8.
[2] Thompson, M.R., "Analytical Methods for Considering Tire Pressure Effects in
Pavement Design," Proceedings, A Symposium/Workshop on High Pressure
Truck Tires, American Association of State Highway and Transportation Officials
and Federal Highway Administration, Austin, Texas, February 1987.
[3] Kim, O., and C.A. Bell, "Measurement and Analysis of Truck Tire Pressures in
Oregon," in Transportation Research Record 1207, Transportation Research
[4] Hudson, S.W. and S.B. Seeds, "Evaluation of Increased Pavement Loading and Tire
Pressures," in Transportation Research Record 1207, Transportation Research
[5] , "High-Pressure Truck Tires," in p~vement Newsletter, Federal
Highway Administration, Washington, D.C., January 1988.
[6] Marshek, K.M., H.H. Chen, R.B. Connell and W.R. Hudson, "Experimental
Determination of Pressure Distribution of Truck Tire Pavement Contact," in
TransPortation Research Record 1070. Transportation Research Board,
Washington, D.C., 1986, pp. 9-14.
[7] Button, J.W., D. Perdomo, and R.L. Lytton, "Influence of Aggregate on Rutting of
Asphalt Concrete Pavements," in Transportation Research Record 1259.
Transportation Research Board, Washington, D.C.,1990, pp. 141-152.
[8] , "Pavement Condition Survey, Distress Identification and Rating
Manual," Bureau of Materials and Tests, Alabama Highway Department,
Montgomery, Alabama, July 1988.
[9] Meininger, R.C., "Proposed Method of Test for Particle Shape and
Texture of Fine Aggregate Using Uncompacted Void Content," National
Aggregates Association, Silver Springs, Maryland, March 1989.
[10] Brown, E.R., J.L. McRae, and A. Crawley, "Effect of Aggregate on Performance of
Bituminous Concrete," Implication of Aggregate in the Design, Construction, and
Performance of Flexible Pavement, SPecial Technical Publication 1016, ASTM,
1989, pp. 34-63.
[11] Brown, E.R. and Charles E. Bassett, "The Effects of Maximum Aggregate Size on
Rutting Potential and Other Properties of Asphalt-Aggregate Mixture," in
Transportation Research Record 1259. Transportation Research Board,
Washington, D.C., 1990, pp.
107-119.
Charles A. Sanders I and Ervin L. Dukatz 2
EVALUATION OF PERCENT FRACTURE OF HOT MIX A S P H A L T GRAVELS IN

INDIANA
REFERENCE: Sanders, C. A. and Dukatz, E. L., "Evaluation

of Percent Fracture of Hot Mix A s p h a l t Gravels in
Indiana", Effects of Aqqreqates and Mineral Fillers on
A s p h a l t mixture Performance, ASTM STP 1147, R i c h a r d C.
M e i n i n g e r Ed., American Society for Testing and Materials,
Philadelphia, 1992.
ABSTRACT: The Indiana Department of T r a n s p o r t a t i o n

prompted by major concerns in 1987 over p e r m a n e n t
d e f o r m a t i o n (rutting) decided to increase the level of one
face fractured gravel coarse aggregate particles in
asphalt mixtures. Previous levels of 40% for base course,
55% for binder course and 70% for surface were e s s e n t i a l l y
increased to 95% for all mixtures used in h i g h e r traffic
v o l u m e work. This paper subsequently addresses two
questions:
i. Is it always necessary to increase gravel fracture

p e r c e n t a g e to offset rutting and if so, what level
is appropriate?
2. Since both stone and gravel asphalt p a v e m e n t s in

Indiana have experienced rutting, what other m i x t u r e
p r o p e r t i e s and/or construction t e c h n i q u e s influence
r u t t i n g potential?
KEYWORDS: rutting, permanent deformation, gravel,

fractured gravel, crushed gravel, sand angularity, fines,
hot mix asphalt, marshall properties.
iDirector of Technical Services at V u l c a n Materials

C o m p a n y ' s M i d w e s t Division, Chicago, Illinois
2Manager of Technical Services at V u l c a n Materials

C o m p a n y ' s Southwest Division, San Antonio, Texas.
90
Dec 22 07:29:10 EST 2011
SANDERS AND DUKATZ ON HOT MIX ASPHALT GRAVELS 91
OBJECTIVES:
i. R e v i e w additional literature that i l l u s t r a t e s the

performance of p a r t i a l l y fractured gravels and
t e c h n i q u e s for rut r e s i s t a n t mixtures.
2. E v a l u a t e gravel sections that have been c o n s t r u c t e d

under the 40-55-70 crush count s p e c i f i c a t i o n and
identify the p e r f o r m a n c e factors.
LITERATURE REVIEW
F r a c t u r e Level
It is d i f f i c u l t to find c o m p r e h e n s i v e w o r k done on
p a r t i a l l y c r u s h e d gravels that doesn't introduce the e f f e c t s of
other variables. Hicks, A l b r i g h t and Lundy at the O r e g o n State
U n i v e r s i t y T r a n s p o r t a t i o n R e s e a r c h I n s t i t u t e in 1985 c o n c l u d e d
that " v a r i a t i o n in p e r m a n e n t d e f o r m a t i o n and m o d u l u s seems to
be i n f l u e n c e d more by asphalt content than by g r a d a t i o n and
fracture"[l]. They also noted that a s p h a l t c o n c r e t e fatigue
lives did not appear to be s i g n i f i c a n t l y a f f e c t e d by f r a c t u r e
level b e t w e e n 50 and 90%. This study was c o n d u c t e d for the
State of Alaska, and their intent was to d e t e r m i n e the
s u i t a b i l i t y of the current Alaska crush count s p e c i f i c a t i o n of
70%, w h i c h they concluded was very suitable. The study,
however, did not focus on an a p p r o p r i a t e t h r e s h o l d value. In
a later 1989 paper for ASTM (STP 1016) c o n t i n u i n g on the same
study by Lundy, Hicks, and McHattie, they c o n c l u d e d m o r e
d e f i n i t e l y that "reduction in p e r m a n e n t d e f o r m a t i o n a s s o c i a t e d
w i t h i n c r e a s i n g fracture levels was not d e m o n s t r a t e d at 10~
(50~ '' [2]. The crushed contents e x a m i n e d in these s t u d i e s
were 50, 70, and 90 percent.
In another study done by W e d d i n g and G a y n o r in 1961 for

A A P T 0, 50, 75, and i00 percent fracture levels were e x a m i n e d
[3]. The p r i m a r y thrust was the effect on t y p i c a l M a r s h a l l
properties. Their c o n c l u s i o n was that " r e q u i r i n g more than 75
p e r c e n t of the coarse aggregate p a r t i c l e s to be c r u s h e d m i g h t
only have the effect of increasing cost w i t h o u t any s i g n i f i c a n t
a d v a n t a g e to the stability of the mix".
H u b e r and Heiman attempted to relate the f r a c t u r e d faces

to both M a r s h a l l and Hveem p r o p e r t i e s as s e c o n d a r y i n f l u e n c e s
on r u t t i n g [4]. They go beyond most studies and c o n d u c t e d a
s t a t i s t i c a l analysis to d e t e r m i n e the m a j o r factors. Rutting
is d e f i n e d as a 20mm rut depth. They c o n c l u d e d that for
d i f f e r e n t levels of fractured faces "no d i v i s i o n a p p e a r s to
exist. A c c e p t a b l e and u n a c c e p t a b l e sites occur w i t h both high
and low fracture counts." After a t t e m p t i n g to c o r r e l a t e the
other influences of what makes a site acceptable or
u n a c c e p t a b l e , they assume that "a t h r e s h o l d v a l u e of 60 p e r c e n t
appears reasonable." One of the most i n t e r e s t i n g f i n d i n g s of
this study is that "fractured faces, voids in the m i n e r a l
92 ASPHALTMIXTUREPERFORMANCE
aggregate (VMA) and Hveem stability appear to be s e c o n d a r y
parameters. V i o l a t i o n of the threshold values for these
p a r a m e t e r s does not necessarily signify poor p e r f o r m a n c e . "
It is important to note that these studies a s s e s s e d

crushed faces on the plus 4.75 mm (No. 4) size m a t e r i a l only.
The gradation and angularity of the sands were not considered.
Sand Composition
Since about 1958, awareness has steadily grown over the

role angular sand plays in rut resistant mixes.
Wedding and Gaynor have focused on the effect of sand

shape on gravel asphalt mixtures [3]. They say, "It a p p e a r s
that in instances where very high percentages of c r u s h e d gravel
are specified, equivalent stabilities can be p r o d u c e d w i t h
crushed sand and moderate percentages of c r u s h e d coarse
aggregate." Also, "the use of crushed gravel sand in place of
natural sand is about equal in effectively raising s t a b i l i t y as
the use of 25% crushed gravel in the coarse aggregate" (Figure
i). The effect of sand particle shape is also r e p o r t e d by
Dukatz [5] and Gronhaug [6]. They both observed t h a t "the
greatest effect on stability was obtained when natural sand was
replaced by angular m a n u f a c t u r e d sand" [5, 6].
8000 -
Crushed
~7500
o
~, 70OO
/~tural Sand
(0
~5500
5000 '''I'TT'I .... [ .... 11'''~ .... I .... I .... ~ .... I i'' I
0 10 20 30 40 50 80 70 80 90 100
%QushedGravel
Figure i. % CRUSHED vs. STABILITY
Lottman and Goetz found that as little as 25% c r u s h e d

gravel fine aggregate produced a significant increase in
s t r e n g t h [7]. In 1979 the M i c h i g a n T r a n s p o r t a t i o n C o m m i s s i o n

undertook a study on "The Effects of Fine A g g r e g a t e on
S t a b i l i t y of B i t u m i n o u s Mixes" [8]. This was done as a follow-
up to their study c o n d u c t e d in 1977 that i n d i c a t e d in m o s t
cases, c r u s h e d content above 50% of the coarse a g g r e g a t e was of
no b e n e f i t to s t a b i l i t y [9]. Their number one c o n c l u s i o n was:
"Both the a n g u l a r i t y of the fine a g g r e g a t e and the

g r a d a t i o n of the m i x t u r e are critical for a h i g h
s t a b i l i t y mix. The more angular the fine aggregate,
the h i g h e r the stability. The a n g u l a r i t y of the
c o a r s e a g g r e g a t e is not decisive."
For any given gradation, VMA's are more d i f f i c u l t to o b t a i n

w i t h g r a v e l s than stone. Meier and Elnicky [I0] tell us that
"voids in the m i n e r a l aggregate increased w i t h i n c r e a s i n g
p e r c e n t a g e of c r u s h e d fine aggregate." Work done by K a l c h e f f
and T u n n i c l i f f show that this VMA d i f f e r e n c e b e t w e e n sands can
be as g r e a t as nearly 2.0% [ii]. Meier and E l n i c k y a g a i n found
that "the H v e e m s t a b i l i t y of asphalt c o n c r e t e m i x t u r e s was
found to be linearly related to the results of tests of fine
a g g r e g a t e shape and surface texture" [I0].
The FHWA suggests that natural sands should not e x c e e d 20%

in a m i x [12]. Other sources [3, 6, 7] say 15 to 30 p e r c e n t
s h o u l d not be exceeded. In Europe the rut r e s i s t a n t stone
m a s t i c (SMA) s p e c i f i c a t i o n s require a m i n i m u m of 50% c r u s h e d
m a n u f a c t u r e d sand in the fines fraction [13]. None q u a l i f y the
a n g u l a r i t y of the r e s t r i c t e d sands. Natural sands v a r y g r e a t l y
in a n g u l a r i t y from source to source. Also, some p r o c e s s e s add
c r u s h i n g s to their sand products from other p a r t s of the
process. Meier and Elnicky [i0] have identified three
a g g r e g a t e shape and surface texture tests that c o r r e l a t e with
H v e e m s t a b i l i t y levels and d e v e l o p e d linear e q u a t i o n s for each.
They are:
1. National Crushed Stone Voids Test

2. Rex and Peck Time Index
3. S p e c i f i c R u g o s i t y by P a c k i n g V o l u m e
Gradation
Rut r e s i s t a n c e of any asphalt m i x t u r e is h i g h l y d e p e n d e n t

u p o n a g g r e g a t e grading. Loss of stability, w h i c h can lead to
rutting, can in general occur when n o n - u n i f o r m g r a d a t i o n s
contain excesses of certain size fractions. The most
d e t r i m e n t a l effect is p e r m i t t i n g a g r a d a t i o n hump w i t h natural,
r o u n d e d sand. Dukatz at the 1989 AAPT S y m p o s i u m on the "Effect
of A g g r e g a t e s in A s p h a l t Mixes" urged that g r a d a t i o n s on a 0.45
p o w e r s e m i - l o g plot be u n i f o r m in shape. He p o i n t e d out that
even 100% c r u s h e d mixes will rut w i t h o u t p r o p e r g r a d i n g [14].
Fines Size and Q u a n t i t y
The fine sand fraction and e s p e c i a l l y the m i n u s 0.075 mm

(No. 200) m a t e r i a l is, as of late, looked upon a l m o s t as an
i m p u r i t y or contaminate. Some people that have seen the loss
of d u r a b i l i t y due to high fines are now o v e r - r e a c t i n g and are
t r y i n g to e l i m i n a t e them. M i x t u r e fines, however, still play
an i m p o r t a n t role in rut r e s i s t a n t d u r a b l e mixes, and m o s t
i n v e s t i g a t o r s agree that minus 0.075 mm (No. 200) levels
b e t w e e n d u s t - t o - a s p h a l t ratios of 0.6 and 1.2 are n e c e s s a r y [i,
2, 12, 15]. Hicks, A l b r i g h t and Lundy's w o r k at O r e g o n State
u n i v e r s i t y found a r e d u c t i o n in p e r m a n e n t d e f o r m a t i o n w i t h the
i n c r e a s e from three to six percent minus 0.075 mm (No. 200),
but the trend did not continue beyond six p e r c e n t [i]. Brown,
M c R a e and C r a w l e y also c o n c l u d e d that "there is e v i d e n c e that
the o p t i m u m amount of minus 0.075mm (No. 200) m a t e r i a l is
b e t w e e n three and six percent for a w e l l - g r a d e d b i t u m i n o u s
c o n c r e t e " [16]. R e s e a r c h by A n d e r s o n [15] has shown that the
v o l u m e c o n c e n t r a t i o n of minus 0.075 mm (No. 200) is a m o r e
r e p e a t a b l e p a r a m e t e r to determine the effect of m i n u s 0.075 mm
(No. 200) in the mix.
A r e p o r t by Khedaywi and Tons [17] states that there is an

o p t i m u m fines size for every type (surface roughness) of c o a r s e
aggregate. "The higher the r u g o s i t y (or s u r f a c e roughness) of
the coarse aggregate, the larger must be the o p t i m u m size of
fines to give the best i n t e r l o c k i n g and the h i g h e s t s t r e n g t h
for the mix." They say a crushed limestone w o u l d need a 0.250
mm (No. 60) to 0.180 mm (No. 80) size m a t e r i a l and an all-
r o u n d e d smooth gravel would need 0.075 mm (No. 200) to 0.053 mm
(No. 270) size fines. L o g i c a l l y speaking, the p a r t i a l l y
c r u s h e d gravel mixes may then throw o p t i m u m fines size into
s o m e t h i n g like the 0.i00 mm (No. 150) to 0.075 mm (No. 200)
range. The report also indicates that a s p h a l t films s h o u l d be
e s t a b l i s h e d in a c c o r d a n c e with the fines thickness.
Other Properties
M a n y r e s e a r c h e r s felt that certain other p r o p e r t i e s were

i m p o r t a n t in c o n t r o l l i n g rutting.
i. H u b e r & Heiman [4]: P r i m a r y P a r a m e t e r s
a. A s p h a l t C o n t e n t - 5.1% max.
b. Air Voids - 4% min.
c. Voids Filled - 70% max.
Secondary P a r a m e t e r s
a. VMA - 13.5% min.
b. F r a c t u r e d Faces - 60% min.
c. Hveem s t a b i l i t y - 37 min.
2. Ford & H e n s l e y [18]: a. Air V o i d s - 4% min.

b. M a r s h a l l Flow - 6 max.
c. M a r s h a l l S t a b i l i t y - 7784 N min.
d. Voids Filled - 65-75%
3. B r o w n & Cross [19]: a. GSI-I.I max. @ 1~ 828 K Pa

(120 PSI), 300 Rev.
b. M a r s h a l l Flow - 16 max.
c. Low Air V o i d s - m a j o r c a u s e of
rutting
d. A s p h a l t C o n t e n t - m u s t be p r o p e r l y
d e s i g n e d and c l o s e l y controlled.
4. Monismith &
T a y e b a l i [20]: a. The creep test a p p e a r s to be a
useful tool in e s t i m a t i n g r u t t i n g if
the Shell r e s e a r c h c o n c e p t s are used.
b. Gravel mix creep v a l u e s a c t u a l l y
exceeded the g r a n i t e value, but at
lower air voids, a s p h a l t content, and
VMA than the granites.
5. Brown [21] : R u t t i n g is p r i n c i p a l l y c a u s e d by
a. Too much asphalt.
b. A g g r e g a t e sizes too fine
(Mixture top size)
c. Q u a l i t y control is l a c k i n g
d. Too m u c h minus 0.075 mm (No. 200)
(Refer to ratio on p. 4)
e. Inadequate compaction
A c o n c l u s i o n that can be r e a c h e d from this s u m m a r y of

f a c t o r s i n f l u e n c i n g mix stability is that p e r m a n e n t d e f o r m a t i o n
(rutting) cannot be tied to only one parameter. Various
r e s e a r c h e r s have p r e s e n t e d data i n d i c a t i n g that the i n f l u e n c e
of a s p h a l t binder p r o p e r t i e s is s e c o n d a r y to the e f f e c t s of
a g g r e g a t e p a r t i c l e shape, top size and g r a d a t i o n in p r e d i c t i n g
rutting.
FIELD I N V E S T I G A T I O N
F i e l d i n v e s t i g a t i o n s made in the fall of 1989 r e v e a l e d

that s e c t i o n n u m b e r I., 1-74 between routes S.R. 25 and U.S. 41
h a d f a i l e d w h i l e sections II., III., and IV. w e r e p e r f o r m i n g
well. S e c t i o n number I. was not c o n s t r u c t e d as an Indiana
Q u a l i t y A s s u r a n c e section, while sections II., III., and IV.
were. All four sections were c o n s t r u c t e d under the I n d i a n a
m i n i m u m c r u s h e d content s p e c i f i c a t i o n of 40% for base course,
55% for b i n d e r course, and 70% for surface course. Information
on the four s e c t i o n s and their p e r f o r m a n c e f a c t o r s are as
follows.
I. 1-74 B e t w e e n State Road 25 and U.S. Route 4~
Age: 4 year
Daily Traffic: 11,200 v e h i c l e s
This s e c t i o n was c o n s t r u c t e d from late 1985 into early

1986 and began showing severe rutting soon after
construction. Surface m i l l i n g had been done in 1987 to
relieve some of the water b u i l d - u p at the edge of
pavement. Rut depth m e a s u r e m e n t s t a k e n in A u g u s t of 1989
a v e r a g e 0.74, 1.12, 0.66 and 1.12 cm in each t r a v e l lane
wheelpath. Of the 40 individual m e a s u r e m e n t s taken, the
w o r s t was 2.54 cm.
Since this project was not c o n s t r u c t e d under the I n d i a n a

Q u a l i t y A s s u r a n c e specifications, t e s t i n g for m a r s h a l l
properties on the produced mixtures was not done.
However, i n d i v i d u a l s and job d o c u m e n t s have p r o v i d e d the
f o l l o w i n g information.
i. Job mix formulas for this section were e s t a b l i s h e d

from h i s t o r y or common p r a c t i c e and not from d e s i g n s
u s i n g d e s i r e d m e a s u r e d m i x t u r e properties.
2. Only one coarse a g g r e g a t e and one sand has been u s e d

w i t h each mixture, thus r e s t r i c t i n g the o p p o r t u n i t y
for a u n i f o r m l y graded mix.
3. Both base and binder actual produced mixtures

e x h i b i t a slight natural sand g r a d a t i o n hump w h e n
plotted, and the gap g r a d i n g of the c o a r s e a g g r e g a t e
does not provide much angular m a t e r i a l to the sand
fraction.
4. A gradation plot encompassing upper and lower

e x t r e m e s of the surface m i x t u r e tests shows the
curve to be much too close to the m a x i m u m d e n s i t y
line to be used for heavy t r a f f i c roadways. The
gradation tests for the finer m i x e s exhibit a
p r o n o u n c e d natural sand hump.
5. Five sets of p a v e m e n t cores that were t e s t e d in the

IDOH laboratory revealed that a significant
difference in actual "in-place" voids existed
b e t w e e n center of lane locations and the w h e e l p a t h s .
A v e r a g e values for the w h e e l p a t h s w e r e lower than
the center of lane by 0.7% for s u r f a c e course, 1.1%
for binder course, and a full 2.0% for base course.
This kind of additional d e n s i f i c a t i o n u n d e r t r a f f i c
d u r i n g the infancy of the p a v e m e n t indicates a
p o t e n t i a l for rutting w h i c h can r e l a t e to d e s i g n
t e c h n i q u e s and field compaction, r a t h e r t h a n just
m a t e r i a l characteristics.
6. C o n s t r u c t i o n conditions were poor d u r i n g p l a c e m e n t

of the base and binder courses. Temperatures ranged
b e t w e e n 15 ~ and -3~ with over half of the days only
r e a c h i n g the 0 ~ to 5~ for a high. The b i n d e r u n d e r
t r a f f i c began showing immediate d i s t r e s s and was
sealed with applied asphalt and a b l o t t e r sand
d u r i n g the w i n t e r months. In early 1986, some
m i l l i n g of the binder surface and p o t h o l e f i l l i n g
o c c u r r e d prior to p l a c e m e n t of the surface. How
m u c h excess sand and asphalt r e m a i n e d in the v o i d s
of the binder course is unknown.
II. 1-74 N.W. of B o o n e - M o n t q o m e r y Co. Line
Age: 1 year (2 summers on binder)

D a i l y Traffic: 11,200 v e h i c l e s
Properties: No. 8 Binder

Air Voids 6.0%
VMA 15.1%
Stability 6672 N
Flow 8.5
Crush Content 55 to 76%
All of the M a r s h a l l p r o p e r t i e s for this s e c t i o n from the

mix d e s i g n are adequate. Plots for the a s - p r o d u c e d mix
c o m p o s i t i o n indicate low v a r i a b i l i t y and good shape. It
a p p e a r s from this picture, along with c o n s i s t e n t a s p h a l t
c o n t e n t s that all the desired p r o p e r t i e s s h o u l d h a v e been
maintained.
The sand fraction has a u n i f o r m gradation, and n e g l e c t i n g

the u n k n o w n a n g u l a r i t y of the natural sand, it is a s s u m e d
that other c o n s t i t u e n t s c o n t r i b u t e about 30% c r u s h e d
m a t e r i a l below the 4.75 mm (No. 4) sieve as a p e r c e n t of
the total minus 4.75 mm (No. 4) material.
A f t e r two summers, the p a v e m e n t structure, w h i c h c o n s i s t s

of 2.54 cm of slag surface over 19.05 cm of No. 8 g r a v e l
binder, shows very little p e r m a n e n t d e f o r m a t i o n . Rut
m e a s u r e m e n t s average 0.15, 0.i0, 0.23 and 0.23 cm. in each
travel lane wheelpath. This minimal amount of
deformation would most likely be attributed to
c o n s o l i d a t i o n of the surface course w h e e l p a t h s u n d e r
traffic.
III. 1-65 N. of T i p p e c a n o e - W h i t e Co. Line
Age: 2 years
D a i l y Traffic: 21,000 vehicles
Properties: No. 5 Base No. 9 Binder

Air Voids 6.0% 6.0%
VMA 14.5% 15.0%
Stability 5671 N 5671 N
Flow 7 7
Crush Content 48 to 66% 64 to 69%
All of the M a r s h a l l p r o p e r t i e s for this s e c t i o n are

acceptable, and g r a d a t i o n plots of the a s - p r o d u c e d mix are
excellent. The asphalt contents are c o n s i s t e n t and the
d e s i r e d p r o p e r t i e s should have been m a i n t a i n e d t h r o u g h o u t
placement.
The sand g r a d a t i o n is u n i f o r m and the a m o u n t of c r u s h e d

m a t e r i a l p a s s i n g the 0.75mm (No. 4) sieve, other than w h a t
is found in the natural sand, e x p r e s s e d as a p e r c e n t a g e of
the minus 0.75 mm (No. 4) material, is a s s u m e d to be about
53% for base course and 32% for binder course.
A f t e r two years of service, the p a v e m e n t structure, w h i c h

c o n s i s t s of 2.54 cm of slag surface over 3.81 cm of b i n d e r
and 6.35 to 12.70 cm of base course, shows only slight
p e r m a n e n t deformation. Rut m e a s u r e m e n t s a v e r a g e 0.i0,
0.15, 0.20 and 0.18 cm in each travel lane wheelpath.
IV. 1-65 N. of State Road 26 to 1.3 mi. N. of State Road 25
Age: 2 years
Daily Traffic: 28,700 v e h i c l e s
Properties: No. 5 Base No. 9 Binder
Air Voids 6.0% 6.0%
VMA 14.5% 15.0%
Stability 5671 N 5671 N
Flow 7 7
Crush Content 48 to 66% 64 to 69%
All M a r s h a l l p r o p e r t i e s are acceptable. P l o t s of the

produced mix indicate good potential and with the
c o n s i s t e n t asphalt contents shown, indicate the d e s i r e d
p r o p e r t i e s were m a i n t a i n e d t h r o u g h o u t placement.
The sand g r a d a t i o n is uniform, and the amount of c r u s h e d

m a t e r i a l p a s s i n g the 0.75 mm (No. 4) sieve other than w h a t
is found in the natural sand, e x p r e s s e d as a p e r c e n t a g e of
the m i n u s 0.75 mm (No. 4) material, is a s s u m e d to be about
53% for base course and 32% for binder course.
A f t e r two years of service, the p a v e m e n t structure, w h i c h

c o n s i s t s of 2.54 cm of slag surface over 3.81 cm of b i n d e r
and 6.35 to 12.70 cm of base course, shows a l m o s t no
p e r m a n e n t deformation. Rut m e a s u r e m e n t s a v e r a g e 0.03,
0.08, 0.00, and 0.03 cm each travel lane w h e e l p a t h .
C O N C L U S I O N S AND R E C O M M E N D A T I O N S
S u m m a r y and C o n c l u s i o n s
Four i n t e r s t a t e h i g h w a y hot mix a s p h a l t o v e r l a y p r o j e c t s

of a b o u t the same age were e x a m i n e d for c u r r e n t c o n d i t i o n and
serviceability. Two p r o j e c t s were two years old, and one was
s l i g h t l y over a year, but has had two summers on the b i n d e r
course. A fourth project was four years old, but had severe
r u t t i n g symptoms in the first year and had to be p a r t i a l l y
m i l l e d to r e m a i n serviceable. Normally, it w o u l d be e x p e c t e d
that at the same age, rutting would increase w i t h the level of
traffic, but as shown in Figure 2, this did not happen. The
o p p o s i t e is shown, indicating that other o v e r r i d i n g v a r i a b l e s
are i n f l u e n c i n g performance.
The w e s t e r l y section of 1-74 is a m o n g s t the lowest in

traffic, but has the worst rutting because:
i. The mixtures were not designed on performance

parameters.
2. The m i x t u r e s are gap graded and lack a n g u l a r c o a r s e
sand.
3. A natural sand hump appears in the lower lifts.
4. The surface m i x t u r e has very low i n - p l a c e v o i d s and
a severe sand hump.
5. The binder mixture did not meet the crushed
r e q u i r e m e n t of the other three projects.
6. Construction conditions did not promote good
p a v e m e n t performance.
On the other hand, the other three s e c t i o n s had b i n d e r and

base m i x t u r e s with d e s i r a b l e levels of air voids, V M A and flow.
Also:
i. Mix g r a d a t i o n s were u n i f o r m w i t h o u t humps.

2. G r e a t e r p e r c e n t a g e s of c r u s h i n g s were e v i d e n t in the
sand fraction.
W i t h m u l t i p l e m a t e r i a l s in the p e r f o r m i n g mixtures, the

c o a r s e a g g r e g a t e and screenings p r o v i d e d s i g n i f i c a n t l y more
m i n u s 0.075 m (No. 4) c o m p a r e d to the r u t t a b l e mixtures. No
a s s u m p t i o n has been made about the natural sands in any of
these mixtures, which can h a v e - - a n d may have had, in some
c a s e s - - e n o u g h fractures and/or a n g u l a r i t y to have a d e q u a t e l y
aided in gravel performance. Since the surface c o u r s e s of the
v a r i o u s p r o j e c t s differed, and since most i n v e s t i g a t o r s agree
that m o s t s i g n i f i c a n t rutting takes place in the u p p e r 7.62 to
10.16 cm, a c o m p a r i s o n of rut depths and b i n d e r c o u r s e m i n u s
0.075 m (No. 4) crushings was made (see Fig.3). As expected,
m o r e r u t t i n g takes place with more r o u n d e d sand.
1.25 "
I. 1-74
-I.|2 RD
t.O0 No Q.A
0.75
Rut Depth
(cm)
0,50
"m'. 1-65
Q.A,
0.20 RD
0.25 n DR RD
00
11250 21000 28750
Average Daily Vehicles
Figure 2. R U T D E P T H vs. T R A F F I C
( M o s t S e v e r e W h e e l Path)
0 5 10 15 20 25 30 85 40
% Crushed Gravel
Figure 3. R U T D E P T H vs. % C R U S H E D SAND

(Most S e v e r e W h e e l P a t h , Binder Course)
Since rutting potential in low shear strength mixes will

usually become known within the first year, an example like 1-
65 between S.R. 25 and S.R. 26 that is two years old and has
28,700 vehicles per day with virtually no rutting to date,
certainly supports this conclusion.
Recommendations
i. Further research is needed to determine how

gradation and the total combined measured angularity
of the mix affects mixture stability.
2. Either natural sand of sufficient angularity or

manufactured sand should be used in all medium and
high traffic volume mixtures. The NCSA Voids Test,
the Time Index Test, or the Packing Volume Test
referenced to the appropriate Hveen stability level
should adequately assess the total sands in a
mixture.
3. The broadly accepted minus 0.075 mm (No. 200) to

asphalt ratio limits of 0.6 to 1.2 should be used.
Without enough "dust" in the mixture for proper
initial asphalt stiffening and dispersion, many
pavements of an early age are at undue risk. Proper
levels of minus 0.075 mm (No. 200) and 0.150 mm (No.
i00) are also needed for gravels because of their
rugosity.
4. Indiana should continue implementing the other

aspects of their current Quality Assurance Program
for the benefit of quality pavement construction.
ACKNOWLEDGEMENTS
Many thanks are extended to Messrs. Don Lucas, Ken Hoover,

Art Rucker, and Ron Walker of the Indiana Department of
Highways for their cooperation in providing information for
this report. Also, the efforts of the Vulcan Materials Company
Research and Development staff in obtaining research literature
is appreciated. The Indiana Department of Highways Research
Section has provided highway traffic data, and the Illinois
Department of Transportation donated use of rut depth
measurement equipment. The advice and guidance of the Indiana
Mineral Aggregates Technical Committee is also appreciated.
REFERENCES
[1] Hicks, R.G.; Albright, S. and Lundy, J.R.; "Evaluation oJ

Percent Fracture and Gradation on Behavior of Asphall
Concrete and Aggregate Base." By the Transportatiol
Research Institute, Oregon State University, Corvallis,
Oregon. Paper prepared for the State of Alaska Departmenl
of Transportation and Public Facilities Research Section,
Fairbanks, Alaska. Report No. AK-RD-86-25A.
[2] Lundy, J.R.; Hicks, R.G. and McHattie, R.; "Evaluation oi

Percent Fracture and Gradation on the Behavior of Asphalt
Concrete Mixtures." ASTM STP 1016, Philadelphia, 1989~
pp. 120-143.
[3] Wedding, P.A. and Gaynor, R.D.; "Effects of using Crushec

Gravel as the Fine and Coarse Aggregate in Dense Gradec
Bituminous Mixtures." Proceedings of the Association oJ
Asphalt Paving Technologists, Volume 30, 1961.
[4] Huber, G.A. and Heiman, G.H.; "Effects of Asphalt Concret~

Parameters on Rutting Performance: A Fielc
Investigation." Paper for the Saskatchewan Department oi
Transportation, Regina, Saskatchewan.
[5] Dukatz, E. L.; "Materials Selection and Mix Design fol

Municipal Use". Presentation at the 2nd Annua3
International Roads and Bridges Maintenance anr
Rehabilitation Conference, Atlanta, GA, October 14, 1988.
[6] Gronhaug, A.; "Evaluation and Influence of Aggregat~

Particle Shape and Form." Paper from the Norwegiar
Research Laboratory, Oslo, Norway; Medd. N.R. 31.
[7] Lottman, R. P. and Goetz, W. H.; "Effects of Crusher

Gravel Fine Aggregate on the Strength of Asphalt Surfacinc
Mixtures". Thesis for Master of science Degree submittec
to Purdue University, January 1956.
[8] Moore, R.B. and Welke, R.A.; "Effects of Fine Aggregate or

Stability of Bituminous Mixes." Michigan Department oJ
Transportation, Lansing, Report No. 78TB-34-79F, Decembe]
1979.
[9] Gaudette, B.E. and Welke, R.A.; "Investigation of Crushec

Aggregates for Bituminous Mixtures." Michigan Departmenl
of Transportation, Bituminous Technical Services Unit, An!
Arbor, Report No. TB-58, April 1977.
[i0] Meier, W.R. and Elnicky, E.J.; "Laboratory Evaluation of

Shape and Surface Texture of Fine Aggregate for Asphalt
Concrete." Transportation Research Board. Prepared for
68th Annual Meeting, Washington, DC, January 22-26, 1989.
[ii] Kalcheff, I.V. and Tunnicliffe, D.G.; "Effects of Crushed

Stone Aggregate and Shape on Properties of Asphalt
Concrete." Proceedings of the Association of Asphalt
Paving Technologists, Volume 51, 1982.
[12] "Report of the AASHTO Joint Task Force on Rutting."

American Association of State Highway and Transportation
Officials, Washington, DC, February 1989.
[13] German Supplemental Technical Specifications and

Guidelines o n Asphalt Surface Courses, ZTVbit S & B 84,
Der Bundesminister, FGR, Verkehr, 1990.
[14] Dukatz, E. L.; "Aggregate Properties Related to Pavement

Performance". Proceedings of the Association of Asphalt
Paving Technologists, Volume 58, 1989.
[15] Anderson, D.A; " Guidelines on the Use of Baghouse Fines".

National Asphalt Pavement Association, Information
Services i01, November 1987.
[16] Brown, E.R.; McRae, J.L. and Crawley, A.B.; "Effect of

Aggregates on Performance of Bituminous Concrete." ASTM
STP 1016, Philadelphia, 1989, pp. 34-63.
[17] Khedaywi, T. and Tons, E.; "Aggregate Rugosity and Size

Effect on Bituminous Mixes." Transportation Research
Board, 68th Annual Meeting, Washington, DC, January 22-26,
1989.
[18] Ford, M.C. and Hensley, M.J.; "Asphalt Mixture

Characteristics and Related Pavement Performance."
Proceedings of the Association of Asphalt Paving
Technologists, Volume 57, 1988.
[19] Brown, E.R. and Cross, S.A.; "A Study of In-Place Rutting
of Asphalt Pavements." Proceedings of the Association of
Asphalt Paving Technologists, Volume 57, 1988.
[20] Monismith, C.L. and Tayebali, A.A.; "Permanent Deformation

(Rutting) Considerations in Asphalt Concrete Pavement Sec-
tions." Proceedings of the Association of Asphalt Paving
Technologists, Volume 57, 1988.
[21] Brown, E.R.; "Asphalt Content, Agg Size, Bad Actors in

Rutting." Roads and Bridges Magazine, March 1989.
Mineral Fillers
A b d u l R a h m a n Al-Suhaibani, l Jamal Ai-Mudaiheem, l
and Fahd A i - F o z a n I
EFFECT OF FILLER TYPE AND CONTENT ON PROPERTIES OF ASPHALT CONCRETE

MIXES
REFERENCE: AI-Suhaibani, A., Ai-Mudaiheem, J., and AI-Fozan, F., "

EFFECT OF FILLER TYPE AND CONTENT ON PROPERTIES OF ASPHALT CONCRETE
MIXES," Effects Of Aqqreqates and Mineral Fillers On Asphalt Mixtures
Performance, ASTM STP 1147, Richard C. Meininger, Ed., A m e r i c a n Society
for Testing and Material, Philadelphia, 1992
ABSTRACT: Some of the newly constructed highway pavements in the

K i n g d o m of Saudi Arabia have shown premature failures with consequential
negative impacts on both roadway safety and economy. A m a j o r type of
these failures is permanent deformation (rutting). Fillers were
suspected to be major contributors to rutting susceptibility. The
effect of filler type and content on the rutting potential of asphalt
concrete as well as other mixes properties was investigated. The wheel
t r a c k i n g test (LCPC) was employed to investigate s u s c e p t i b i l i t y of these
mixes to rutting. The mechanical properties of mixes were studied using
tests such as resilient modulus, indirect tensile strength, H v e e m
stability and Marshall criteria. Results of this study indicate that
partial replacement of limestone dust by hydrated lime or portland
cement aggravates resistance of the mixes to rutting. Furthermore, an
analytical model was developed for predicting rutting potential based on
various mixes properties.
KEYWORDS: permanent deformation, rutting, limestone dust, hydrated

lime, p o r t l a n d cement, resilient modulus, indirect t e n s i l e strength,
wheel tracking test
INTRODUCTION:
Saudi Arabia underwent a rapid rate of development in every aspect

of life that was dominated by the c o n s t r u c t i o n boom that started two
decades ago. This resulted in increased traffic volume and higher than
design load magnitudes. These heavy loads coupled with the high ambient
temperature, are undoubtedly the primary factors c o n t r i b u t i n g to the
development of premature pavement distresses of which the permanent
d e f o r m a t i o n (rutting) is the most severe.
R u t t i n g is defined as the progressive a c c u m u l a t i o n of permanent
d e f o r m a t i o n of each layer of the pavement structure under repetitive
loading. However, rutting in the K i n g d o m was found to be confined to
asphalt bound layers only [!,~]- The magnitude and rate of pavement
rutting depends on several factors. The amount of fillers used and their
IAssistant and Associate Professors, and Graduate Student,

Civil Eng. Dept., College of Eng., King Saud University, P.O.Box 800,
Riyadh 11421, Saudi Arabia.
107
Copyright by
Copyright 9 ASTM
1992 Int'l
(all rights
by ASTM
International
07:29:10 EST 2011
c h a r a c t e r i s t i c s are among the major factors w h i c h affect the pavement

p e r f o r m a n c e and resistance to rutting [!,~-~]-
Fillers are included in paving mixtures to impart greater
stability and strength. There are two theories to e x p l a i n the
stabilizing effect of fillers. A c c o r d i n g to the first one, filler
(because of its fine particles) serves to fill the voids b e t w e e n
aggregate particles, thereby increasing the density and strength of the
compacted mixture. The second theory presumes that the fine particles of
filler become suspended in the asphaltic binder forming a mastic. The
suspended filler particles adsorb asphaltic components, thereby
increasing the viscosity of the binder and, consequently, to toughness
of the mixes [~]. However, it is believed that both roles are played by
fillers s i m u l t a n e o u s l y [ 6 ] .
It is an accepted fact that tHx mechanical p r o p e r t i e s of an
asphaltic mix to a large extent depend on the p r o p e r t i e s of the binder.
R e s i s t a n c e of an asphaltic mix to deformation can be favorably
influenced by using a h i g h - v i s c o s i t y binder. Filler type is found to
affect the susceptibility of asphaltic mix to permanent d e f o r m a t i o n [3].
This is due to the fact that different types of fillers have different
effects on viscosity of binder. It is reported that some fillers
s u b s t a n t i a l l y affect the viscosity of the asphalt binder which, in turn,
increases the stiffness of asphalt concrete mixes [6-13]. However, stiff
mixes may require an extra compaction to avoid rutting. It is,
therefore, p o s t u l a t e d that there is a critical filler content which
reduces the rutting potential affecting the other properties.
This study is undertaken to investigate the effect of filler type
and content on the properties of asphalt concrete mixes, e s p e c i a l l y the
rutting potential. The other properties of asphaltic mixes studied are
resilient modulus, indirect tensile strength, H v e e m stability and void
properties.
The rutting potential was studied using the wheel t r a c k i n g
equipment (LCPC) which simulates actual field conditions. D e t e r m i n a t i o n
of the resistance to rutting in the laboratory by a wheel t r a c k i n g test
has some advantages [14]. Firstly, it permits testing of large size
specimens. Secondly, the method of compaction simulates that of the
field.
MATERIALS CHARACTERIZATION
The materials used in this study, namely, asphalt cement,

aggregates, and fillers were characterized using routine type of tests
and results were compared with Ministry of C o m m u n i c a t i o n (MOC)
specifications.
Asphalt Cement
The asphalt cement used in this study is of 60/70 p e n e t r a t i o n

grade. It was obtained from the Riyadh refinery. The asphalt p r o p e r t i e s
are shown in Table i.
Aqqreqates
The aggregate used in this study was crushed limestone

o b t a i n e d H x r o m the Manwa Crusher Plant located in Wadi Laban, west of
Riyadh. This aggregate is widely used in the Riyadh area for asphaltic
mixes. The coarse and fine aggregates used in this study were sieved and
r e c o m b i n e d in the proper proportions to meet the w e a r i n g course
g r a d a t i o n required by MOC specifications [15]. The gradation curve for
the aggregates is shown in Figure i.
R o u t i n e tests were performed on the aggregates to evaluate their
physical properties. The results together with the s p e c i f i c a t i o n limits
as set by the MOC [15] are summarized in Table 2. Tests results show
that the chosen aggregate met the MOC specifications.
AL-SUHAIBANI ET AL. ON ASPHALT CONCRETE MIXES 109
Fillers
The filler, as defined in this study, is the material p a s s i n g the

# 200 (0.075 mm) sieve. Three different filler types; limestone dust,
hydrated lime, and portland cement were selected for this study.
Particle size of all the fillers was finer than # 200 (0.075 mm) sieve.
L i m e s t o n e dust was obtained from the Manwa Crusher Plant, w h i l e
hydrated lime (calcium hydroxide Ca (OH)2) was obtained from Saudi Lime
Brick and B u i l d i n g Materials Company. Portland cement was obtained from
A i - Y a m a m a h Cement Company. The specific gravities of these fillers were
2.713, 2.307, and 3.150 respectively.
TABLE 1--Properties of Asphalt Cement
Test ASTM Test MOC"

Designa- Results Specifica-
tion tion
Oriqinal Asphalt
P e n e t r a t i o n @ 25~ D-5 63 60-70
i00 gm, 5 seconds, 0.i
mm
Kinematic viscosity @ D-2170 377.5 200 min.
60~ poises
Absolute viscosity
@ 60~ poises D-2171 2203 NA--
Softening point, ~ D-36 49.0 NA
Ductility @ 25~ D-II3 i00+ Ii0 min
5 cm/min, cm
Flash point, ~ D-92 335 232.2 min.
Fire point, ~ D-92 361 NA
Specific gravity D-70 1.034 NA
Residue from TFOT =

Penetration @ 25~ D-5 41 NA
i00 gm, 5 seconds,
0.1 m m
R e t a i n e d Penetration, %
of original 65.1 52 min.
Kinematic viscosity @
135~ centistoke
A b s o l u t e viscosity D-2170 503 NA
@ 60~ poises D-2171 4692.6 NA
*MOC = M i n i s t r y of Communications, K i n g d o m of Saudi A r a b i a
**TFOT = Thin Film Oven Test
***NA = Not A p p l i c a b l e
The control mixes were prepared using limestone dust at contents

of 3, 5.5, and 8 % by total weight of aggregate. These contents cover
the filler range set by M O C specifications. M i x e s in w h i c h limestone
dust was partially replaced by either hydrated lime or p o r t l a n d cement
were also prepared. The replacement percentages were I, 2, and 3% by
total weight of aggregate. It should be m e n t i o n e d here that the %
p a s s i n g each sieve has been adjusted to reflect the change in the filler
content. This was done to satisfy the requirement that the total
p e r c e n t a g e retained should be 100%. Table 3 shows the gradations
c o r r e s p o n d i n g to each filler content.
EXPERIMENTAL WORK
The experimental work was started by d e t e r m i n i n g the o p t i m u m

asphalt content for all the mixes using the Marshall mix design method
100
Spec. Limits /~
80 5etected Grad. //// I
6O
///
a
a. 40
l/~f//
20
I I i i _ I I
O
U.5. Standard Sieve Size

Fig. l_-Aggregate gradation
TABLE 2--Physical Properties of A q q r e q a t e s
Property ASTM Test MOC"

Designa- Results Specifica-
tion tion
Coarse aqqreaate
Bulk S.G. C-127 2.561
Bulk S.G. (S.S~ 2.601
Apparent S.G. 2.668
Absorption, % 1.57
Fine aqqreqates
Bulk S.G. C-128 2.486
Bulk S.G. (S.S.D.) 2.564
Apparent S.G. 2.696
Absorption, % 3.12
Specific gravity of D-854

filler 2.713
Sand equivalent, % D-2419 45 Min 45%
Percent loss by Los

Angeles abrasion C-131
Grade B 24.3 Max. 40%
Grade C 24.6
Soundness by sodium
sulfate solution C-88
Percent loss in
Coarse agg. 4.2
Fine agg. 2.3 Max. 10%
* MOC = M i n i s t r y of Communications, K i n g d o m of Saudi A r a b i a
[16] w i t h a compaction effort of 75 blows. Mix design data is reported

elsewhere [17]. Mixes at their o p t i m u m asphalt content were then made
and tested for resilient modulus, H v e e m stability, and indirect tensile
strength. The potential of these mixes to rutting was i n v e s t i g a t e d using
the wheel tracking test (LCPC). The other properties, such as air voids,
VMA, unit weight, were also investigated. Duplicate specimens were used
for each mix.
TABLE 3--Aqqreqate Gradations Used for V a r i o u s Filler Contents
Percent Passing
Sieve Size JMF* JMF JMF

m m (#) for 3% for 5.5% for 8%
Filler Filler Filler
19 I00 i00 I00

12.5 83.6 84 84.4
9.5 70.8 71.5 72.2
4.75 (#4) 47.2 48.5 49.8
2.0 (#i0) 28.2 30 31.8
0.425 (#40) 11.2 13.5 15.7
0.180 (#80) 6.6 9 11.4
0.075 (#200) 3.0 5.5 8.0
* Job Mix Formula
H v e e m Stability
The test was performed for each mix at its o p t i m u m asphalt content
according to the ASTM Test Methods for Resistance to d e f o r m a t i o n and
C o h e s i o n of Bituminous Mixtures by Means of Hveem A p p a r a t u s (D 1560).
Hveem stability values were calculated using the following equation
[16]:
22.2
S =
Ph D (i)
+ 0. 2 2 2
Pv-Ph
where;
S = Relative stability
D = D i s p l a c e m e n t on specimen
Pv = Vertical pressure at 5000 ib (22.24 kN), t y p i c a l l y
400 psi (2.76 MPa)
Ph = Transmitted lateral pressure c o r r e s p o n d i n g to Pv of
5000 ib (22.24 kN).
Split Tensile Strenqth
The indirect tensile test is the most widely used test for
determining the tensile properties of highway materials. Specimens
tested for resilient modulus were tested for the indirect t e n s i l e
strength at 25~ The load was applied using a v e r s a t e s t e r m a c h i n e at a
constant rate of deformation of 2 inches (51 mm) per minute. Both
vertical load and d e f o r m a t i o n v a l u e s at failure were r e c o r d e d using a
digital display.The indirect tensile strength was c a l c u l a t e d using the
following equation [18].
2 Pmax (2)
ST -
~ t D
where;
ST = T e n s i l e strength, psi
V e r t i c a l load at failure, ib
t = Specimen's thickness, inches
D = S p e c i m e n ' s d i a m e t e r (4 inches).
Resilient Modulus {MR)
R e s i l i e n t m o d u l u s w a s d e t e r m i n e d for all m i x e s at r o o m t e m p e r a t u r e
(25~ by m e a n s of a d i a m e t r a l m o d u l u s d e v i c e in a c c o r d a n c e w i t h A S T M
M e t h o d of I n d i r e c t T e n s i o n T e s t for R e s i l i e n t M o d u l u s of B i t u m i n o u s
M i x t u r e s (D 4123). S p e c i m e n s w e r e s u b j e c t e d to a 450 Ib d y n a m i c load of
0.i sec d u r a t i o n a n d 30 r e p e t i t i o n s p e r m i n u t e f r e q u e n c y . T h e 450 Ib
p u l s e load is w i t h i n t h e r a n g e of I0 to 50% of t h e t e n s i l e s t r e n g t h as
r e c o m m e n d e d b y A S T M D 4123. A s e a t i n g l o a d of i0 ib w a s a p p l i e d to h o l d
t h e s p e c i m e n in place. By m e a s u r i n g t h e r e s i l i e n t d e f o r m a t i o n M R v a l u e
w a s c a l c u l a t e d as follows:
M~ = P (~ + 0 . 2 7 3 4 ) (3)
t (A~
where;
M R = R e s i l i e n t modulus, p s i
P = D y n a m i c (pulse) load, ib
= P o i s s o n ' s ratio, a s s u m e d to be 0.35
t = T h i c k n e s s of specimen, i n c h e s
AH = Total recoverable horizontal deformation, inches.
Wheel Trackinq Test
The pavement rutting machine, manufactured by the Laboratory

Central Ponts et C h a u s s e e s , France, w a s u s e d to t e s t t h e s e l e c t e d
mixes. T e s t s w e r e c a r r i e d out at t h e M i n i s t r y of C o m m u n i c a t i o n s
L a b o r a t o r y , R i y a d h , S a u d i Arabia. T h e p r o c e d u r e o u t l i n e d in t h e
O p e r a t i n g M a n u a l [19] was f o l l o w e d for p r e p a r a t i o n and t e s t i n g of
a s p h a l t c o n c r e t e slabs. T h e m a t e r i a l p r e p a r a t i o n w a s in g e n e r a l s i m i l a r
to t h a t of t h e M a r s h a l l m e t h o d b u t on a l a r g e r scale. Slabs; 50 cm long,
18 cm wide, a n d i0 cm h i g h w e r e c o m p a c t e d u s i n g a l a b o r a t o r y t i r e
compactor.
B e f o r e testing, t h e slabs w e r e m a i n t a i n e d at a t e m p e r a t u r e of 60~
for 12 hours. A s e n s o r set i n s i d e t h e s l a b w a s u s e d to m o n i t o r t h e
t e m p e r a t u r e w h i c h w a s m a i n t a i n e d at 60~ d u r i n g t h e test.
T h e slabs w e r e s u b j e c t e d to a w h e e l load of 5 kN w i t h i n f l a t i o n
p r e s s u r e of 6 b a r s (0.6 MPa). T h e t i r e p r e s s u r e w a s c h e c k e d f r e q u e n t l y .
T h e w h e e l w a s c o n f i n e d to f o l l o w t h e same t r a c k on t h e slab.
Two slabs w e r e p l a c e d s i m u l t a n e o u s l y on t h e left and o n t h e r i g h t
sides of t h e r u t t i n g m a c h i n e . T w o d a y s a f t e r p r e p a r a t i o n t h e slabs w e r e
s u b j e c t e d to a p r e l o a d i n g of i000 c y c l e s (i load c y c l e = 1 t r a v e l and
r e t u r n of t h e tire). The r u t t i n g t e s t w a s t h e n p e r f o r m e d in t h e
f o l l o w i n g day. T h e rut d e p t h m e a s u r e m e n t s w e r e c a r r i e d out a f t e r i00,
300, i000, 3000, and i 0 , 0 0 0 l o a d i n g cycles. At t h e e n d of e a c h
m e a s u r e m e n t , t i m e w a s a l l o w e d for t h e t e m p e r a t u r e i n d i c a t e d by t h e
s e n s o r i n s i d e t h e slab to r e t u r n to its p r e s c r i b e d value. T h e rut d e p t h
at a given loading cycle was then calculated as the average of the 5

dial gauges measurements.
R E S U L T S AND D I S C U S S I O N
As reported earlier, mixes at their o p t i m u m asphalt content for

various types and contents of filler were tested for indirect tensile
strength, resilient modulus, and wheel tracking test (to m e a s u r e the rut
depth). The results are summarized in Table 4. The table also includes
other properties such as VMA, air voids, Marshall flow, and stability.
Effect of Fillers on A s p h a l t - F i l l e r Mixes
Effect on Softeninq Point--A softening point test was conducted on

selected mixes of filler/asphalt in order to study the changes in
softening point with filler type and content. The results of softening
points are shown in Table 4 and shown g r a p h i c a l l y in Figure 2. The
figure shows that the softening point becomes higher as the amount of
limestone dust increases from 3% to 8%.
It was found that replacing limestone dust with hydrated lime
s i g n i f i c a n t l y raises the softening point, w h e r e a s replacement with
portland cement has little effect. This suggests that the hydrated lime
has a great stiffening effect on asphalt cement w h i c h increases with
hydrated lime contents.
For mixes made with 8% and 5.5% limestone dust, the percent
elevation of softening point, upon replacing three percent limestone
dust by hydrated lime, was about 59 and 53, respectively.
The percent elevation of softening point due to the increase in
115
110 _'/, Fitler Content
3 (LSD)
105
n 5.5 (LSD+HL)
100 Z~ 8 (LSD + HL )
95 X 5.5 (LSO+ PC)
O 8 (LSI)§
90 n
""
c 85
"5
" 80
O~
._c 75 A
C
70
"6
~n o
65
[] o
60 x
55
5O
45 I I I
0 1 2 3
"1, of HL or PC in the Mix
Fig. ~ - - E ~. e. u. ~. .
z of filler type and content
on softening point
the content of limestone dust from 3 to 8% was about 17%. The highest
value of softening point (>IO0~ was obtained for mixes made w i t h 8%
filler composed of 5% limestone dust and 3% hydrated lime.
Effect of Fillers on the Properties of Asphalt Concrete Mixes
Effect on O p t i m u m Asphalt C o n t e n t - - F i g u r e 3 indicates that the

o p t i m u m asphalt contents for mixes containing different fillers vary
over a wide range. Mixes made with 8% limestone dust p a r t i a l l y replaced
with hydrated lime possess higher o p t i m u m asphalt content than those
replaced with portland cement e s p e c i a l l y at high p e r c e n t a g e of
replacement. The difference in optimum asphalt content of
mixes with hydrated lime filler over those with portland cement,
increases with increasing the amount of
TABLE 4--Summary of the Properties of Asphalt Concrete Mixes at O p t i m u m

Asphalt Content
Mix RD(mm) MR ITS Marsh. Flow Unit

Desig- @IO000C (lO~kP,) (kP,) Stab. (mm) Wt.
nation ycles (kN) (g/cc)
3L 4.51 3.302 981.05 14.50 4.39 2.314
IH2L 3.179 977.64 14.85 4.25 2.327
2HIL 2.918 970.92 16.54 3.96 2.322
3H 5.43 2.973 1099.39 16.87 4.03 2.332
IP2L 2.615 88.62 13.28 3.90 2.310
2PIL 2.159 878.09 15.08 4.17 2.322
3P 2.402 930.36 14.63 3.73 2.314
5.5L 3.83 3.491 1005.21 15.92 4.11 2.356
IH4.5L 5.46 2.882 001.7 16.39 3.69 2.364
2H3.5L 2.884 1020.1 16.79 4.12 2.358
3H2.5L 6.48 2.726 982.39 16.00 4.22 2.322
IP4.5L 4.88 3.016 1029.1 13.80 4.28 2.346
2P3.5L 2.616 989.84 16.22 3.96 2.371
3P2.5L 5.28 2.560 988.24 16.83 4.06 2.363
8L 6.25 3.088 1118.5 14.26 5.10 2.385
IH7L 8.05 3.802 1235.3 17.52 4.21 2.369
2H6L 3.101 1094.7 16.26 3.97 2.340
3H5L 11.53 2.759 1037.3 15.78 4.14 2.340
IP7L 6.80 3.254 1159.2 17.74 3.67 2.376
2P6L 2.749 1120.8 17.16 4.12 2.383
3P5L 6.21 2.884 1151.4 19.13 3.90 2.375
*Description of abbreviation is given in a b b r e v i a t i o n list.
**No standards for these numbers are currently available from MOC.
replacements. This is attributed to the higher specific surface area

and asphalt absorption of the hydrated lime particles.
For mixes made with 5.5% limestone dust, there is no big
difference in o p t i m u m asphalt content b e t w e e n hydrated lime and portland
cement, whereas, mixes made with 3% limestone dust replaced with
h y d r a t e d lime generate higher optimum asphalt content than those
replaced with portland cement except in case of two percent hydrated
lime. The highest value of optimum asphalt content (5.15%) was obtained
w i t h 3% hydrated lime, while the lowest value (4.31%) was obtained with
8% limestone dust partially replaced with three percent p o r t l a n d cement.
The increase was about 20%. This is attributed to the effect of such
fillers on the air voids. Mixes containing 3% h y d r a t e d lime p o s s e s s e d
higher air voids than those containing 8% limestone dust replaced with
three percent portland cement.
Effect on Unit W e i o h t - - T h e effect of both filler type and their

contents on the unit weight of compacted m i x e s is shown in Figure 4.
M i x e s m a d e with 3% limestone dust r e p l a c e d p a r t i a l l y by h y d r a t e d lime
showed a t r e n d of increase in unit weight. The o p p o s i t e t r e n d was
o b s e r v e d for mixes p a r t i a l l y replaced by p o r t l a n d cement. Mixes made
w i t h high contents of limestone dust (5.5% and 8%) showed a d e c r e a s e in
unit weight w i t h increasing p e r c e n t a g e replaced by h y d r a t e d lime. These
are a t t r i b u t e d to the increase in air voids r e s u l t i n g from insufficient
compaction. Since the increase in v i s c o s i t y of filler-asphalt mixes
made w i t h hydrated lime is higher than that o b t a i n e d for those made with
p o r t l a n d cement t h e former t e n d to be stiffer and h e n c e r e q u i r e greater
c o m p a c t i o n effort.
It is shown that the replacement of limestone dust w i t h p o r t l a n d
cement had the same effect on unit weight for the three levels of filler
contents. One percent replacement by p o r t l a n d cement results in a
r e d u c t i o n in unit weight value. Two percent r e p l a c e m e n t showed a
TABLE 4--(Contd.)
Mix AV VMA VFA Hveem Opt. Soft.

Desig- % % % Stab. AC. Point
nation % (~
3L 5.42 12.74 59.00 55.00 4.71 53.9

IH2L 4.11 12.71 67.75 60.05 5.06
2HIL 4.70 12.46 61.30 61.43 4.63
3H 3.91 12.61 69.60 60.57 5.15 67.2
IP2L 5.56 13.17 57.38 57.27 4.75
2PIL 4.80 13.04 64.30 56.83 5.14
3P 5.86 13.12 56.40 56.31 4.80
5.5L 4.16 11.20 63.00 62.53 4.38 58.9
IH4.5L 4.00 11.68 66.00 61.35 4.58 62.6
2H3.5L 3.73 11.60 69.30 59.29 4.83
3H2.5L 5.50 13.18 59.30 60.00 4.75 90.5
IP4.5L 4.48 11.84 63.00 59.40 4.56 59.2
2P3.5L 3.11 11.20 71.50 59.37 4.85
3P2.5L 4.00 11.14 66.40 60.31 4.68 57.6
8L 2.38 10.59 78.80 45.01 4.68 63.1
IH7L 3.92 10.97 65.60 61.22 4.35 76.0
2H6L 3.52 12.08 70.50 61.35 4.50
3H5L 4.54 12.48 64.20 54.40 4.95 i00
IP7L 3.76 10.50 65.00 62.63 4.34 66.1
2P6L 3.38 10.31 68.50 61.09 4.43
3P5L 3.04 10.54 71.00 60.57 4.31 64.1
m o d e r a t e increases in unit weight values. Beyond that, the unit weight

decreased. It is shown that mixes made w i t h 3% limestone dust p a r t i a l l y
r e p l a c e d by hydrated lime possess higher unit weight than those
p a r t i a l l y replaced by p o r t l a n d cement. However, the o p p o s i t e is true
for m i x e s with high limestone amounts (8%).
Effect on Air V o i d s - - A s can be seen in Figure 5 the air voids

decrease as the amount of limestone dust increases. The d e c r e a s e in air
voids upon increasing limestone dust content from 3% to 5.5% is
a p p r o x i m a t e l y 30% w h i l e from 3% to 8% is a p p r o x i m a t e l y 128%.
The effect of the replacement process depends p a r t i a l l y on the
total filler content as was the case with most mixes properties. The
highest v a r i a t i o n b e t w e e n h y d r a t e d lime and p o r t l a n d cement is observed
for mixes m a d e w i t h three percent.
In general, for high filler content (5.5 and 8%) h y d r a t e d lime
mixes possess higher air voids than p o r t l a n d cement mixes. This is
p r o b a b l y due to the fact that hydrated lime stiffens m i x e s more than
p o r t l a n d cement.
'1, Fit%er Content

-o-3 -B-S5 -v.-8
5.2
c 5.0
o
u
4.8
o.
in
4.6
E
.--
E
4.4
O
4.2 HL
..... PC
4.0 I I I
0 1 2 3 4
9/ , o f H L o r PC i n t h e Mix
Fig. 3--Effect of filler type and content

on o p t i m u m asphalt content
2.41 I
2 o~0 I- "1, F i l t e r Content
/
2.39L -o- 3 -m-5.5 ~ 8

t
2.38 "~~..~ ......._x....__.
"~,
2-34 ~ --
2.33
2.32
2.31 - ..... "
HL
2.30
...... PC
2,29 I I I
0 1 2 3 4
% of HL or PC in the Mix
Fig. 4 - - E f f e c t of filler type and c o n t e n t

on unit w e i g h t
Indian Institute of Technology Kharagpur pursuant to License Agreement. No further reproductions authorize
Effect on Voids in Mineral A g g r e q a t e s - - F i g u r e 6 d e m o n s t r a t e s the

effect of filler type and their contents on voids in mineral aggregates
(VMA). The V M A decrease as the amount of limestone dust increases. The
V M A values of mixes made with 3% filler are slightly affected by
hydrated lime or portland cement as a partial replacement for limestone
dust. On the other hand, for those with higher filler content (5.5% and
8%), the VMA values are highly affected due to partial replacement of
limestone dust by hydrated lime but not by p o r t l a n d cement.
The highest variation b e t w e e n hydrated lime and p o r t l a n d cement
fillers is observed at three percent partial replacement of limestone
dust. At this level, and for 5.5% and 8% filler contents, the increase
in V M A value due to using hydrated lime instead of p o r t l a n d cement is
about the same for both filler contents and is equal to about 18%. It is
clear that all the V M A values are lower than the limits set by MOC
specifications [15]. Lower values of VMA result in less spaces to
accommodate the required asphalt to produce good coating and durable
mix.
Effect on Marshall S t a b i l i t y - - F i g u r e 7 shows the results of

Marshall stability as a function of varying the filler content and type.
It is observed that for mixes made with limestone dust, the Marshall
stability increases as the filler content increases from 3% to 5.5%, and
then decreases as the filler content increases. This is due to the fact
that voids are too high at 3% and the aggregates are finer at 8%, both
tend to reduce stability. It is shown that for mixes made with 3%
filler, the stability increases with the increase of percentage replaced
with hydrated lime. On the other hand, for hydrated lime mixes with high
filler contents (5.5% and 8%), the stability increases with the increase
in the amount of replacement up to a m a x i m u m value and then decreases.
This can be easily explained by the fact that the hydrated llme is finer
than limestone dust, hence it stiffens and, consequently, increases
Marshall stability as its content increases. After a certain level,
mixes become stiffer and require greater compaction effort.
For a given percentage of p o r t l a n d cement there is an increase in
Marshall stability as filler content increases from 3 to 8%.
Effect on Marshall Flow--Figure 8 shows that for mixes made with

hydrated lime as a partial replacement for limestone dust there is an
o p t i m u m value of filler content w h i c h yields the m i n i m u m Marshall flow
values. A d d i t i o n of hydrated lime higher than this o p t i m u m starts
increasing Marshall flow value. This is the result of increasing air
voids and then increasing the asphalt cement required. The general trend
is such that there is a decrease in flow as the contents of both of
hydrated lime and portland cement increase and p a r t i c u l a r l y for 5.5 and
8% filler contents.
Effect on H v e e m Stability--The effect of filler type and its

contents on H v e e m stability is depicted in Figure 9. H v e e m stability
values of mixes made with 5.5% limestone dust were the highest followed
by mixes made with 3% and then those made with 8%. This may be explained
by the following: increasing filler content from 3% to 5.5% fills the
voids among aggregates particles thus p r o d u c i n g dense mixes, hence
increasing Hveem stability, whereas increasing filler content beyond
5.5% reduces the contact among coarse aggregate particles, hence
reducing the Hveem stability.
It is clear that varying the amount of hydrated lime or portland
cement as a partial replacement for limestone dust has little effect on
H v e e m stability values. All mixes studied satisfied the m i n i m u m Hveem
stability criteria of 40 set by MOC specifications [I_55].
The highest increase in Hveem stability was o b t a i n e d for mixes
made w i t h 8% filler. This can be explained by the fact that mixes made
with 8% limestone dust possess relatively low internal friction, hence
low H v e e m stability, but by partially replacing limestone dust by either
6.5 | Filler C o . t e n t
-e-3 -e-~5 -x-8
6.0
2
/
5.5
" 5.o
"O
"5 4.5
._~
< z,.O
3.S
/," \\'x.j_,
3.0
/"
// ~ "'--x
2.S HL
..... PC
2.0 I I I
0 1 2 3 4
' / , o f H L o r P C inthe Mix
Fig. 5--Effect of filler type and
content on air voids
14.0 % Filler Content

-e-3 -Q-5.5 -~-8
13.5
13.0
12.5
12.0
11.5
11.0
105
...... PC
10.0 I I I
0 I 2 3
91, of HL or PC in the Mix

content on V.M.A.
2 1 - -
% Filler Content
20 -e-3 -B-~S -=-8
19 X
/
b
"~ 17
u3
_.-16
O
1&
~',NI/
13 HL
..... PC
12 I I I
0 ! 2 3
9/ , o f HLor PC in!he Mix
Fig. 7 - - E f f e c t of f i l l e r t y p e a n d
c o n t e n t on M a r s h a l l s t a b i l i t y
55"~ "1. Filter Content

9 ~ -e-3 --e-5.5 --x-8
4.8~
.\
a.8 \\ , / ",--
-,~ ,~
3.6 HL
..... PC
3.& I I I
0 1 2 3
Fig. 8 - - E f f e c t of f i l l e r type and

content on flow
p o r t l a n d cement or hydrated lime, the mixes become stiffer, hence the

increase in H v e e m stability.
Effect on Resilient M o d u l u s - - C h a n g e s of resilient m o d u l u s with

filler type and amount is d e m o n s t r a t e d in Figure I0. It can be seen that
replacing small amounts of limestone dust with hydrated lime or portland
cement generally reduces resilient modulus of most mixes. However the
reduction is more in the case of portland cement mixes. The most
obvious exception is for 8% filler mix with 1% hydrated lime w h e r e there
is an increase in M R values up to 1% replacement then a decrease after
that. The decrease in resilient modulus with the increase in the amount
of replaced limestone dust by either hydrated lime or portland cement
could be attributed to the effect of such fillers on stiffness.
Increasing hydrated lime or portland cement may have extended the
asphalt and hence caused reduction in resilient modulus.
Effect on Indirect Tensile Strenqth--The effect of filler type and

content on split tensile strength is d e m o n s t r a t e d in Figure ii. It can
be seen that filler type and filler replacement have little effect on
split tensile strength. It is clear that the highest value of split
t e n s i l e strength was obtained for mixes made with 8% filler content
whereas the highest values of Marshall and Hveem stabilities were
obtained for mixes made with 5.5% filler content. This may be due to
effect of both the air voids and the asphalt content on the tensile
strength of asphalt concrete mixes. This is in agreement with Hadley et.
al. [20].
Permanent Deformation Test--The wheel tracking test was conducted

on asphalt concrete mixes made with different filler types and contents
to study permanent deformation (rutting). The detailed results of this
test are reported elsewhere [17].
Figure 12 shows the accumulated rut depth at I0,000 loading cycles
for various mixes selected. It can be seen that the rut depth is
decreased from 4.51 m m to 3.83 mm when the limestone dust increased from
3% to 5.5% (i.e., 15% reduction), whereas the rut depth increased to
6.25 m m when the limestone dust increased to 8% (Figure 13). The
increase was about 63%. There seems to be an o p t i m u m filler content for
the least rut depth. This behavior could be explained by the following:
increasing filler content from 3% to 5.5% reduces air voids, hence
d e c r e a s i n g rut depth. Whereas, increasing the filler content beyond 5.5%
increases fines in the mix which decreases the internal friction, hence
increasing the rut depth.
Mixes made with hydrated lime partially replacing limestone dust
possess higher rut depth than those replaced with p o r t l a n d cement. This
is attributed to the fact that mixes containing hydrated lime have
higher air voids than those containing portland cement, hence they are
more susceptible to rutting. This may indicate that such mixes need more
compaction.
So from this investigation, it is clear that the compactness has a
large effect on mix behavior. This agrees with the studies done by Barbe
et al. [21] and A i - D h a l a a n et al. [22] in w h i c h they found that
increased compaction has a positive effect on reducing rutting of
asphalt concrete mixes. Also, Hughes et al. [23] concluded that the
rutting during the first 12 months of pavement service life appears to
be influenced by the density obtained during c o n s t r u c t i o n (the lower
densities associated with greater increase in rutting).
The asphalt content seems to be another factor c o n t r i b u t i n g to
rutting. This is in agreement with the study done by Bolk et al. [24],
in w h i c h they found that the b i t u m e n content has a s u b s t a n t i a l l y greater
effect than the filler content on the deformation resistance. Rutting
decreases with decreasing bitumen content.
An example of the effect of number of load repetitions on rutting
of the asphalt concrete mixes is shown graphically in Figure 14. It can
be seen that rutting increases with increasing number of load
6 7 l-
/ AL-SUHAIBANI ET AL. ON ASPHALT CONCRETE MIXES
% Filler Content
121
s
61 ~--
-r
s2
49
46
43 HL
..... PC
40 I I
0 1 2 3
Fig. 9 - - E f f e c t of f i l l e r type and
c o n t e n t on H v e e m stability
4.2
"h Filter Content
4.0
-o-3 -a-5.5 -~8
3.8 X
3.6
3.4
0
o. .X.
3.2
r
3.0
IZ
2.8
2,6 '~ "E}-. . . .
2.4 %%% ~4
2.2
2.0 HL
..... PC
1. I I I
0 1 2 3 4
"1, of HL or PC in the Mix
Fig. 1 0 - - E f f e c t of f i l l e r type and

c o n t e n t on M R
,=,
1350
*l. Fitter Content
1300 -e-3 -e-~5 -,,-8
1250
X
1200
1150
1100'
1050
10001
950
900
850 HL
..... PC
80O ! I I
0 1 2 3
*/,of HLor PC inthe Mix
Fig. l l - - E f f e c t of f i l l e r type and
c o n t e n t on ITS
13 % FiUer Content
-e-3 -a-5.5 --~-8
12
X
11
10
8
7
4 HL
...... PC
3 I I I
0 I 2 3
content on Rut Depth
9.0
8.5
8.0
7.5
,,, 7.0
E
v
E 6.5
t-"
~. 6.0
Q 5.5
"" 5.0
4.5
4.0
3.5
3.0 1 I I I 1 I I I
1D 2.0 3.0 4.0 5.0 6D 70 8.0 9.0 10.0

9/. of L i m e s t o ~ Dust
Fig. 13--Effect of limestone dust
content on Rut Depth
12
11 9 3L
10 n 3H
9
--, 8
E
6
,-.--s
o 4
3
2
1
0 I, | I I |
0 1 2 3 4 5 6 ? 8 9 10 11
N u m b e r of Load Cycles(Thousands)
Fig. 14--Effect of loading cycles on Rut Depth
repetitions. The rate of increase is rather high up to a p p r o x i m a t e l y

1,000 loading cycles and then decreases and the r e l a t i o n s h i p between the
rut depth and the number of load cycles become r e l a t i v e l y linear.
R u t t i n g is highly affected by the amount of hydrated lime but not
by the portland cement. It is also shown that mixes made with limestone
dust exhibit lower rut depth than those containing either p o r t l a n d
cement or hydrated lime.
The rut depth increases as the percentage of limestone dust
replaced by hydrated lime or portland cement increases. H i g h e s t values
were found for mixes made with 8% limestone dust replaced by three
percent hydrated lime, while the lowest values were found for those made
with 5.5% limestone dust. The reason that mixes made with partial
replaced filler showed higher rutting depths could be a t t r i b u t e d to
insufficient compaction. Upon replacing part of the limestone dust with
hydrated lime or portland cement in asphalt-filler binders it was found
that the viscosity of such binders increases and, consequently, the
stiffness of asphalt concrete mixes made with these binders increases
too. This necessitates increasing compaction efforts to o v e r c o m e
increasing stiffness. However, since the compaction effort was
constant throughout this study, the stiffer mixes were not as well
compacted.
To be meaningful, the rut depth for each mix is also reported as
percent of the slab thickness. The rut depth as % of slab thickness can
be best described by the number of loading cycles (see Figure 15). From
linear regression analysis of logarithms of the rut depth as % of slab
thickness versus number of load cycles, an equation of the following
form can be obtained [19]:
RD~ = A Ns (4)
where;
RD~ = Rut depth as a p e r c e n t a g e of slab
thickness
A = Rut depth associated with one load cycle
N = Number of load cycles
B = The slope of the best fitting straight
line when R D s ~ and N are p l o t t e d on log-
log plot.
The values of A and B were calculated for the various mixes and
are given in Table 5. The coefficient of determination, R 2 is also shown
in this table. Example of the best fitting straight lines is shown in
Figure 15.
The slope of the deformation line (B value) is of great importance
in comparing rutting susceptibility of different mixes because the slope
of the line indicates the progress of rutting over a period of time
(load repetitions). The higher the B value, the more susceptible the mix
to rutting is. The A value (intercept) indicates the initial deformation
(associated with one load cycle).
ANALYTICAL MODEL FOR PREDICTION OF PERMANENT DEFORMATION
An attempt was made to correlate the permanent d e f o r m a t i o n with

other properties of asphalt concrete mixes such as Marshall stability,
flow, unit weight, air voids, voids in mineral aggregates, voids filled
with asphalt, Hveem stability, asphalt content, split tensile strength,
resilient modulus, softening point of filler-asphalt mixes, filler type
and amounts. The purpose was to develop an analytical model for the
p r e d i c t i o n of permanent deformation based on some or all of these
properties.
A stepwise regression analysis, using the Statistical Analysis
System (SAS), p r o g r a m was conducted to develop a r e l a t i o n s h i p b e t w e e n
the rut depth as % of slab thickness and other mix properties. The
results of the analysis are shown in Table 6. The results show that the
softening point (SP), asphalt content (AC), and voids in mineral

aggregates (VMA) are the most significant independent variableB, of all
the independent v a r i a b l e s investigated, w h i c h give t h e h i g h e s t
c o r r e l a t i o n w i t h RDs~. The general form of the model is as follows:
RD~ = 0.1398 SP + 1.9143 AC - 1.0462 VMA (5)

with R ~ = 98.6%
where;
RD~ = Rut depth as p e r c e n t a g e of slab t h i c k n e s s

SP = Softening point of f i l l e r - a s p h a l t mixes, ~
AC = Asphalt Cement content, % by total weight
of mix
VMA = Voids in mineral aggregates, %.
TABLE 5--Ruttinq Parameters, RD~ = A * NB
Mix Regression Data

Designation
A B R2 T-Value
3L 0.4330 0.255 0.9999 246.95

3H 0.4549 0.269 0.9997 135.37
5.5L 0.3273 0.269 0.9996 114.84
IH4.5L 0.2919 0.323 0.9969 40.35
3H2.5L 0.1999 0.381 0.9986 59.81
IP4.5L 0.3192 0.298 0.9995 95.32
3P2.5L 0.2373 0.340 0.9987 61.76
8L 0.3363 0.323 0.9969 39.82
IH7L 0.2196 0.397 0.9953 32.61
3H5L 0.2128 0.439 0.9960 35.48
IP7L 0.4297 0.303 0.9988 65.46
3P5L 0.4713 0.281 0.9997 134.17
TABLE 6--AnalTsis of Variance and R e q r e s s i o n Model for_RD~
Source DF Sum of Mean F Value Prob>F

Squares Square
Model 3 502.249 167.416 205.019 0.0001
Error 9 7.349 0.817
U Total 12 509.599
Root MSE 0.9044 R-Square 0.9856

Dep Mean 6.226
C.V. 14.515 ADJ R-SQ 0.9808
Variable DF Parameter Standard T for Ho Prob >T

Estimate Error Para-
meter=0
SP 1 0.1398 0.0214 6.519 0.0001
AC 1 1.9143 1.0522 1.819 0.1022
VMA 1 -1.0462 0.4312 -2.427 0.0382
The model predicts an increase in rut depth as the softening point

of the binder increases, as the asphalt content increases, and as the
voids in mineral aggregates decrease. From the developed model, it is
clear that the softening point of the binder has the most significant
effect on rut depth.
The reason that the relationship between softening point of the
binder and rutting is unexpectedly proportional is that rutting in
asphalt concrete mixes depends not only on the softening point of the
binder but also on several other factors which interact with the
softening point. The softening point of the binder does increase upon
addition of different fillers with different contents. However, by
increasing the softening point of the binder the stiffness of the
asphalt concrete mix is increased. The increase in mix stiffness
requires higher compaction effort to achieve the same air voids as the
other mixes. In this study, however, the compaction effort was
constant, hence air voids are higher for mixes containing high softening
point binders. It is those mixes with high air voids which exhibited
higher rut depth.
For each slab, the unit weight was calculated before and after
being tested in the wheel tracking test. Figure 16 shows the
differences in unit weight values between mixes made with hydrated lime
and portland cement. The figure also shows the differences after
performing the rutting test. It is clear that mixes made with hydrated
lime show lower unit weight values than those made with portland cement.
It is shown that the unit weight decreases with the increase of percent
of replacement with hydrated lime. It can be seen from Figure 17 that,
in general, mixes compacted by a laboratory tire compactor produce lower
unit weight values than those compacted by Marshall compactor. This was
attributed to the different compaction effort used.
CONCLUSIONS
Based on the findings of this study, the following conclusions are

drawn:
i. Replacing limestone dust with hydrated lime has caused a
considerable increase in softening point of the mastic, whereas,
replacing limestone dust with portland cement has little effect on it.
2. For high filler contents (5.5% and 8%), mixes containing
hydrated lime replacing limestone dust possess higher optimum asphalt
contents, higher air voids, and lower unit weights than those containing
portland cement. The differences in optimum asphalt content, air voids,
and unit weight increase with the filler content.
3. VMA values of mixes are not sensitive to portland cement
replacement.
4. Hveem stability is not affected by the type of filler used in
the mixes.
5. Resilient modulus values of mixes were found to be decreasing
with the increase in the amount of replacement of limestone dust by
either hydrated lime or portland cement.
6. Mixes with high contents of limestone dust (5.5% and 8%) showed
a decrease in unit weight with increasing percent replaced by hydrated
lime.
7. It was found that mixes compacted by laboratory tire compactor
produce lower unit weight than those compacted by Marshall compactor.
8. Replacing part of limestone dust by hydrated lime or portland
cement aggravates the resistance of mixes to rutting. The rut depth
increases as the percentage replaced increases. Highest values were
found for mixes made with 8% limestone dust partially replaced by three
percent hydrated lime, while the lowest values were found for those made
with 5.5% limestone dust.
9. Regression analysis showed that rutting in asphalt concrete
mixes is highly dependent on softening point of binder (asphalt +
filler) which depends on filler type and concentration, VMA, and asphalt
100 '
9 3L RD('I.)=0.4338.N-0.2663 R-SQ-0.9996
Z~ 3H RD('I.)=0.4549.N'0.2688 R 50"0.9966
,_"Z
JE
.,0
l:l
i/I
"6
10
,,C:
la
S
O
3
I:I:
1 i I ,,,,,,I I I I IIILI[ I i I I Rj''I L , , ,,,,,I J | i , , J,i
10 100 1000 10000 1OOOOO

Number of Load Cycles
Fig. 15--Effect of load c y c l e s on Rut D e p t h
2.36 -
2.3&
J~Gmb before ~Gmb after
,_. 2.32
2.30
9~ 2 . 2 8 -
.~_c~2 . 2 6 -
~2. N
.-_ 2.22
C
m 2.20
2.18 -
2.16-
2.14
2.12
, . . . . . .
2.10 i
3L 3H 6.6L 1H4.6L3H2.6L~4.6L3P2.6L 8L 1H7L 3H6L 1PTL 3 6L
Mix Designation
Fig. 16--Effect of Rutting test on unit w e i g h t
content (AC). However, other properties were not highly correlated with
rut depth.
2.52
2 .t,9
2.&6 Ji m Grab (tyre comp.) ~ Grab(Marshall corer ) J
U 2.&3
O~ 2.&0
2.2~8
.E 2.37
i l[ ii,
2.34
2.31
2.25
2.22
2.19
2.16
2.13
2.10
3L 3H 6.6L 1H&.6L3H2.6L1P&.6L3P2,6L 8L 1H7L 3H6L 1P7L 3P6L
Mix Designation
Fig. 17--Difference between unit weight of compacted

and Marshall compacted speciraens
REFERENCES
[!] Balghunaim, F., A1-Dhubaib, I., Khan, S., Fatani, M., AI-
Abdulwahhab H., and Babshait, A. "Pavement Rutting in the Kingdom
of Saudi Arabia: A diagnostic Approach to the Problem",
Proceedinqs of 3rd IRF Middle East Reqional Meetinq, Vol. 6,
Riyadh, February 1988, pp. 210-232.
[21 Abdulshafi, A., "Rutting-Review of Existing Models and Some

Application to Saudi Arabia", Proceedinqs of 3rd IRF Middle East
Reqional Meetinq, Vol. 6, Riyadh, February 1988, pp. 244-256.
[3] Rao, S.K., and Sen, B.R., "Evaluation of Mineral Fillers for
Asphalt Paving Mixtures", Hiqhway Research Record 468, HRB, 1973,
PP. 100-108.
C41 Kalcheff, I.V., and Tunnicliff, D.G., "Effects of Crushed Stone

Aggregate Size and Shape on Properties of Asphalt Concrete",
Proceedinqs of the Association of Asphalt Pavinq Technoloqists,
Vol. 51, 1982, pp. 453-474.
[_s] Yao, Z., and Monismith, C.L., "Behavior of Asphalt Mixtures with
Carbon Black Reinforcement, Proceedinqs of the Association of
Asphalt Pavinq Technoloqists, Vol. 55, February 1986, p. 582.
[_6) Puzinauskas, V.P., "Filler in Asphalt Mixtures", Asphalt

Institute, RR 69-2, 1969, p. 14.
[7] Eick, J.H., and Shook, J.F., "Effect of Baghouee Fines on Asphalt-
Fines and Asphalt-Aggregate Mixtures", Asphalt Institute, RR 78-3,
1978, p. 14.
[8_] Kallas, B.F., and Puzinauskas, V.P., "A Study of Mineral Fillers
in Asphalt Paving Mixtures", Proceedinqs of the Association of
Asphalt Pavinq Technoloqists, Vol. 30, 1961, pp. 493-525.
[9_] Dukatz, E.L., and Anderson, D.A., "The Effect of Various Fillers
on the Mechanical Behavior of Asphalt and Asphaltic Concrete",
Proceedinqs of the Association of Asphalt Pavinq Technoloq~@t~,
Vol. 49, 1980, pp. 530-548.
[10] Geitz, R.H., and Lamb, D.R., "Effect of Filler Composition on

Binder Viscosity and Mix Stability", Hiqhwa 7 Research Record 256,
HRB, Washington, D.C., 1968, pp. 1-13.
[11] Tunnicliff, D.G., "Binding Effects of Mineral Filler", Proceedinqs
of the Association of Asphalt Pavinq Technoloqist~, Vol. 36, 1967,
pp. 149-150.
[12] Button, J.W., Little, D.N., Gallaway, B.M. and Epps, J.A.
"Influence of Asphalt Temperature Susceptibility on Pavement
Construction and Performance", NCHRP Report No. 252, Washington,
D.C., December 1983, p. 59.
[13] Anderson, D.A., and Tarris, J.p., "Adding Dust Collector Fines to
Asphalt Paving Mixtures", NCHRP Report NO. 252, Washington, D.C.,
December 1982, pp. 8-9.
[14] Nievelt, G., and Thamfald, H., "Evaluation of the Resistance to

Deformation of Different Road Structures and Asphalt Mixtures
Determined in the Pavement-Rutting Tester", Proceedinqs of the
Association of Asphalt Pavinq Technoloqists, Vol. 57, 1988, pp.
320-328.
[Ls] Ministry of Communications (MOC), "General Specifications for

Roads and Bridges Construction: Revised 1987", Circular No. 2403,
Riyadh, Saudi Arabia, 1987.
[16] The Asphalt Institute, "Mix Design Methods for Asphalt Concrete
and Other Hot-Mix Types", Manual Series No. 2 (MS-2), May 1984.
[17] Ai-Fozan, F. "A Study of The Effect of Fillers on The Mechanical
Properties of Asphalt Concrete Mixes", Master Thesis, Civil
Engineering Dept., College of Eng., King Saud University, July
1991.
[18] Kennedy, T.W., and Hudson, W.R., "Application of the Indirect
Tensile Test to Stabilized Materials", Hiqhway Research Record
235, HRB, Washington, D.C., 1968, pp.36-48.
[19] Operating Manual, "Pavement Rutting Tester (1/89)", Map-23, rue de
la Sinne -B.P. 1492 - 68072 Mulhouse Cedex - France.
[20] Hadley, W.O., Hudson, W.R. and Kennedy, T.W., "An Evaluation of
Factors Affecting the Tensile Properties of Asphalt-Treated
Materials", Research Report No. 98-2, Center for Hiqhwa 7 Research,
The University of Texas at Austin, March 1969.
[21] Barbe, B., Caroff, G., Maia, A. and Hiernaux, R., "Permanent
Deformation in Bituminous Mixes Effect of C o n s i s t e n c y of Bitumen,
Type of Aggregate, and Mix Composition", Proceedinqs of the
A s s o c i a t i o n of Asphalt Pavinq Technolouists, Vol. 57, 1988, pp.
197-212.
[22] AI-Dhalaan, M., Balghunaim, F, Ai-Dhubaib, I, and Noureldin, A.S.,

"Field Trials with Polymer Modified Asphalts in Saudi Arabia",
ASTM, Philadelphia, 1991.
[23] Hughes, C.8., and Maupin Jr., G.W., "Experimental Bituminous Mixes
to Minimize Pavement Rutting", Proceedinus of the A s s o c i a t i o n of
Asphalt Pavinq Technoloqists, Vol. 56, 1987, p. 22.
[24] Bolk, N.J.N.A., Van der Hifide, J.P.J. and Van Zantvliet, M.C.,
"Basic Research into the Effect of Filler on the M e c h a n i c a l
Properties of Dense Asphalt Concrete", Proceedinqs of the
A s s o c i a t i o n of Asphalt Pavinq Technologists, Vol. 51, 1982, pp.
398-448.
ABBREVIATION LIST OF SYMBOLS
HL H y d r a t e d lime
PC Portland cement
3L 3% Limestone dust
IH2L 1% Hydrated lime + 2% Limestone dust
2HIL 2% Hydrated lime + 1% Limestone dust
3H 3% Hydrated lime
IP2L 1% Portland cement + 2% Limestone dust
2PIL 2% Portland cement + 1% Limestone dust
3P 3% Portland cement
5.5L 5.5% Limestone dust
IH4.5L 1% Hydrated lime + 4.5% Limestone dust
2H3.5L 2% Hydrated lime + 3.5% L i m e s t o n e dust
3H2.5L 3% Hydrated lime + 2.5% Limestone dust
IP4.5L 1% Portland cement + 4.5% Limestone dust
2P3.5L 2% Portland cement + 3.5% Limestone dust
3P2.5L 3% Portland cement + 2.5% Limestone dust
8L 8% Limestone dust
IH7L 1% Hydrated lime + 7% Limestone dust
2H6L 2% Hydrated lime + 5% Limestone dust
3H5L 3% Hydrated lime + 5% Limestone dust
IP7L 1% Portland cement + 7% Limestone dust
2P6L 2% Portland cement + 6% Limestone dust
3P5L 3% Portland cement + 5% Limestone dust
D. A. A n d e r s o n I, H. U. Bahia 2, and Raj Dongre ~
R H E O L O G I C A L PROPERTIES OF M I N E R A L F I L L E R - A S P H A L T M A S T I C S AND ITS

IMPORTANCE TO PAVEMENT PERFORMANCE
REFERENCE: Anderson, D. A., Bahia, H. U., and Dongre, Raj, "Rheological

Properties of M i n e r a l F i l l e r - A s p h a l t M a s t i c s and Their R e l a t i o n s h i p tD
Pavement Performance," Effects of A q q r e u a t e s and Mineral Fillers on
A s p h a l t M i x t u r e Performance: A S T M STP 1147, Richard. C. Meininger,
editor, A m e r i c a n Society for T e s t i n g and Materials, Philadelphia, 1992.
ABSTRACT: The rheological and failure properties of eight asphalt-

filler mastics were c h a r a c t e r i z e d using n e w testing techniques b e i n g
d e v e l o p e d w i t h i n the SHRP p r o g r a m for testing asphalt binders. The
m a s t i c s were p r e p a r e d using four SHRP core asphalts and two fillers
(calcite and quartz). The rheological m e a s u r e m e n t s were used to
construct rheological master curves and temperature shift functions.
The failure properties were m e a s u r e d at different t e m p e r a t u r e s and
strain rates and were shifted using t i m e - t e m p e r a t u r e s u p e r p o s i t i o n to
construct stress and strain-to-failure master curves. A subset of the
mastics was also tested after oxidative aging with the pressure a g i n g
vessel (PAV) and after isothermal aging at low temperatures. The
p r o p e r t i e s of the mastics were compared with the asphalt binder
p r o p e r t i e s to describe the changes resulting from adding the fillers.
The fillers were found to change the shape of the rheological m a s t e r
curves and to s i g n i f i c a n t l y increase the failure stress at all
combinations of temperatures and loading times. The changes in
rheological properties were o b s e r v e d only in the time d e p e n d e n c y while
the temperature shift functions showed only slight changes at the
highest temperatures. The relative stiffening effects were o b s e r v e d to
be asphalt-specific, especially at the lower frequencies or higher
temperatures. Oxidative aging was shown to change the rheological type
of the mastics, to be asphalt-specific, and to be independent of filler
type. Physical h a r d e n i n g was o b s e r v e d to result in shifting the m a s t e r
rheological curves to longer loading times without affecting the shape
of the m a s t e r curves. The a n t i c i p a t e d influence of m a s t i c p r o p e r t i e s on
the main distress types of asphalt p a v e m e n t s was considered. The
influence of fillers on low-temperature cracking was h y p o t h e s i z e d to be
independent of filler type. Therefore, it is e x p e c t e d that the asphalt
b i n d e r will control the cracking mechanism. B a s e d on the energy
concept, the addition of the filler is expected to improve resistance to
s t r e s s - c o n t r o l l e d fatigue while for s t r a i n - c o n t r o l l e d fatigue the
improvement, if any, is expected to be minimal. Fillers are e x p e c t e d to
s i g n i f i c a n t l y improve rutting resistance as a result of increasing
moduli (viscous component). Effect on rutting is e x p e c t e d to be
asphalt-filler-specific.
IProfessor, Civi- Engineering Department, P e n n s y l v a n i a State

University, U n i v e r s i t y Park, Pa - 16802.
2Research Associate, The P e n n s y l v a n i a T r a n s p o r t a t i o n Institute,

P e n n s y l v a n i a State University, U n i v e r s i t y Park, Pa - 16802.
3Research Assistant, The P e n n s y l v a n i a T r a n s p o r t a t i o n Institute,

P e n n s y l v a n i a State University, U n i v e r s i t y Park, Pa - 16802.
131
Copyright by
Copyright 9 ASTM
1992 Int'l
(all rights
by ASTM
International
07:29:10 EST 2011
KEYWORDS: A s p h a l t mastics, theological properties, complex modulus,

phase angle, b e n d i n g beam rheometer, failure properties, direct tension
test, p a v e m e n t performance, thermal cracking, fatigue cracking, rutting,
oxidative aging, low-temperature physical hardening.
INTRODUCTION
The mineral filler fraction is one of the important components of

h o t - m i x asphalt (HMA), h o w e v e r its importance is often overlooked. The
mineral filler fraction is commonly c o n s i d e r e d part of the aggregate
system, and, for m a n y pavement engineers, this fraction is simply an
inert material whose m a i n role is that of filling the voids b e t w e e n
coarser aggregate particles. However, b a s e d upon a m i c r o s c o p i c
c o n s i d e r a t i o n of HMA, it is apparent that a m a j o r part of the filler
fraction is e m b e d d e d in the asphalt cement in such a way that the H M A
b i n d e r is not asphalt cement but i n s t e a d a m a s t i c consisting of m i n e r a l
fines and asphalt cement. This hypothesis, which has been long held b y
the authors [!], is r e i n f o r c e d by the work being conducted at the Danish
National Road L a b o r a t o r y as part of a Strategic H i g h w a y R e s e a r c h Program
(SHRP) Idea Project, SHRP AIIR-13, M i c r o s c o p i c A n a l y s i s of A s p h a l t
A g g r e g a t e Mixtures R e l a t e d to Performance [~]. In the A I I R study,
optical t r a n s m i s s i o n m i c r o s c o p y is being u s e d to p h o t o g r a p h thin
sections of HMA. B y impregnating the H M A with a f l u o r e s c i n g polymer,
voids, cracks, and fissures within the H M A can be c l e a r l y identified.
From t h i n - s e c t i o n p h o t o g r a p h s several o b s e r v a t i o n s can be made:
9 In terms of p h y s i c a l - c h e m i c a l interactions b e t w e e n the

asphalt and the mineral surface, the p r o p e r t i e s of the
fine fraction should dominate because the fines are
imbedded in the asphalt and, therefore, have the
closest p r o x i m i t y to the asphalt.
9 The m a j o r i t y of the surface area g e n e r a t e d b y the

aggregate is w i t h i n the fines. As shown later, the
specific surface area of the fines m a y be as large as
I m2/g, while the specific surface area of the plus
No. 200 fraction m a y only be in the order of a
fraction of a m2/g. Thus, the fines should be the
m a j o r contributor to any p h y s i c a l - c h e m i c a l interaction
b e t w e e n mineral aggregate surface and asphalt cement.
9 In terms of a c c e l e r a t e d aging, aging in thin films is

not representative of the HIW.A because the asphalt
cement does not exist as a thin film in the HMA. The
mean distance from a single asphalt m o l e c u l e to
ambient air (oxygen) may be in the order of several
thousand ~m and not the i0 to 20~m often a s s u m e d b y
m a n y p a v e m e n t technologists [~].
9 The above hypotheses do not imply that the coarse

aggregate fraction does not p l a y a role in d e t e r m i n i r g
the mechanical p r o p e r t i e s or the aging c h a r a c t e r i s t i c s
of HMA. The coarse aggregate fraction can be e x p e c t e d
to dominate the a b s o r p t i o n c h a r a c t e r i s t i c s of the
aggregate and, to that extent, also p l a y a role in the
aging characteristics of HMA. It is well a c c e p t e d
ANDERSON ET AL. ON MINERAL FILLER-ASPHALT MASTICS 133
that both the coarse and fine fraction should also

affect the moisture sensitivity of H M A [!] 9
In order to u n d e r s t a n d the relation b e t w e e n asphalt binder

p r o p e r t i e s and H M A properties, it is n e c e s s a r y to define the p r o p e r t i e s
of the mineral filler-asphalt m a s t i c that serves as the binder for the
HMA. The importance of the m a s t i c p r o p e r t i e s in d e f i n i n g the H M A
p e r f o r m a n c e is r e f l e c t e d in the asphalt literature by the large n u m b e r
of studies c a r r i e d out to define these p r o p e r t i e s and to identify the
factors that control them [5-i_~0] . These studies indicate that the
physical p r o p e r t i e s of fillers, such as gradation, surface area, and
shape, as well as compositional factors such as surface mineralogy, can
be important variables. The importance of p h y s i c a l - c h e m i c a l interaction
b e t w e e n the fillers and asphalts has been r e c o g n i z e d for a long time and
the d e p e n d e n c y of the mechanical p r o p e r t i e s of the m a s t i c s on the nature
of this interaction has been shown in a n u m b e r of studies [6,11,12] .
The m a j o r i t y of the r e p o r t e d studies of mineral f i l l e r - a s p h a l t

mastics have relied on high temperature routine type m e a s u r e m e n t s such
as penetration, capillary viscosity, or softening point temperature to
describe flow p r o p e r t i e s and to explain the interaction b e t w e e n asphalts
and mineral filler [13,14]. With the new testing techniques b e i n g
d e v e l o p e d within the SHRP p r o g r a m for testing asphalt binders, a n e w
look at the r h e o l o g y and failure p r o p e r t i e s of asphalt m a s t i c s is now
possible. The p u r p o s e of this p a p e r is to report on the rheological and
failure p r o p e r t i e s of a selected group of asphalt binders when m i x e d
with two mineral fillers. The p a p e r also addresses the a n t i c i p a t e d
changes in H M A p e r f o r m a n c e that m a y be a t t r i b u t e d to the mineral filler
p o r t i o n of H M A and its interaction with the asphalt cement.
MATERIALS A N D T E S T I N G P R O G R A M
Four asphalts and two mineral fillers were used in this study.
Table I gives the SHRP identification code for the asphalts, routine
c o n s i s t e n c y measurements, and the chemical p r o p e r t i e s of these asphalts.
The mineral fillers used were calcite and quartz, selected to represent
the range of mineral aggregates b e t w e e n acidic and b a s i c commonly used
as fillers in asphalt concrete p a v i n g mixtures. Both the quartz and the
calcite are c o m m e r c i a l l y p r e p a r e d b y crushing q u a r r i e d stone. Neither
is used as a commercial H M A filler--the quartz is u s e d in the
m a n u f a c t u r e of glass and the calcite is u s e d in industrial applications.
Their gradations are typical of commercial H M A f i l l e r s - - t h e y were chosen
for use in this study because of their relative p u r i t y and the
similarity of their gradation.
Table 2 gives the physical properties of the m i n e r a l fillers used

in this study. The grain-size analysis for the fillers was conducted
using a M i c r o t r a c k small-size particle analyzer (Leeds and Northrup
Model 7995-30), w h i c h applies light scattering techniques to a dilute
suspension of powders in d i s t i l l e d water. The surface area was m e a s u r e d
using a BET Q u a n t a c h r o m e Sorption System (Model S-7) with n i t r o g e n gas
as the adsorbate. The specific gravities were m e a s u r e d u s i n g h e l i u m gas
and a Q u a n t a c h r o m e s t e r e o - p y c n o m e t e r (Model SPY-2).
For each a s p h a l t - f i l l e r m a s t i c the f i l l e r - t o - a s p h a l t ratio was

held constant at 0.5 by volume. The ratio was selected to represent a
typical ratio used in a d e n s e - g r a d e d HMA [~] . The m i x i n g of the mastic
was done after separately h e a t i n g the asphalt and filler in an oven--the
actual m i x i n g was a c c o m p l i s h e d in a container immersed in an oil bath
m a i n t a i n e d at 150~ (302 ~ . A mechanical m i x e r was u s e d to
c o n t i n u o u s l y stir the asphalt as the filler was added.
Table i-- Properties of asphalts used in the study.
Property I AAD-I
Kinematic Viscosity at 135aC,

cSt, ASTM D2170 316 246 582 590
Absolute Viscosity at 60~ P, 1140 1950 3320 2040

A S T M D2171
Penetration at 25~ 0.01 mm, 133 53 70 63
A S T M D5
Softening Point Temperature, ~ 44 50 49 50
ASTM D36 |
Corbett Component Analysis,

Percent
Asphaltenes 23 6 21 4
Polar Aromatics 41 51 42 50
Napthene Aromatics 25 33 30 42
Saturates 9 9 5 2
Table 2-- Physical properties of mineral fillers used in the study.
.- .,, ,
Gradation Calcite I Quartz
Size, ~m Percent Passing Percent Passing
42.2 100 100

39.9 88 93
21.1 76 79
14.9 66 63
10.6 57 47
7.5 47 37
5.3 36 26
3.7 26 19
2.6 16 13
1.7 12 8
1.0 8 5
Specific
Surface, m2/g 2.013 1.699
r
Specific
Gravity 2.705 2.645
The t e s t i n g p r o g r a m included a rheological c h a r a c t e r i z a t i o n of each of

the m a s t i c s at high a n d low temperatures. The m a s t i c s were t e s t e d after
m i x i n g in the u n a g e d c o n d i t i o n as well as a f t e r a g i n g w i t h the SHRP
p r e s s u r e a g i n g vessel (PAV) at 71~ (159.8 ~ for 6 days a n d at 100~
(212 ~ for 20 h. Four mastics, two asphalts each m i x e d with the
quartz and calcite, were selected for the f a i l u r e characterization. The
failure t e s t i n g was done at d i f f e r e n t t e m p e r a t u r e s a n d strain rates to
o b t a i n the failure envelopes.
Rheoloqical Evaluation A dynamic m e c h a n i c a l analyzer, R h e o m e t r i c s

RMS-800, was u s e d to m e a s u r e the dynamic shear m o d u l u s and phase angle
of the m a s t i c s at 25, 35, a n d 45~ (77, 95, and 113 ~ A frequency
sweep covering the range of 0.1 radians/s to i00 radians/s was o b t a i n e d
at each test temperature. The details of this e q u i p m e n t and m e t h o d of
c h a r a c t e r i z a t i o n have b e e n d e s c r i b e d in detail elsewhere [16,17].
Basically, the test p r o c e d u r e consists of p l a c i n g the test s a m p l e b e t w e e n
two p a r a l l e l plates and applying a s i n u s o i d a l l y a l t e r n a t i n g shear
d i s p l a c e m e n t to the two plates. The amount of torque r e q u i r e d to
p r o d u c e the selected strain amplitude is then r e c o r d e d and, by d i v i d i n g
the m e c h a n i c a l shear stress by the shear strain, c o m p l e x modulus, G*,
that is a function of test temperature a n d a n g u l a r f r e q u e n c y can be
determined. Because the m a s t i c behaves as a v i s c o e l a s t i c fluid at these
temperatures, the a p p l i e d strain a n d the r e s u l t i n g stress will be out of
p h a s e w i t h each other. This phase difference is c h a r a c t e r i z e d by a
p h a s e angle, 6.
The R h e o m e t r i c s Model 800 mechanical a n a l y z e r does not p r o d u c e

sufficient torque to y i e l d reliable l o w - t e m p e r a t u r e rheological
m e a s u r e m e n t s and, therefore, the Penn State b e n d i n g b e a m r h e o m e t e r was
u s e d for the l o w - t e m p e r a t u r e rheological measurements. The b e n d i n g b e a m
device was u s e d to obtain 4-minute creep curves at -5, -15, and -35~
(23, 5, and -31 ~ . The Penn State B e n d i n g B e a m R h e o m e t e r is d e s c r i b e d
elsewhere [18]; however, the device is quite simple in terms of its test
configuration. The b e a m to be tested is s u p p o r t e d on its ends a n d a
constant creep load is a p p l i e d at the m i d p o i n t of the beam. By
e m p l o y i n g the c o r r e s p o n d e n c e p r i n c i p l e a n d m e a s u r i n g the d e f l e c t i o n of
the m i d p o i n t as a function of time it is p o s s i b l e to calculate the creep
compliance as a function of loading time.
In order to facilitate the analysis of the rheological data, the

creep a n d d y n a m i c shear data were c o m b i n e d into a single m a s t e r curve
u s i n g t i m e - t e m p e r a t u r e superposition. This technique is i l l u s t r a t e d in
Fig. i, where the l o g a r i t h m of flexural stiffness is p l o t t e d versus the
l o g a r i t h m of loading time or frequency. By "shifting" the data a l o n g
the time or f r e q u e n c y axis it is p o s s i b l e to generate a "master curve"
at a p r e s e l e c t e d reference temperature. In this case, the reference
temperature was s e l e c t e d as 25~ A l s o shown in Fig. 1 are the series
of "shift factors" that are a function of temperature. Thus, the m a s t e r
curve defines the time d e p e n d e n c y of the stiffness a n d the shift
functions define the temperature d e p e n d e n c y of the stiffness.
In order to p r o v i d e c h a r a c t e r i s t i c values for each of the m a s t e r c u r v e s

so that the effect of f i l l e r - a s p h a l t interactions on the stiffnes ~ of
the m a s t i c can be c o m p a r e d on a q u a n t i t a t i v e basis, a m o d i f i e d
h y p e r b o l i c model was u s e d to describe the mastercurves. This model has
been u s e d at Penn State to a c c u r a t e l y model the stiffness m a s t e r c u r v e of
asphalts (19,20). The model is r e p r e s e n t e d as follows:
D
o
E
o
U
o
.J
Temperature
Loq Reduced Frequency !
Fig. 1 - - T i m e - t e m p e r a t u r e superposition concept and construction of

r h e o l o g i c a l m a s t e r curve.
log(G') = {G~-G~} { 0.5 [~. - ( ~ + 412) '~] } + G'~
where
G'(~) = complex modulus as a f u n c t i o n of f r e q u e n c y , Pa
Gg = g l a s s y modulus, Pa
G~ = m o d u l u s at c r o s s o v e r f r e q u e n c y , d e f i n e d as the p o i n t
on the m o d u l u s axis w h e r e the g l a s s y a s y m p t o t e a n d the
viscous asymptote intersect
~ = f r e q u e n c y at w h i c h the g l a s s y asymptote a n d the v i s c o u s

asymptote intersect
I = 2 / [ e x p (x) +exp (-x) ]
x = .339 ~, + 0 . 0 0 6 3 7 ~
~n = [~o + l o g ( ~ ) + aT(T)] /[G,* - G~*]
= loading frequency, rad/s
aT(T) = temperature shift factor, log r a d i a n s / s
The above model is b a s e d on the observation that all asphalts show a

viscous asymptote at v e r y long times where the b e h a v i o r is c o m p l e t e l y
Newtonian. A t very short loading times the b e h a v i o r reaches an elastic
b e h a v i o r with moduli that are a p p r o x i m a t e l y the same for all asphalts.
The point of i n t e r s e c t i o n of the elastic and viscous asymptotes is
called the crossover point with an a s s o c i a t e d crossover modulus and
frequency. Fig. 2 illustrates the basic concepts involved in d e v e l o p i n g
the model. The p a r a m e t e r s selected for the model have physical m e a n i n g
as follows:
G~__ [ Glassy Asymptote,

5
ul
x j ',
._1
l o g Reduced Frequency
Fig. 2 - - S c h e m a t i c d i a g r a m depicting the p a r a m e t e r s of the rheological

model u s e d to fit the binder LVE properties.
9 ~ locates the m a s t e r c u r v e along the loading time scale
9 G'o controls the shape of the mastercurve and represents the

t i m e - d e p e n d e n c y of the modulus
A typical m a s t e r curve for a m a s t i c showing all the data points,

for both the b e n d i n g b e a m creep tests and the dynamic shear tests are
shown in Fig. 3. The solid line represents the fitted model and
demonstrates the v a l i d i t y of the model when a p p l i e d to voidless mineral
filler-asphalt mastics.
M A S T ~ TYPE: AAK--1 + C A L Q T E
REFERENCE "rEMERATURE: 2 5 C
11
+ I~._.A~RED - - F I T T E D
"6"
9
0
o 5
3 J i 1 I
-5 -2 I 4 7 ~3
LOG REDUCED FEQLE~Y. UO(;(rod/,)
Fig. 3--Typical example of m e a s u r e d data and fitted function for one of
the study mastics.
Fracture Characterization The rheological m e a s u r e m e n t s d e s c r i b e d

above account for the s t r e s s - s t r a i n b e h a v i o r of asphalt cement; however,
these m e a s u r e m e n t s do not account for strength or failure behavior. The
complete c h a r a c t e r i z a t i o n of the fatigue and fracture p r o p e r t i e s
requires an extensive testing p r o g r a m and a c o n s i d e r a t i o n of fracture
mechanics. This is beyond the scope of the current paper and instead a
l e s s - s o p h i s t i c a t e d e v a l u a t i o n of the low-temperature direct tensile
strength p r o p e r t i e s was conducted using the same test protocol that is
being considered for the SHRP binder specification. B a s i c a l l y this
protocol consists of conducting direct tension tests at the following
test temperatures and rates of elongation: 25 n~n/m at 5~ 15 m m / m at
0~ 7.5 and 25 m m / m at -5~ 7.5 rmn/m at -10~ and 2.5 mm/m at -30~
A p h o t o g r a p h of the test specimen is shown in Fig. 4. The

specimen is cast in a silicone rubber m o l d with Plexiglass
(Polymethylmethacrylate or PMMA) end pieces. The straight section of
the test specimen is 6-mm b y 6-~n in c r o s s - s e c t i o n and 18-mm in length.
The resulting data, stress, strain, and energy to failure, are shifted
along the time axis using the same shift factors u s e d for the
rheological m e a s u r e m e n t s [21] . This procedure results in a m a s t e r c u r v e
of the shape shown in Fig. 5.
A q i n q Procedures Conventional aging tests in accordance with A S T M

Standard Test M e t h o d for Effect of Heat and Air on A s p h a l t i c Materials
(Thin-Film Oven Test) (D 1754-87) and the A S T M S t a n d a r d Test M e t h o d for
Effect of Heat and Air on a M o v i n g Film of A s p h a l t (Rolling T h i n - F i l m
Oven Test) (D 2872) cannot be used for filled systems because the
fillers settle from the asphalt at the high temperatures, 163~
(325.4 ~ used in these tests. The Pressure A g i n g Vessel (PAV) test
p r o c e d u r e d e v e l o p e d by others and as used by SHRP can be u s e d with

filled systems as a result of the lowered test temperatures that do not
reduce the v i s c o s i t y of the asphalt cement to the extent that settlement
is a problem. In the PAV test, asphalt cement or m a s t i c is p l a c e d in a
conventional thin film oven test (ASTM D 1754) p a n to a depth of 6 m m
(50 ml of mastic) and aged for 6 days at 71~ (159.8 ~ The vessel is
p r e s s u r i z e d with air to 2.07 MPa. These conditions m i m i c aging that
t y p i c a l l y occurs in the field in moderate climates a f t e r 5 to i0 years
of service.
Fig. 4--Direct tension test device and specimen configuration.
20 , I I I I I I I I I I
U
+Q
15
o~ CC < ~.__:~,"
+' /
,s 10
;/
CO
5 -1
b-
t__ I I I I I I I I I
-5
10-B10-~10-'+-10-~10-210 -+ 100 10 + 10 z 103 104. 105 10 e
Reduced Time, s
Fig. 5--Typical example of failure strain m a s t e r curves of u n f i l l e d and

filled asphalts.
Work conducted as part of the SHRP A - 0 0 2 A project, B i n d e r

C h a r a c t e r i z a t i o n and Evaluation, has shown that at low t e m p e r a t u r e s
asphalt cement can u n d e r g o a time-dependent collapse of free volume with
a concomitant increase in modulus (22). This type of aging is called
physical aging in the p o l y m e r a n d plastics industry (23) but has been
called physical hardening b y the authors to clearly d i s t i n g u i s h the
p h e n o m e n o n from oxidative aging. With time, over a m a t t e r of days or
weeks, physical h a r d e n i n g m a y result in 3- to 4 - f o l d increases in
stiffness, which is completely reversible with h e a t i n g to room
temperature.
Previous work has shown that the effect of the physical h a r d e n i n g

is to shift the mastercurve to longer loading times and that physical
h a r d e n i n g can be treated as a shift along the loading time axis. The
magnitude of the shift is dependent upon the storage temperature and
time of storage. Thus, to q u a n t i t a t i v e l y describe physical hardening,
it is possible to develop an isothermal storage shift factor that is a
function of storage time and temperature only. A temperature of -15~
and a storage time of 2 h were a r b i t r a r i l y chosen as reference points
and the data were shifted with respect to these conditions to determine
the physical h a r d e n i n g shift factors.
DISCUSSION OF T E S T RESULTS
A typical set of c o m p l e x modulus, G*, m a s t e r c u r v e s for one of the

systems, SHRP asphalt AAM-1 f i l l e d with quartz and calcite, is shown in
Fig. 6. In this Fig. (and in subsequent Fig.s) the individual data
points, n e a r l y one t h o u s a n d in number, have b e e n o m i t t e d for the sake of
clarity 9 From the sets of curves generated for the o t h e r systems,
asphalts AAK-I, AAG-I, and A A D - 1 w i t h calcite a n d quartz, similar
results were observed. Consistently, the a d d i t i o n of the filler h a d
little effect on the overall c o m p l e x modulus, G*, ( w h i c h can be
s e p e r a t e d into two components-- a storage component c a l l e d the storage
m o d u l u s or G' a n d the loss component called loss m o d u l u s or G" ) at low
temperatures. However, the loss component of the c o m p l e x modulus, G",
i n c r e a s e d at the h i g h e r temperatures. This is f u r t h e r i l l u s t r a t e d in
Fig. 7 where the m o d u l u s ratio, filled/unfilled, is given for G" a n d G"
for the quartz and calcite m i x e d with asphalt AAK-I. The m o d u l a r ratio
for G" is n e a r l y equal to that for G', and this was true for all of the
systems tested. In summary, the a d d i t i o n of the filler h a d little
effect on the e l a s t i c response at all t e m p e r a t u r e s but i n c r e a s e d the
viscous component of the response only at the u p p e r range of test
temperatures. These trends are f u r t h e r s u b s t a n t i a t e d by the data
p r e s e n t e d in Table 3, where the g l a s s y m o d u l u s a n d the c r o s s o v e r m o d u l u s
for each m a s t i c are tabulated.
12
FILER TYPE
9- CALCITE .---Q U A R T Z -- NONE
Q. I0 B0
V
0
0
-.I
Q@
W
D_J 8
"5
W
U~
X
"'
--I
Q_
6
:f/ %
C.)
C.9
0 20
_I
!
2 ' 0
0 4 8
LOG REDUCEDFREQUENCY,LOG(r~:l/s)
Fig. 6 - - C o m p a r i s o n of m a s t e r curves for asphalt AAM-I before a n d a f t e r
m i x i n g with study fillers at a reference t e m p e r a t u r e of 25 ~ (77 ~
Table 3-- Calculated rheological parameters of mastics.
Binder Treat- G~, Glassy G:, Cross Over tc, Cross Over
ment M~dulus Log Modulus log (Pa) Time log(s)
(Pa)
AAK-I Unaged 9.418 (0.0613) a 7.78 (0.0332) 2.233 (0.0699)
AAK-I +
Calcite Unaged 9.896 (0.0389) 8.116 (0.0207) 2.899 (0.0489)
Quart z Unaged 9.968 (0.0441) 8.149 (0.0231) 2.958 (0.0553)
AAG-I Unaged 9.192 (0.0333) 8.289 (0.0230) 12.962 (0.0688)

AAG-I +
Calcite Unaged 9.663 (0.0066) 8.737 (0.0058) 13.290 (0.0189)
Quartz Unaged 9.674 (0.0105) 8.748 (0.0089) 3.341 (0.0301)
AAD-I Unaged 9.334 (0.0205) 7.679 (0.0108) 1.661 (0.0205)

AAD-I +
Calcite Unaged 9.897 (0.0246) 8.126 (0.0139) 2.197 (0.0289)
Quartz Unaged 9.797 (0.0262) 8.156 (0.0153) 2.182 (0.0318)
AAM-I Unaged 9.074 (0.0246) 7.570 (0.0149) 3.640 (0.0446)

KAM-I +
Calcite Unaged 9.548 (0.0178) 7.877 (0.0110) 4.496 (0.0391)
Quartz Unaged 9.600 (0.0122) 8.007 (0.0079) 4.147 (0.0260
AAK-I +
Calcite iPAV 9.968 (0.0476) 7.947 (0.0206) 3.933 (0.0596)
Quartz PAV 10.050 (0.0201) 8.003 (0.0201) 3.911 (0.0577)
AAG-I +
Calcite PAV 9.481 (0.0134) 8.552 (0.0084) 4.203 (0.0531)
Quartz PAY 9.516 (0.0107) 8.577 (0.0067) 3.999 (0.0352)
a Numbers in parentheses represent the standard error of parameter.

Note: Master curve reference temperature = -15 ~ (-5 ~ Reference
Isothermal Age = 2 h.
ANDERSON ET A L ON MINERAL FILLER-ASPHALT MASTICS 143
20
- - C+ cdcite
-- - C-, q u o r t z
- - - G" ( : d c i t e
.... G" q u o r t z
..J
LIJ
_J
.J
i,
lo
d
<
n,"
i I , , i i I i J i I
0 '
-5 0 5 I0
LOG REDUCED FREQUENCY, LOG(rud/s)
Fig. 7--The ratio of G" and G" of filled to u n f i l l e d as a function of

frequency for asphalt AAK-I at 25~ (77 ~ .
For the two fillers that were tested, quartz and calcite, the
relative increase in the loss modulus with the addition of the filler
was a s p h a l t - s o u r c e as well as m i n e r a l - t y p e - d e p e n d e n t . This is
i l l u s t r a t e d in Fig. 8 where the relative modulus, G" f i l l e d / ~ unfilled,
for calcite is shown. The moduli ratios (calcite) at 7 rad/s are 0, I,
2, and 3 for asphalts AAK-1, AAG-I, AAD-I, and AAM-I, respectively. The
c o r r e s p o n d i n g ratios (not shown in Fig. 8) for the quartz m a s t i c were I,
2, 3, and 4, respectively. At the low temperatures the o b s e r v e d moduli
ratio are a p p r o x i m a t e d b y the law of m i x t u r e s when the m e a s u r e d glassy
modulus of the asphalt and typical h a n d b o o k values for the e l a s t i c
modulus of the mineral fillers are used:
i/~m = (c+)(I/E,) + (i-~)(I/E,)
where Em, E~, and E 2 are the moduli of the m i x t u r e and components 1 and
2, respectively, and cv is the volume c o n c e n t r a t i o n of component 1.
At the other temperature extreme, the o b s e r v e d modulus ratios are

in agreement with the ratios p r e d i c t e d by various equations reported b y
others [19], ranging from 16 to 20. None of the systems tested were
h i g h l y reinforcing, w h i c h w o u l d require m o d u l a r ratios a p p r o a c h i n g or
e x c e e d i n g i00.
A l t h o u g h the time d e p e n d e n c y of the theological response is

important, the temperature d e p e n d e n c y of the response is also important.
Temperature shift factors p r o d u c e d from the g e n e r a t i o n of m a s t e r c u r v e s
for two of the asphalts with and without filler are shown in Fig. 9.
The shift factors are r e l a t i v e l y u n c h a n g e d by the a d d i t i o n of the fines.
Thus it can be concluded that temperature d e p e n d e n c y of the asphalt and
the m a s t i c are e s s e n t i a l l y the same but the time d e p e n d e n c y of the

v i s c o u s component of the response is d e c r e a s e d at the u p p e r range of
service temperatures.
F I ( F R TYPE: CALCITE A T 5 0 ~ B Y V O L U M E
20
ASR-LALT T Y P E
-- AAK-I
-- A A G - I
tm -- A A D - I
i,i
9--- AAM--I
\
o
LIJ
\
._I
\
12
=<
i,
O
0 I I I I I I I I I I I I '
-~ -5 0 5
R E D U C E D F R E Q U E N C Y . LOG(rod/s)
Fig. 8--Ratio of moduli of f i l l e d to u n f i l l e d asphalts as a function of

f r e q u e n c y for the four s t u d i e d asphalts.
A typical direct t e n s i o n m a s t e r c u r v e was shown in Fig. 5.

A l t h o u g h the l o w - t e m p e r a t u r e r h e o l o g y was c h a n g e d little w i t h the
a d d i t i o n of the fillers, the effect on the tensile failure test results
are m o r e dramatic. The a d d i t i o n of the fines tends to increase the
tensile strength with a m o d e s t increase in the tensile strain at
failure. The net result is a significant increase in the stress to
failure as shown in Fig. i0 (note that the scales in Fig. i0 are
logarithmic). A l s o of importance is the o b s e r v a t i o n that the a d d i t i o n
of the fillers tends to shift the s t r a i n - t o - f a i l u r e m a s t e r c u r v e s (not
shown) to shorter times. This shift implies that the a d d i t i o n of the
fillers gives the appearance that the f i l l e d m a t e r i a l s are r e s p o n d i n g as
if they are b e i n g strained at a more r a p i d strain rate than the u n f i l l e d
asphalt.
The last c o n s i d e r a t i o n is that of the influence of the mineral

surface on the o x i d a t i v e aging a n d p h y s i c a l h a r d e n i n g of the asphalt.
Table 3 lists the rheological p a r a m e t e r s for the P A V aged mastics.
C o m p a r i n g these p a r a m e t e r s to those of the u n a g e d m a s t i c s it is clear
that the m a s t i c s are not o n l y b e c o m i n g h a r d e r but the rheological type
is changing. To clarify this point, a n d to study the relative effect of
the two different fillers, the ratio of G" of aged to the u n a g e d is

p l o t t e d in Fig. II. The plot indicates that the ratio is changing with
f r e q u e n c y v e r i f y i n g that aging ratios b a s e d on 6 0 ~ v i s c o s i t y are n o t
reliable p r e d i c t o r s of modulus changes due to aging. The Fig. also
indicates that the effect of asphalt source was g r e a t e r than the effect
of mineral type. Physical h a r d e n i n g was d e f i n e d by the isothermal shift
factors. Fig. 12 depicts an e q u a l i t y plot of shift factors of u n f i l l e d
and filled asphalts at different isothermal ages. The plot indicates
that the rate and magnitude of the physical h a r d e n i n g of the u n f i l l e d
asphalts is not a f f e c t e d by the presence of the calcite fillers. The
same b e h a v i o r was observed for the quartz filler.
5
BNDs TYPE
AAD-1 -i- . V a - 1 + ,:=kee -6- A~-I +
0
I--
0
<
I,
td
C~
,q
< -.5
I--
--1~ I I I I i I ! t l t I
--6O -3O 0 3O 6O
TEMPERATURE. C
Fig. 9--Temperature shift factors of asphalts AAM-I and AAD-I before and
after m i x i n g with study fillers.
In closing, a number of findings have been stated with regard to

mineral filler-asphalt interactions. It must be clearly u n d e r s t o o d that
these findings are b a s e d on a v e r y limited data set--additional sources
of filler as well as asphalt must be c o n s i d e r e d before the findings can
be considered comprehensive.
101 I I I I i L I
Calcite
~.~-..................
................ -.--.~... / 9
---.-..T .............~ ~ , w ...... '...
Q ............. "-T..,~ " "', ".
0
Q_
(n
i0 ~
U3
CD
AAG-1
i,
1 0 -- 1 ........ [ ........ ~ ....... ~ ....... J ....... ~ ...... a ........ I ........ I ....... d ....
10-5 10 -4 10-3 10-2 10 -1 10 0 101 10 2 10 3 10 4 10 5
Reduced Time, s
Fig. lO--Change in s t r e s s - t o - f a i l u r e master curve of asphalt AAD-I after

m i x i n g with the two fillers.
O 81
0
MASTICTYPE
< K & C.dcite
Z
-- K & Ouortz
--- G & Calcite
---- G & ( ~ r t z
6
8
<
< 4.
.....
0 i , I I I I i I I I l i I ,
-10 -5 0 5
REDUCED FREQUEt~Y. LOG(r~l/s)
Fig. ll--G" ratio for asphalt AAK-I and AAG-1 m i x e d with the two fillers
before and after aging in the PAV.
__o 150
._o
125
O
I00
t)
(D .7.5
.50
t)
.25
o
U_
.00
c- f
~p I! I i I I ' i i , 1 I I I I , Is l l i I ~ , I , s * i I I * , *
-15
b
-r --~5 .00 .25 .50 .75 tOO t25 150
Horden{ng Foctors for Unfi~ed Asphdts, log(s)
Fig. 1 2 - - E q u a l i t y plot of physical hardening shift factors o b t a i n e d for

the four asphalts before and after m i x i n g with calcite.
EFFECT O N PAVEMENT PERFORMANCE
Several factors are g e n e r a l l y identified as the m a i n causes of

p a v e m e n t failures: moisture damage, premature aging, low-temperature
cracking, fatigue cracking, and rutting. Except for moisture damage
each of these factors is to some degree related to the rheology or
failure p r o p e r t i e s of the asphalt binder. U s i n g this hypothesis, a
n u m b e r of the p a r a m e t e r s m e a s u r e d in this study were related on a
qualitative basis to potential pavement performance.
Low-Temperature Thermal Shrinkaqe Crackinq Thermal cracking is the

m a i n distress types o b s e r v e d at low in-service temperatures. The
buildup of thermal stress is caused b y thermal cooling a n d cracking is
b e l i e v e d to occur when the thermal shrinkage stresses e x c e e d the tensile
strength of the mix. In a number of previous studies the stiffness
value at long loading times, ranging b e t w e e n 7200 s and 20000 s, has
been u s e d to relate binder p r o p e r t i e s to cracking [24]. The hypothesis
is that pavement cracking initiates in the b i n d e r and that b i n d e r
stiffness is an indicator of p r o p e n s i t y for thermal cracking.
In k e e p i n g with the literature, the modulus at 104 rad/s (i000 s)

was u s e d as an indicator of sensitivity to thermal cracking. To compare
the binders and mastics tested in this study, the moduli at -15~ and 10-
4 rad/s were compared (see Table 4). For all of the mastic systems the
relative increase in modulus is r e l a t i v e l y small when compared to high
temperatures and v e r y similar for the different systems, a p p r o x i m a t e l y a
5- to 6-fold increase. Small differences are seen b e t w e e n the two
different fillers: for AAK-1 and AAD-1 (highly polar), the moduli of
quartz mastics are slightly higher than the the moduli of calcite
mastics. For AAG-1 and AAM-I (less polar), the quartz mastics show
lower moduli.
A l t h o u g h the modulus of the m a s t i c s at low t e m p e r a t u r e s is higher,

the tensile strength p r o p e r t i e s also show a p r o n o u n c e d change, as shown
in Table 4. The energy to failure for the mastics is c o n s i d e r a b l y
higher than for the u n f i l l e d asphalt. At -10~ and i000 s the energy to
failure of both mastics made with asphalt AAD-1 is 5 to 6 times that of
/~zd9-1, which offsets the increase in modulus. The increase in toughness
of mastics relative to u n f i l l e d asphalt has been r e p o r t e d before [25]
and should result in an increased resistance of H M A to cracking.
Fatique Craekinq This type of cracking m a y be the least

u n d e r s t o o d distress m e c h a n i s m in terms of the role of the binder. An
energy approach to fatigue cracking is being a d o p t e d by other SHRP
researchers to explain fatigue [26, 27]. This approach considers the
energy d i s s i p a t e d in cracking (or damage in general) as an indicator of
the fatigue life, (i.e., the Z W ~ = W~) . Therefore, to extend
fatigue life, the energy d i s s i p a t e d per cycle must be minimized. For a
v i s c o e l a s t i c material the W ~ is d i r e c t l y p r o p o r t i o n a l to the p r o d u c t s
of stress, strain, and sin (a) . Consequently, to minimize the W~c~ it
is always preferable to reduce the phase angle (a) . However, selecting
a criterion for the modulus is not so straightforward: in stress-
c o n t r o l l e d fatigue a higher modulus is p r e f e r r e d because it results in a
lower strain and thus longer fatigue life. For c o n t r o l l e d - s t r a i n
fatigue a lower modulus is p r e f e r r e d because a lower stress will be
p r o d u c e d for a given strain.
To compare mastics, Table 4 lists the values of G" and a at 0.0 ~

and 1 rad/s for the binders. The values show an increase of G" b y an
average of 4 to 5 while the a is dropping b y 3 to 4 degrees. For
Table 4--Derived rheological parameters reflecting possible effects on p a v e m e n t performance.
>
Low-Temperature Cracking Fatigue Cracking ~utting Z
CJ
G* 6 Failure Failure G* 6 Gv m
Binder ~O
(-15 ~ (-15 ~ Energy Energy (0 "C, (0 ~ 45 ~ 0"~
104 tad/s) (-i0 ~ (-30 ~ 1 rad/s) 1 rad/s) rad/s) 0
i0 ~ rad/s) Z
MPa deg i000 S) K P a I000 s) K P a KPa deg KPa
m
--i
>
AA.K- 1 5.9 54 167 37 1.65 r-
'AAK- I + Calcite 33. 49 770 33 27.2
0
AAK-1 + Quartz 39. 49 680 33 37. Z
AAK-I + Calcite Aged 81. 42 ii00 29 176.
AAK-I + Quartz Aged 91. 43 1240 29 246.
m
~J%G 1 - 49. 52 55. 0.97 670 23 1.05 >
AAG-I + Calcite 244 . 47 112 . 7.44 2360 21 9.2 i--
AAG 1 - + Quartz 219. 48 345 . 4 .I0 2470 19 i0.7 '11
458. 35 2290 9 38. F
AAG-I . Calcite Aged l--
AAG- 1 + Quartz Aged 387. 38 2280 Ii 40. m
JJ
AAD-I 1.9 59 1.2 1.19 74 41 0.67 ..Q

AAD-I + Calcite 12.6 54 8.8 6.53 360 37 7.5 "r
AAD-I + Quartz 12.7 55 6.6 6.14 380 37 4.2 >
q
AAM-1 24. 44 250 26 2.47!
AAM-1 + Calcite 130. 38 960 22 56. >
AAM-1 + Quartz 120. 40 1000 23 21. --t
09
fatigue cracking both of these changes are favorable under controlled-

stress conditions. In addition, the energy to failure increases
s i g n i f i c a n t l y (not shown in Table 4). Therefore, for s t r e s s - c o n t r o l l e d
fatigue the addition of filler is e x p e c t e d to improve resistance to
fatigue and, on relative bases, the improvements are similar for all
fillers.
For c o n t r o l l e d - s t r a i n fatigue it appears that adding the filler is

not favorable because G" is increasing while @ is decreasing. The
e n e r g y to failure, however, is also increasing significantly, w h i c h
indicates that fillers m a y not affect p r o p e n s i t y to fatigue cracking.
This effect will be n e g l i g i b l e at the lower t e m p e r a t u r e s and be more
evident as the pavement temperature increases. L a b o r a t o r y fatigue
and fracture testing v a l i d a t e d by field trial are required to validate
these hypotheses.
Rutting Rutting in the wheel tracks is a m a j o r form of pavement

distress in the U n i t e d States as well as in other countries. Although
aggregate interlock and interparticle friction are r e c o g n i z e d as being
the p r i m a r y mixture variables that control resistance to rutting, the
role of the binder is also well r e c o g n i z e d [26]. Certainly, before the
interparticle friction within the aggregate m a t r i x can be mobilized, the
b i n d e r must be d e f o r m e d and hence the flow p r o p e r t i e s of the binder
influence the rutting resistance of HMA. The specific rheological
p a r a m e t e r s that are responsible for resisting a c c u m u l a t e d p l a s t i c
d e f o r m a t i o n are not well defined; however, the viscous component of the
m o d u l u s is most likely an important c o n t r i b u t i o n to this resistance
[15].
The viscous component of the modulus Gv at 45~ for the asphalts and
m a s t i c s that were c h a r a c t e r i z e d in the study are given in Table 4.
Failure energy values given at 1000 s were d e t e r m i n e d from failure
e n e r g y m a s t e r curves. The viscous component of the modulus, Gv, was
d e t e r m i n e d b y dividing the steady state viscosity, ~o, by loading time,
t, where the loading time is a p p r o x i m a t e l y equal to the inverse of the
f r e q u e n c y in radians. For asphalt AAK-I the increase is a p p r o x i m a t e l y
20-fold. For other asphalts, it is at least 10-fold. The asphalt
sources-mineral filler specificity of the modulus is also d e m o n s t r a t e d
in Table 4. The specificity of this increase of ~ has been reported by
others and is well r e c o g n i z e d [6, 10].
For all asphalts tested the filler effect is favorable at the

volume c o n c e n t r a t i o n used, and it should s i g n i f i c a n t l y decrease the
a c c u m u l a t i o n of n o n r e c o v e r a b l e deformation. The improvement of
resistance to rutting caused by the filler is not a new subject:
mineral fillers were used in 1905 by R i c h a r d s o n in Washington, DC to
stabilize a w e a r i n g course and make it more resistant to d i s t o r t i o n
(rutting) [22]. Mineral fillers, however, can result in unsuitable HMA
if not well controlled: excessive stiffening of m i x has b e e n reported
b y some researchers [5]. This can be the result of a r e i n f o r c i n g effect
or the result of an excessive q u a n t i t y of filler, such that the filler
is no longer "floating" in the asphalt cement. In other r e p o r t e d
instances the mineral filler has served as an asphalt extender resulting
in an apparent "softening" of tk~ mix. It is apparent that correct
v o l u m e t r i c p r o p o r t i o n i n g of the mix as well as the r h e o l o g y of the
asphalt are v e r y important. A s p h a l t mastic should be c o n s i d e r e d in
d e v e l o p i n g a rational explanation of the role of mineral fillers and in
d e s i g n i n g H M A for increased rutting resistance.
Aqina The presence of mineral filler does not affect the magnitude or
the rate of l o w - t e m p e r a t u r e physical hardening. A l s o the data r e p o r t e d
in this p a p e r show that the m i n e r a l o g y of the filler has n e g l i g i b l e
effect on the level of oxidative hardening. H y d r a t e d limes r e p o r t e d l y
can retard oxidative hardening but this finding was not included in
other studies [28]. A d d i t i o n a l study is n e e d e d before conclusions can
be made regarding the role of mineral filler in the oxidative aging of
asphalt cement. The PAV procedure, e s p e c i a l l y when o p e r a t e d at 71~
(159.8 ~ is a satisfactory m e t h o d for studying the long-term aging of
asphalt-mineral filler mastics.
FINDINGS AND CONCLUSIONS
Two mineral fillers, quartz and calcite were a d d e d to four asphalt

cements and the rheological p r o p e r t i e s and failure p r o p e r t i e s of the
resulting m a s t i c s were d e t e r m i n e d using test methods d e v e l o p e d for neat
asphalt cement as part of the SHRP A - 0 0 2 A project, "Binder
C h a r a c t e r i z a t i o n and Evaluation". The test m e t h o d s developed for
c h a r a c t e r i z i n g the rheological p r o p e r t i e s (dynamic shear and flexural
creep) and fracture p r o p e r t i e s (direct tension) were found to be
applicable to voidless filler-asphalt cement mastics. B a s e d upon the
limited n u m b e r of materials studied, it was found that:
9 The m o d i f i e d h y p e r b o l i c model developed for p l a i n asphalt cements

can be used to describe voidless mineral filler-asphalt cement
mastics.
9 At v e r y low temperatures the simple law of m i x t u r e s accounts for

the stiffening effect of the mineral filler. At the upper range of
service temperatures the complex modulus is s i g n i f i c a n t l y increased
by the presence of the mineral filler.
9 The addition of the mineral filler does not affect the temperature
shift factors of the rheological response but does change the
frequency d e p e n d e n c y by shifting the relaxation times to longer
times, thereby stiffening the asphalt.
9 For the systems studied, the presence of the mineral filler did not
s i g n i f i c a n t l y affect the rate or level of oxidative or physical
hardening.
9 At low temperatures the mineral filler imparts a l e a t h e r y - l i k e

b e h a v i o r to the mastic, enhancing the strain and e n e r g y - t o - f a i l u r e
c h a r a c t e r i s t i c s of asphalt cement.
The qualitative e v a l u a t i o n of the p e r f o r m a n c e - r e l a t e d p r o p e r t i e s of the

m a s t i c s in this study led the authors to conclude that asphalt m a s t i c s
can p l a y a m a j o r role in defining the performance of the HMA. The data
collected also led the authors to believe that voidless mastics, similar
in volume c o n c e n t r a t i o n to the mineral filler-asphalt fraction in
typical HMA, can be c h a r a c t e r i z e d with the same test m e t h o d s as
d e v e l o p e d for neat asphalt cement. Further, when these test methods are
a p p l i e d to such mastics, fundamental information n e e d e d to characterize
the b e h a v i o r of H M A and improve their p e r f o r m a n c e can be developed.
REFERENCES
[i] Anderson, D. A., "Mechanical Behavior of A s p h a l t - M i n e r a l Powder

Composites and A s p h a l t - M i n e r a l Interaction," Ph.D. Dissertation, Purdue
University, 1971
[2] Larsen, H. J., Microscopic Analysis of Asphalt Aggregate Mixtures

Related to Performance, as part of a Strategic Highway Research Program
(SHRP) Idea Project, SHRP AIIR-13, Danish Ministry of Transport,
National Road Laboratory, Denmark, 1991.
[3] Halstead, W. J., and Zenewitz, J. A., "Changes in Asphalt Viscosity
During the Thin Film Oven and Microfilm Durability Tests," Symposium on
Microviscometry, ASTM STP No.309, American Society for Testing and
Materials, 1961, pp. 133-147.
[4] Ishai, I., and Craus, J.,"Effect of the Filler on Aggregate-Bitumen
Adhesion Properties in Bituminous Mixtures," Proceedings of the
Association of Asphalt Paving Technolouists, Vol. 46, 1977, pp. 228-258.
[5] Anderson, "D. A., "Guidelines for Use of Dust in Hot-Mix Asphalt
Concrete Mixtures," Proceedings of the Association of Asphalt Pavin~
Technologists, Vol. 56, 1987, pp. 492-516.
[6] Anderson, D. A., and Goetz, W. H., "Mechanical Behavior and
Reinforcement of Mineral Filler-Asphalt Mixtures," ProceedinGs of the
Association of. Asphalt Paving Technoloqists~ Vol. 42, 1973, pp. 37-66.
[7] Tunicliff, D. G., "Binding Effects of Mineral Fillers," Proceedings
of the Association of Asphalt Paving Technologists, Vol. 36, 1967, pp.
i14-156.
[81Winniford, R. S., "The Rheology of Asphalt-Filler Systems as Shown
by the Microviscometer," Symposium on Microviscometry, ASTM STP No.309,
American Society for Testing and Materials, 1961, pp. 109-120.
[9] Heukelom, W., and abd Wijga, P. w., "Viscosity of Dispersions as
Governed by Concentration and Rate of Shear," Proceedings of the
Association of Asphalt Paving TechnoloaistsL Vol. 40, 1971, pp. 418-437.
[10] Huschek, S., and Angst, C., "Mechanical Properties of Filler-

Bitumen Mixes at High and Low Service Temperatures," Proceedings of the
Association of Asphalt Paving TechnoloaistsL Vol. 49, 1980, pp. 440-475.
[ll] Craus, J., and Ishai, I., "Some Physico-Chemical Aspects of the
Effect and Role of the Filler in Bituminous Paving Mixtures,"
Proceedings of the Association of Asphalt Paving Technologists, Vol. 46,
1978, pp. 558-588.
[12] Ensley, E. K., and Scholz, H. A., "A Study of Asphalt-Aggregate
Interactions by Heat of In~nersion," Journal of Institute of Petroleurq,
Voi.58, 1972, pp. 95-I01.
[13] Anderson, D. A., and Chrismer, M., "Evaluation of Tests for
Characterizing the Stiffening Potential of Baghouse Dust in Asphalt
Mixes," Paper Presented at the Annual Meeting of the Transportation
Resrach Board, Washington, D.C., 1984.
[14] Tunnicliff, D. G., "Binding Effects of Mineral Filler," Proceedings
of the Association of Asphalt Paving Technologists, Vol. 36, 1967, pp.
i14-156.
[15] Anderson, D. A., Christensen, D. W., Bahia, H. U., "Physical

Properties of Asphalt Cements and the Development of Performance-Related
Specifications," Paper presented at the 1991 AAPT meeting in Seattle,
Washington March 2-4, 1991.
[16] Christensen, D. W., and Anderson, D. A., "Interpretation of Dynamic
Mechanical Test Data for Paving Grade Asphalt Cements," Paper Presented
at the 1992 AAPT Meeting, March 1992, Charleston, South Carolina.
[17] Bahia, H. U., Anderson, D. A., and Christensen, D. W., "The Bending
Bean Rheometer: A Simple Device for Measuring Low Temperature Rheology
of Asphalt Binders," Paper Presented at the 1992 AAPT meeting, March
1992, Charleston, South Carolina.
[18] Bahia, H. U., "Low-Temperature Physical Hardening of Asphalt

Cements," Ph.D. dissertation, Department of Civil Engineering, The
Pennsylvania State University, University Park, PA, 1991.
[19] Anderson, D. A., Christensen, D. W., Dongre, R., Sharma, M. G., and
Jordhal, P., "Asphalt Behavior at Low Service Temperatures," Report No.
FHWA-RD-88-078, U.S. Department of Transportation, FHWA, March 1990.
[20] Bahia, H. U., and Anderson, D.A.," Isothermal Low-Temperature

Physical Hardening of Asphalt Binders," Proceedinqs, International
Symposium for Chemistry of Bitumens, 5-8 June, Rome, Italy, 1991.
[21] Struik, L. C. E, Physical Aqinq in Amorphous Polymers and Other

Materials, Elsevier, Amsterdam-New York, 1978.
[22] Readshaw, E. E., "Asphalt Specifications in British Columbia for

Low Temperature Performance," Proceedinqs of the Association of Asphal~
Pavinq Technoloqists, Vol. 42, 1972 pp. 562-581.
[23] Rigden, P. J., and Lee, A. R., ,,The Brittle Fracture of Tars and
Bitumens," Journal of A~plied Chemistry, Vol. 3, 1953 p. 62.
[24] Van Dijk, W., "Practical Fatigue Characterization of Bituminous

Mixes," [roceedinqs of the Association of Asphalt Pavinq Technoloqists,
Vol. 44, 1975 pp. 37-74.
[25] Pell, P.S., and Cooper, K. E., "The Effect of Testing and Mix
Variables on the Fatigue Performance of Bituminous Materials,"
Proceedinqs of the Association of Asphalt Pavinq Technoloqists, Vol. 44,
1975 pp. 1-37.
[26] Valkering, C. P., Lancon, D. J. L., deHilster, E., and Stoker, D.

A., "Rutting Resistance of Asphalt M/xes Containing Non-Conventional and
Polymer Modified Binders,"Proceedinqs of the Association of Asphalt
Pavinq Technoloqists, Vol. 59, 1990 pp. 590-609.
[27] Richardson, C., Asphalt Construction for Pavements and Hiahwavs,

McGraw-Hill Co., New-York, 1913.
[28] Petersen, J. C., Plancher, H., and Harnsberger, P. M., "Lime

Treatment of Asphalt to Reduce Age Hardening and Improve Flow
Properties," Proceedinqs of the Association of Asphalt Pavinq
Technoloqists, Vol. 56, 1987 pp. 632-649
David A. Anderson, I Raj Dongre, 2 Donald W. Christensen III, 3
and Ervin L. Dukatz 4
EFFECTS OF MINUS NO. 200-SIZED AGGREGATE ON FRACTURE

BEHAVIOR OF DENSE-GRADED HOT-MIX ASPHALT
REFERENCE: Anderson, D. A., Dongre, R., Christensen, D. W.

III, and Dukatz, E. L. "Effects of Minus No. 2 0 0 - s i z e d
Aggregate on F r a c t u r e Behavior of D e n s e - G r a d e d Hot-Mix
Asphalt," Effects of Aqqreqates and Mineral Filler8 on
Asphalt Mixture Performance: ASTM STP 1147, Richard C.
Meininger, editor, American Society for Testing and
Materials, Philadelphia, 1992.
ABSTRACT: The effect of minus No. 200-sized material

(mineral filler) on the fundamental mechanical properties of
hot-mix asphalt is not well understood. In the work
reported in this paper a series of minus No. 200 (75 ~m)
mineral fillers were used to prepare and characterize
filler-asphalt mastics and hot-mix asphalt concrete. The
mineral fillers were blends of dust of fracture and baghouse
dust was sampled from seven sources. The gradation and the
void-filling characteristics of the mineral fillers were
determined. Mastics prepared with the fillers were
characterized with viscosity measurements at 60~ (140~ and
dynamic mechanical properties, storage modulus, loss
modulus, and tan delta, over a wide temperature range.
Mixtures containing the mineral fillers were prepared using
several different filler-asphalt ratios. Originally, the
behavior of the mixes was to be evaluated with a fatigue
test in which a test beam is supported on an elastic (low-
modulus rubber) foundation. This test method proved
unacceptable because the failures that occurred were in
shear rather than bending fatigue. As an alternative, the
IProfessor, Civil Engineering Department, Pennsylvania

State University, University Park, Pa-16802
2Research Assistant, Pennsylvania Transportation

Institute, Pennsylvania State University, University Park,
Pa-16802.
3Research Associate, Pennsylvania Transportation

Institute, Pennsylvania State University, University Park,
Pa-16802.
4Senior Materials Engineer, Vulcan Materials Company,

Birmingham, Alabama-35253-0187
154

Dec 22 07:29:10 EST 2011
ANDERSON ET AL. ON MINUS NO. 200-SIZED AGGREGATES 155
single-edge-notched-beam (SENB) fracture toughness test was

used. With this procedure it was possible to determine the
fracture toughness of the mixes as a function of
temperature. The source of the mineral filler, as reflected
by the properties of the mineral filler-asphalt mastics, was
correlated with the fracture toughness properties of the
mixes.
KEYWORDS: Fracture, fatigue, mineral fillers, mastics,

rheological behavior, viscoelasticity, dynamic mechanical
properties
INTRODUCTION
The effect of mineral filler and, in particular,

baghouse dust on the fatigue and fracture properties of
asphalt concrete mixtures is neither well documented in the
literature nor well understood by asphalt technologists.
The objective of this study was to generate test results
that will contribute to a better understanding of the
effects of mineral filler and baghouse dust on the fatigue
and fracture properties of hot-mix asphalt concrete.
Mixtures were prepared in the laboratory from seven
different aggregate sources and baghouse fines were added to
these mixes at two levels. Rheological properties of the
mineral filler-asphalt cement mastics and fracture
properties of the mixes were then measured.
The aggregates used in the mixes were sampled from

seven hot-mix production facilities in Georgia and dense-
graded wearing course mixes were prepared in accordance with
job-mix formulas (control mixes) from the respective plants.
Baghouse dust was then added to these control mixes at two
levels, generating 21 mixtures. Dynamic mechanical
properties were measured for the mastics using a resonant
beam technique at sub-ambient temperatures and capillary
v i s c o m e t r y at 60~ (140~ The original intent was to
evaluate the fatigue properties of the mixes by supporting
beams on an elastic foundation. This foundation support,
which consisted of multiple layers of rubber, did not yield
satisfactory resul~s. As an alternative, fracture
properties were measured for the mixtures using a notched
beam technique. The fracture mechanics parameter called the
critical J integral (or elastic plastic strain energy
release rate), J1c, was determined for the mixtures over a
range of test temperatures.
MATERIALS USED IN STUDY
Asphalt Cement
A single AC-20 asphalt was used in preparing the

mastics and hot-mix asphalt mixtures. Viscosity
measurements resulted in values of 2,360 Poises at 60 ~
(140 ~ using A S T M Standard Test Method for Kinematic
Viscosity of Asphalts (Bitumens) (ASTM D 2170) and 430 cSt at
135 ~ (275 ~ using ASTM Standard Test Method for Viscosity
of Asphalts by V a c u u m Capillary Viscometer (ASTM D 2171).
As described below, a DuPont 9900 Thermal Analysis System
was used to obtain dynamic mechanical measurements for the
asphalt cement at the sub-ambient test temperatures.
AuqreQate and Mineral Filler
The aggregate and baghouse dust samples were obtained

from seven different hot-mix plants operated by Vulcan
Materials Company: Dalton (D), Fairmount (F), Grayson (G),
Kennesaw (K), Lithia Springs (LS), Norcross (N), and
Stockbridge ( S ) . The high-calcium hydrated lime was
obtained locally. The aggregate sampled from the cold feed
was separated in the laboratory onto individual sieves and
recombined according to the respective job-mix formulas to
produce the seven control mixtures (see table i). The minus
No. 200 (75 ~un) fraction in the control mixes consisted of
the dust of fracture and 1 percent hydrated lime as used in
the production mixes. Each of the job-mix formulas
specified 6 percent passing the No. 200 sieve (75 ~n),
including the hydrated lime, expect for the Lithia Springs
mix, which contained 5 percent.
Additional mixture gradations for each source were

generated from the job-mix formulas by adding baghouse dust
(see table 2), to increase the total amount of material
finer than the No. 200 sieve (75 ~m) to 8 and i0 percent.
For these mixes the gradation above the No. 30 sieve was
opened up slightly to allow for the increase in minus No.
200 (75 ~n) material. This was necessary to provide
sufficient voids in the mineral aggregate (VMA) to
accommodate the additional mastic (dust plus asphalt).
Thus, for each aggregate source, mixes containing three
different minus No. 200 sieve (75 ~m) fractions were
produced. The mixture gradations are referred to by the
proportions of dust of fracture, hydrated lime, and baghouse
fines--5:l:0, 5:1:2, and 5:1:4, respectively. Since
aggregates from seven different hot-mix plants were used in
the project, this resulted in a total of 21 possible mastic
or hot-mix gradations (see table 3).
TABLE 1--Aggregate properties.
>
z
o
m
-n
0
Gradation, Percent Passing z
m
Source Type Sp. Gr. 3/8
in NO*
4 I NO"
8 I ~O.
16 >
100 200 F-
i|
r.:0 (.:o I.o [.o
iii i
0
Dalton I Limestone 2.69 97 75 46 28 19 12 8 6 z
E
Fatrmount Limestone 2.74 99 73 45 27 17 12 7 6 z
c
G0
Grayson G r a n i t i c gneiss 2.65 100 68 45 36 28 18 11 6 z
.o
Kennesaw Granitic gneiss 2.79 100 66 45 3.5 27 21 11 6
0
,o
25 17 10 $ 6O
Lichia Springs G r a n i t i c gneiss 2.63 100 66 45 32
N
m
Norcross G r a n i c l c gneiss 2.76 100 71 46 30 22 15 9 6
Scockbridge Granitic gneiss 2.63 100 67 65 36 28 20 12 6 X1

m
._z
ol
-4
158 ASPHA~ MIXTURE PERFORMANCE
Mastics
The mastics were prepared by mixing the AC-20 asphalt

c e m e n t w i t h the pre-blended minus No. 200 sieve (75 ~m)
TABLE 2--Baghouse dust gradation. All sources of fines

were i00 percent passing No. 200 (75 ~m).
Source Percent Percent Percent Percent DS0 D,,

Passing Passing Passing Passing (~)
40 ~m 20 ~m i0 ~m 5~m
Dalton 90 73 8 3 17
Fairmount 45 21 9 8 24
Grayson 57 34 20 12 35 14
Kennesaw 60 24 I0 7 33 21
Lithia Spring 93 63 46 32 11 4
Norcross 70 31 17 12 33 16
Stockbrldge 97 82 56 32" 9 a
f r a c t i o n s (dust of fracture, hydrated lime, a n d baghouse

dust) d e s c r i b e d above. The asphalt cement a n d m i n e r a l
f i l l e r w e r e mixed in one-quart paint cans i m m e r s e d in an oil
b a t h m a i n t a i n e d at 149 to 153 ~ (300 to 308 OF). Sufficient
a s p h a l t was added to the preblended mineral f i l l e r to
p r o d u c e the same asphalt-mineral filler ratios in the
m a s t i c s as in the corresponding mixes. The m a s t i c samples
w e r e m i x e d with a small propeller-type mixer o p e r a t i n g at a
m a x i m u m speed of 3000 rpm for at least 15 m i n but until all
v i s i b l e lumps of mineral filler were dissipated. After
mixing, a portion of each of the mastics was p o u r e d into
s e v e r a l ointment tins for further testing. T h e cans in
w h i c h the mastics were mixed were tightly s e a l e d and stored
at r o o m temperature, as were the ointment tins.
Considerablematerial was consumed during the attempt to
r e f i n e the b e a m on elast'c foundation fatigue t e s t i n g
procedure. As a consequence, the quantity of available
a s p h a l t cement and fines was insufficient to p r o d u c e the
5:1:0 a n d 5:1:4 mastics for the LS and N sources, although
all of the hot-mix beams were made.
TABLE 3--Properties of minus No. 200 (75 ~m) asphalt mastic.
I I I Ill I I
Ratio-- Bulk
Dust of Denos165 7~ee A i p h a l c
Percent Fracture: ot Dry in Haetio with Viscosity
Aggregate Mlnue NO. Baghouse Duets Compacted 5.9t Aaphalt a t 60 "C
Source 200 i n Hix Hydrated lime Yillo: in Hix ~STH D 2171 >
( v o l . %) (Pa-e) I z
1! m
Dalton Control. 6 580:1 1.47 50.9 1,210
co
B 55251 1.45 39.0 3,660 9
24,900 z
10 5;4:1 1.46 29.6
m
-i
Pairmount Control. 6 5:051 1.48 50.9 lp970
8 552s1 1.52 41.4 3,780 F
10 554,1 1.50 31.4 10,400 0
z
Grayio~ Control, 6 5JOel 1.36 47.8 1,820

8 552:1 1.42 36.5 4,790 c
1.40 28.2 26,100 09
10 5s451
z
0
Kenneiaw Control. 6 5:0sl 1.45 49.8 1,430
8 5;2sl 1.47 39.0 4,000 O
I0 5s4sl 1.47 29.7 15,100 co

N
. * B
m
LLthia Springs Control, 6 5:0s1 1.45 49.9 o
8 5;251 1.43 37.9 4,320
5z4zl 1.40 26.6 e e l
l0
m
Norcroao Con, roLe 6 5s0sl 1.38 47.3
8 5:2:1 1.38 35.4 4.900
m
1.40 26.2 a . o o9
10 5s4zl
Stockb.idge Control, 6 5:0:1 1.42 49.1 1,420 .-L
8 5;2:1 1.42 37.9 4,510 on

10 5:4;1 1.42 28.1 18,600
Hot-Mix Asphalt Concrete
The 21 different aggregate-gradation combinations

described above were used in preparing the asphalt concrete
mixtures for the fracture testing (table 3). The asphalt
content used for each of the 21 mixtures was 5.9 percent by
total mix weight as per the job-mix formulas. Beam
specimens measuring 76 m m by 76 nm~ by 330 m m (3 in by 3 in
by 16 in) were compacted with a Cox Model CS-1000 kneading
compactor. Mixing and compaction temperatures were
controlled to produce viscosities of 170 • cSt and 280 z30
cSt, respectively. The compaction procedure followed ASTM
Standard Practice for Preparation of Bituminous Mixture Beam
Specimens by Means of the California Kneading Compactor
(ASTM D 3202-83) and ASTM Standard Method for Preparation of
Bituminous Mixture Test Specimens by Means of California
Kneading Compactor ASTM ( A S T M D 1561) and is described in
detail elsewhere.[l] Particular care was taken to ensure
equal levels of compaction throughout the depth and length
of the beams. Three beams were manufactured for each
mixture-filler combination and each beam was notched at its
midspan with a diamond saw.
After compaction and cooling to room temperature, the

beams were placed in a freezer held at approximately -18 ~
(0 ~ for storage. This was done to minimize differences
in mechanical properties caused by oxidation and steric
hardening. In more recent work conducted at Penn State, it
has been observed that storage at low temperatures can
result in considerable time-dependent, albeit reversible,
stiffening of asphalt cement. [2] In retrospect, better
control of the storage temperature and time may have
improved the repeatability of the test results for both the
mastic and hot-mix beams.
EXPERIMENTAL METHODS
Tests were conducted on the neat asphalt cement, the

minus No. 200 sieve (mineral filler) fraction, the asphalt-
mineral filler mastic, and hot-mix asphalt prepared with the
different aggregates and mineral fillers. The experimental
p r o g r a m that was followed is described below.
Characterization of the Mineral Filler
Hydrometer tests were performed on the baghouse dusts

from each plant in order to determine their particle size
distribution (table 2). Contrary to common belief the
gradation of baghouse dust varies considerably from source-
to-source, depending on the aggregate being processed, the
presence or lack of primary dust collectors such as cyclones
or knockout boxes and other factors. [3] This is evidenced
ANDERSON ET AL ON MINUS NO. 200-SIZED AGGREGATES 161
by the baghouse dusts used in this study. The particle size

distributions for the various baghouse dusts are shown in
figure 1. The gradation of the F, G, K, and N baghouse
dusts are very similar, well-graded from 75 to less than l0
~m. The LS and S dusts are also w e l l - g r a d e d but are
100
90
80
7O
O~
r"
f/'l 60
O
o.
.,., 50
C
0
U
~-
D
411
30
20
10
0
10 100
Porticle Size. D./4 m

FIG. 1--Gradation of baghouse dust.
considerably finer than the F, G, K, and N dusts. The D

baghouse dust is essentially one sized with 80 percent of
the dust between ii and 20 ~m. Thus, the dusts represent
three gradings: coarse well-graded (F, G, K, and N); coarse
fine-graded (LS and S); and fine one-sized (D).
The compacted dry bulk density of mineral filler has

been used by others as measure of stiffening potential. [4,5]
In this procedure, the mineral filler is compacted into a
small metal container and the bulk specific gravity of the
dry compacted filler and th~ bulk specific gravity of the
mineral solids are used to cc.npute the voids in the dry
compacted mass.[6] The asphalt required to fill these voids
is considered as "fixed" asphalt. Only when asphalt in
excess of the fixed asphalt is added to an asphalt-mineral
filler mastic is the asphalt free to lubricate or reduce the

v i s c o s i t y of the mastic.
The dry compacted density for the 21 different blends

of dust of fracture, baghouse dust, and hydrated lime are
shown in table 3. Also shown in table 3 is the percent free
asphalt calculated by assuming that the fillers are added to
asphalt cement to form a mastic with the same filler-asphalt
ratio as used in the 21 mixes and that the mastic is
voidless. There is surprisingly little variation in the dry
bulk density values, both with respect to the source of the
filler or with respect to the amount of added baghouse dust.
This w o u l d indicate that when the different fillers are
added to asphalt cement, the consistency of the ensuing
mastic should be very similar.
Rheological Properties of the Asphalt and Asphalt-Mineral

Filler Mastics
The absolute viscosity at 60 ~ (140 ~ for each of the

asphalt mineral-filler mastics was determined in accordance
with A S T M D 2171 and the results are presented in table 3.
In order to characterize the rheological behavior at lower
temperatures, dynamic mechanical measurements were obtained
for the neat asphalt and the asphalt cement-mineral filler
mastics. These measurements were obtained in duplicate with
a DuPont 9900 Thermal Analysis System equipped with a 982
Dynamic Mechanical Analysis (DMA) Module and a liquid
m i t r o g e n cooling accessory. The D M A m o d u l e operates by
vibrating a small beam of material in bending at its
resonant frequency (see ref. 1 for details). The computer
and its associated software are used to compute the modulus
of the material in bending. For viscoelastic materials,
both the complex bending modulus and the phase angle are
determined and, by conducting the test at different
temperatures, the modulus and phase angle may be determined
as a function of temperature. Because the frequency of the
measurements is limited to the resonant frequency of the
beam, it is not possible to vary the frequency of the
measurements except by varying the dimensions of the beam.
The moduli and b e a m dimensions for the beams that were
tested resulted in frequencies that ranged from 10 to 30 Hz.
Additional details regarding the use of this equipment for
testing asphalt cement may be found elsewhere. [i]
Fatigue Testinq of Asphalt Concrete Beams
Initially, beams were tested in fatigue using a

procedure as reported in the literature in which the beams
are supported on an elastic (rubber) foundation and loaded
repeatedly at their midpoint until failure occurs. [7] The
beams were supported on 25.4-mm (1-in) thick slabs of
rubber, equal in plan dimensions to the dimensions of the
beams. Considerable effort--including varying the hardness

of the rubber beams, varying the number of slabs, and fixing
the ends of the beams--was expanded in attempts to obtain a
bending fatigue failure. Despite these efforts only shear
failures, immediately under the loading strip, were
observed, (see figure 2). The two major problems
encountered with this procedure are summarized below.
Beam on Elastic
Foundation
n ,',ad
Shear Cr
Beam
Rubber
FIG. 2--Schematic of beam on elastic foundation fatigue

test.
9 First, the ends of the beam rose upward when

loaded, thereby losing contact with the rubber
supports. With increasing load cycles the beams
took on a curved shape with both ends unsupported.
This partial support, which varies both during each
load cycle and over the loading history, makes a
closed-form analysis of the stresses within the
beam intractable and therefore eliminates the
possibility of relating the stresses to the number
of cycles to failure.
9 A second major problem with this configuration is

the presence of very large shear stresses under the
loading strip. The cracks that were observed were
most likely the result of shear stresses and not

bending stresses.
Because of the problems with the above procedure it was

abandoned, and with the remaining resources for the study
were directed to beams tested to determine fracture
parameters.
Fracture Testing of H0t-Mix Asphalt Concrete
Notched beam fracture tests have been used previously

by the authors to characterize the fracture properties of
asphalt concrete. Given the limited number of beams that
were available for testing and the desire to obtain
fracture/failure data at various temperatures, the notched
b e a m fracture testing was chosen instead of conventional
b e a m fatigue testing.
The fracture tests were conducted at four temperatures:

60 ~ (16 ~ 40 ~ (4 ~ 20 ~ (-4 ~ and -5 ~ (-20 ~
in accordance with a procedure previously developed at Penn
State.[l] In this procedure, a notch is cut in the
underside of the beam at its midpoint and the b e a m is loaded
in bending in 3-point loading until failure. The specimen,
MTS loading frame, and the three-point loading frame were
all kept in a constant temperature chamber at the test
temperature for 18-24 h before the beams were tested.
Details of the test equipment and the test protocol are
given elsewhere.
The notched beam test was used to determine J~, the

initial strain-energy release rate. This parameter is
indicative of the ease with which a crack can be propagated
in a material and is considered a fundamental material
property. Critical values of J integral in plane strain,
J1c, can be determined from load-deflection data of the
notched beams as follows:
9 Total energy, U T, under the load deflection curve

for each of the beams tested is computed by summing
the area under the curve up to the point of
failure, defined as the m a x i m u m load.
9 UT/b, the total energy per unit thickness, is then

plotted against notch depth, a. The slope of the
reciprocal of the notch depth versus the unit
energy per unit width, (i/b) (dUT/da) , is then
obtained through regresJion. This is repeated for
each test temperature.
9 The critical strain-energy release rate, also

called the elastic-plastic fracture resistance, J1c,
can be d e t e r m i n e d for each test temperature using

the following equation:
J,= = kUT/b(d-a) (I)
where
J~= = critical strain-energy release rate,
j/m 3
k = g e o m e t r i c constant
UT = unit energy, J
b = w i d t h of beam, n~n
d = d e p t h of beam, mm
a = n o t c h depth, m m
S u m p t e r and Turner have d i s c u s s e d equation 1 in

detail. [8] For l e n g t h - t o - d e p t h ratios equal to 4 a n d notch-
t o - d e p t h ratios b e t w e e n 0.5 and 0.7 the constant, k, in
e q u a t i o n 2 is equal to 2.00. Therefore, for these
conditions, equation I can be s t a t e d as [8]:
J1~ -- 2UT/b (d-a) (2)
E q u a t i o n 2 is v a l i d o n l y for the l e n g t h - t o - d e p t h and

n o t c h - t o - d e p t h ratios g i v e n above. Since the specimens u s e d
in this study s a t i s f i e d these requirements, e q u a t i o n 2 was
u s e d to calculate J~c u s i n g test results for a single n o t c h
depth, 12.7 mm (1.5 in). The results of the J~c c a l c u l a t i o n s
for the individual mixes are s u m m a r i z e d in table 4.
RESULTS AND DATA ANALYSIS
Rheological Characterization of A s p h a l t - M i n e r a l Filler

Mastics
T e s t i n g of the m a s t i c s i n v o l v e d absolute v i s c o s i t y at
60 ~ (140 ~ and d y n a m i c m e c h a n i c a l analysis at
t e m p e r a t u r e s ranging from -5 ~ (23 ~ to 15 ~ (59 ~ It
was found that the m i n e r a l fillers i n c r e a s e d the v i s c o s i t y
of the asphalt following r e l a t i o n s h i p s p r o p o s e d by p r e v i o u s
investigators. [9] The increase in v i s c o s i t y was found to be
almost c o m p l e t e l y a f u n c t i o n of v o l u m e c o n c e n t r a t i o n of
m i n e r a l filler, and was l a r g e l y i n d e p e n d e n t of source or
type of filler (see figure 3). The p e r c e n t free a s p h a l t
(see figure 3) affords a s l i g h t l y b e t t e r c o r r e l a t i o n w i t h
the m e a s u r e d v i s c o s i t y than the v o l u m e concentration. The
authors suspect that this is b e c a u s e this p a r a m e t e r also
a c c o u n t s for the p a c k i n g c h a r a c t e r i s t i c s of the fillers.
Unfortunately, the fillers that w e r e included in the test
p r o g r a m have v e r y similar p a c k i n g c h a r a c t e r i s t i c s as
i l l u s t r a t e d by the b u l k d e n s i t y of the dry c o m p a c t e d filler
(see table 3). From these data it was concluded that none
of the fillers that were tested are r e i n f o r c i n g - - n o
TABLE 4--Measured values of e l a s t i c - p l a s t i c fracture O~
toughness, Ju0-
>
J,., l b - i n / I n 3 -o
I
>
Percent Passlng
-5 "F 20 "F 40 ~ 60 ~
Source No. 200 in Mix
Dalton Control, 6 1.74 8.9 17.7 6.8 x

8 2.2 . e l . s a 6.9 c
i0 2.6 11.9 25. 12.2 m
Control, 6 1.82 11o0 15.4 11.2 m

Falrmont
8 2,6 . l e e o . 11.8 TI
O
10 3.9 17.0 17.3 10.8
>
Orayson Control, 6 3.2 6.5 17.0 13.2 z
3.9 . t o 11.6 c')
8 rtl
10 4.3 7.2 14.5 12.0
K e n n e e ~ i w Control, 6 2.2 7.9 15.8 9.2

8 2.9 18.1 10.1
10 3,3 9.9 19.'/ 11.5
Lithia Spring Control, 5 2.0 8.0 18.9 14.0

i a a 10.1
8 2.?
10 3,7 11.2 24. 12.4
, ,
NororosB Control, 6 1.6 5.6 16.8 9.7

o e e ~ e o
8 2.0
10 2.1 7.2 17.2 9.1
1.85 8.6 20. Q D Q

Stockbrs Control, 6
e 9 e Q O @
8 2.5 e e *
6.7 29. O G i
10 3.3
ANDERSON ET AL. ON MINUS NO, 200-SIZED AGGREGATES 167
physical-chemical interactions were demonstrated, since the

stiffening was the result of volume filling effects alone.
The resonant beam testing and the capillary viscosity

data were combined to construct plots of modulus versus
temperature, assuming that the capillary viscosity
measurements represent a coefficient of Newtonian viscosity,
7, and, therefore, that the complex modulus, G*(~) is
approximately equal to i/(3~/t) where t - I/~. A typical
plot of the complex modulus versus temperature is shown in
figure 4 for one of the mastic systems. The magnitude of
the stiffening effect is time- and temperature-
dependent. [I0] At shorter loading times and lower
temperatures where the phase angle approaches zero the
stiffening effect obeys the law of mixtures resulting in
stiffening ratios of three to five. [i0] At the higher
temperatures, as viscous flow is approached, the stiffening
effect is much greater, approaching i00 or more.
In order to compare the moduli for the different

mastics, statistical analyses were used to test the
hypothesis that the source and percentage of baghouse dust
affect the magnitude of the phase angle and the loss and
storage moduli. Although the hypothesis was supported, the
statistical analyses showed that the testing error
(coefficient of variation) for the resonant frequency device
is very large, averaging 27 and 20 percent for the storage
and loss moduli, respectively. This relatively poor
repeatability is the result of errors from various sources:
(I) difficulty in uniformly clamping the specimen, and the
occurrence of small cracks during clamping; (2) thermal
gradients within the specimen during the temperature ramp;
and (3) lack of precise control over the thermal history
prior to testing, because of the time required to clamp the
specimen. Based on the results, the authors cannot
recommend the resonant beam device for this purpose. The
simple bending b e a m rheometer or the torsion bar fixture in
a dynamic rheometer are preferred test techniques for these
materials. [Ii,12]
Normalized plots of the stiffening ratio formed by

dividing the modulus of the asphalt mastic by the modulus of
the neat asphalt are shown in figures 5 and 6 for the
storage and loss modulus, respectively. Stiffening ratios
for both -5 and 15 ~ are shown in these plots. Much
greater scatter is observed in the correlation between the
cv and the stiffening ratios for the storage and loss
modulus than the corresponding correlation with viscosity
stiffening ratios (figure 3). Some of this variability may
be the result of source effects identified in the
statistical analyses, but the variability in the test
results prohibited a more specific identification of the
variability. However, the general trend is the same:
O~
co
Free Aspholt,
20 40 60 BO 100 120 >
2 95 I * i , , 9 z ,. z 9 ~ 2.5 1011 i I I I I I I I ..... i J -u
-r
>
101~
X
I~ ~ ~'~TL -~ -. C
"-n
m
l -1 "o
m
-i1
k ",, -I o
:D
~>
Z
0
m
~" Aephal
t Eworo/Heukolom 10 "~
-0.5F. , , . , . , . , -0.5 102 l ' ' ' ' ' ' ' 7~ B'0
-0.2 0.0 0.2 0.4 0.6 0.6 t.O -10 '0 10 20 30 40 50 ..... 60 0 0
r'flective Filler Volume, C., X Temperoture, "C
FIG. 3--Relative viscosity of mastic versus FIG. 4--Complex modulus of mastic as a function
free asphalt and volume concentration of temperature, Dalton aggregate.
of filler.
10 I ' "1 I I 10 I I I --'
O
8 9O
UI 15"Cm~DO ~/~ 15 "C z
-i O m
D
"El 10
O O O o 0
O i
,U rTl
t~ O O --I
,~ O >
o8 3 r-"
o~ 0
4 o "c Z
O 9 _o
o n-
oJ C
,w
Z
.o
o
! .... I I, I
_w
I I I I 0 N
0 0.8 m
-0.2 0.0 0.2 0.4 0.6 0.8 -0.2 0,0 0.2 0.4 0.6
Filler Concentrotion. Volume 8osis. Cv, X Filler Concenlrolion. Volume Bosis, Cv,
0
m
FIG. 5--Relative storage modulus of mastic FIG. 6 - - R e l a t i v e loss modulus of m a s t i c versus
versus volume concentration, -5~ (23~ v o l u m e concentration, -5~ (23~ m
c~
and +15~ (59~ and +15~ (59~
..&
o~
increasing volumes of filler always result in increasing

storage and loss moduli. Statistical analyses indicated
that the stiffening ratios for both the loss and storage
moduli are statistically dependent upon the source of the
added baghouse dust, although this effect could not be
attributed to any of the measured physical properties of the
dust.
Tan delta was significantly decreased by the addition

of the mineral fillers, indicating that the fillers not only
stiffened the binder, but increased the relative magnitude
of the elastic component of the response. In figure 7 tan
delta is shown as a function of filler content for all seven
plants for data at 15 ~ Data for tan delta at -5 ~ were
not considered to be reliable owing to limitations in the
experimental technique and equipment.
Results and Analysis of Fracture Toughness Testinq of

AsPhalt Concrete Beams
The fracture toughness of hot-mix asphalt concrete

typically increases with temperature, reaches a maximum, and
then decreases as the temperature is increased. [13] At the
low temperatures the hot-mix behaves in a brittle manner and
J1c is synonymous with klc, the linear elastic fracture
toughness. As the temperature is increased the fracture
toughness increases, undergoing a transition from brittle to
brittle-ductile behavior. With increase in temperature, the
fracture toughness reaches a maximum, beyond which the
failure is primarily of a ductile nature and the resistance
to crack propagation is primarily related to the stiffness
of the asphalt. This transition from brittle to brittle-
ductile to ductile is illustrated in figures 7 and 8 where
the fracture toughness, JL~, is plotted versus temperature
for the seven aggregate sources, with and without the
addition of baghouse fines.
The critical region with respect to the value of J1c is

most likely the point of transition from brittle to ductile-
brittle behavior. [Ii] Below this transition the asphalt
cement, or mastic, behaves in a brittle fashion and the goal
of the materials/pavement engineer should be to control the
properties of the asphalt or mastic such that the service
temperature--loading rate are well above this transition
temperature. At temperatures above the point where Jlc
obtains an optimum, the strain to failure is very large and
the propagation of cracks is most likely the result of
plastic in the asphalt/mastic.
The fracture toughness is generally increased

significantly with the higher level of baghouse dust,
although the shape of the curves is not altered with the
addition of baghouse dust. The different behavior exhibited
A N D E R S O N ET AL. ON MINUS NO. 200-SIZED A G G R E G A T E S 171
Filler : 6
140 I I '1 I ! I
1.1
E 120
"u 100
"3
M
f~ 80
U
C:
r"
".d.
60
0
I-"
m
t.
40
:3
U
0
t., 20
b.
0 I I I I ! I
-30 -20 -10 0 10 20 30 40
Temperoture. ~
FIG. 7--Fracture toughness of mixtures with

no baghouse dust versus temperature.
Filler : 10 X
140 I I I I ' I I
E 120
% 100
m
80
I)
!-
ol 60
o
I-
ra
4.
40
:3
U
0
t.. 20
b.
0 I I I I I I
-3( -20 -10 0 10 20 30 40
Temperature, ~
FIG. 8--Fracture toughness of mixture with
4 percent baghouse dust versus
temperature.
by the Fairmount material is obvious in these two plots--the

peak in fracture toughness occurs at a temperature 6 to 8 ~
(i0 to 15 ~ lower than that seen for the other mixtures.
At the 6 percent total mineral filler level (no baghouse
fines), the fracture toughness values of the various
mixtures, other than the Fairmount mixes, are fairly close.
However, at the I0 percent level (4 percent baghouse dust),
the differences among the various material sources become
much larger. Note the relatively low fracture toughness
values observed for the Grayson and Norcross mixtures made
with 4 percent baghouse material. Apparently, baghouse dust
from these sources does not increase the fracture toughness
of the asphalt concrete mixes to the same extent as dust
from the other plants.
As a further comparison the temperature at which the

fracture toughness falls below 45 J / ~ (10 ib-in/in 3) was
calculated for the 6 percent and i0 percent mixtures, and is
shown in table 5. For the mixtures with no baghouse fines,
the transition temperature ranges from -8 t o -i ~ (18 to
30~ with an average of -4 ~ (25 ~ For the mixes with
4 percent baghouse dust, the transition temperature ranges
from -14 to -3 ~ (6 to 27 ~ with an average of -8 ~ (18
~ Thus, the use of the baghouse dust in general improves
the fracture toughness of the mixtures at low temperatures.
On the average, the transition temperature is lowered by
3.5~ (6.3 ~ when 4 percent baghouse fines are used, as
compared to the reference mixtures with no baghouse
material. However, for the G r a y s o n m a t e r i a l , the critical
temperature for the two mixtures is essentially identical,
again indicating that the Grayson baghouse material does not
improve the fracture toughness in the same way as baghouse
fines from the other plants.
In summary, the addition of the baghouse fines

increases the fracture toughness of the mixtures and lowers
the temperature at which a preselected value of fracture
toughness is attained. The particular value of the fracture
toughness depends upon the source of the aggregate as well
as the amount of filler in the mix--the effect of source is
as large as the effect of filler content, which ranged from
6 to I0 percent. Although source was a significant class
variable in the statistical analysis, no significant
relationships were identified between the g r a d a t i o n and
void-filling dry compacted bulk density or void content
properties of the filler and fracture toughness. Of the
mastic rheological properties, only the phase angle was
statistically significant when related to the fracture
toughness (see figure 9). Based on this very limited study
it has been shown that the properties of the filler can
affect the fracture toughness of hot-mix asphalt concrete.
A much larger study is required in order to fully define the
TABLE 5--Temperature at which elastic-plastic fracture

toughness, Jr,, equals 1 0 ib-in/in 3.
6I Minus No. 200 I l O l Minus No. 200 Difference

Source (OZ Baghouse)(P.ani (4z B a ~ o u s , ) O ~ . ~ )
Dalton 23 (3) 16 (2) 7 (2)
FairmonC 18 (1) 6 (1) 12 (1)
Grayson 28 (6) 27 (7) 1 (o
Kennesaw 25 (5) 20 Ca) 5 (5)
L l c h i a Spring 24 (4) 17 (3) 7 (2)
NoEcrol8 3O (7) 24 (6) 6 (4)
SCockbrldge* 22 (2) 23 (5) -1 (7)

(Estimated)
q
~2C) ........ ~ ........ ~ ........ ,~176176176176176 ...... --oo oo -o-o, oooooooo~o ooo-ooo
........I"........"I
9
.........,1-11
~
......9~.........lI .........9I........".~........
'
! ! : : ;
: ?."=
~c,,\ 9: i: 9.
Jo
.k
9 9 9 , o
IP .................. ""-/'"': ........ "'"~" " ........ " ........ T ........
9 9
30 ......... ~ ......... i- "'r'" . . . . . . . . . ~" . . . . . . -'. ...... x.t ........ ..:. . . . . . . . . .
!
0
0 .1 .2 .3 .4 3 .6 .7 .8
Tm Ddta
FIG. 9--Fracture toughness of mixtures versus tan delta of

mastics.
174 ASPHA~ MIXTURE PERFORMANCE
effect of mineral filler on both the fracture and the

fatigue properties of hot-mix asphalt.
FINDINGS AND C O N C L U S I O N S
The main purpose of this investigation was to determine

if the addition of baghouse fines affects the failure or
fracture properties of hot-mix asphalt concrete. Problems
were encountered with the fatigue test procedure and no
conclusions could be drawn with respect to fatigue.
However, with respect to fracture properties, on the basis
of the limited study and the specific material tested, the
following findings were established:
i. The empirical relationship between mastic viscosity

and volumetric filler concentration developed by Ewers
and Heukolom was verified and found to be valid,
although the free asphalt content afforded a better
correlation.
2. Gradation does not necessarily relate to

stiffening--the finest dust acted in much same manner
as the coarser dust.
3. The "beam on elastic foundation" method, in its

present form, is unsuitable for characterizing the
flexural fatigue behavior of hot-mix asphalt.
4. The fracture toughness, J1c, of the hot-mix tested

in this study appeared to be sensitive to the source of
the aggregate as well as to the amount of added
baghouse dust. In general, the addition of the dust
increased the fracture toughness of the hot-mix.
5. Physical properties of the filler fraction did not

explain the effect of variations in J,~ that were
observed. Only the tan delta of the mastic correlated
with the J1c data for the mixtures.
6. The resonant beam device was less than satisfactory

for characterizing the low-temperature rheology of the
asphalt-mineral filler mastics. This is the result of
problems associated with controlling thermal history
and mounting the beam in the test fixture. Other test
methods are more suitable.
7. A technique for measuring fracture toughness of

hot-mix asphalt concrete was demonstrated. The method
appears to be accurate and can also differentiate
between materials of different compositions.
8. Differences among mixes made using different

aggregate sources were as great as the differences
between mixes made with and without the addition of 6
percent baghouse dust. In other words, adding 6
percent baghouse dust to the mix gave similar or lesser
values of Jlc than switching aggregate sources.
It is concluded that the mineral filler fraction can

have a significant effect on the fracture properties of hot-
mix asphalt concrete. A comprehensive study is needed to
establish the significance of various mixture and filler
properties on the fracture behavior of asphalt concrete and
explain these effects in a rational manner using fundamental
material properties.
REFERENCES
[I] Anderson, D. A., Christensen, D. W., Dongre, R.,

Sharma, M. G., and Jordhal, P., Asphalt B~havior at Low
Service Temperature, Report No. FHWA-RD-88-078, U. S.
Department of Transportation, FHWA, March 1990.
[2] Bahia, H. U., and Anderson, D. A., "Isothermal Low-

Temperature Physical Hardening of Asphalt Binders,"
Proceedings, International Symposium for Chemistry o f
Bitumens, 5-8 June, Rome, Italy, 1991.
[3] Anderson, D. A., and Tarris, J. P., Adding Dust

Collector Fines to Asphalt Paving Mixtures, National
Cooperative Highway Research Program Report 252.
Washington, DC: Transportation Research Board, 1982.
[4] Rigden, P. J., and Lee, A. R., "The Brittle Fracture of

Tars and Bitumens," Journal of Applied Chemistry, Vol. 3, p.
62, 1953.
[5] Kandhal, P. S., "Evaluation of Baghouse Fines in

Bituminous Paving Mixtures," Proceedings of the Association
of Asphalt Paving Technolouists, Vol. 50, 1981, pp. 150-210.
[6] Anderson, D. A., Guidelines o~ the Use of Baghouse

Fines, National Asphalt Pavement Association, Information
Series 101-11/87, Riverdale, Maryland.
[7] Barksdale, R. D., "Practical Applications of Fatigue

and Rutting Tests on Bituminous Base Mixes," Proceedings of
the Association of Asphalt Paving Technologists, 1978, pp.
115-159.
[8] Sumpter, J. D. G., and Turner, C. E., "Method for

Laboratory Determination of J,=," Cracks and Fracture, A S T M
STP 601, American Society for Testing and Materials, 1976,

pp. 3-18.
[9] Heukelom, W., and abd Wijga, P. W., "Viscosity of
Dispersions as Governed by Concentration and Rate of Shear,"
Proceedings of the Association of Asphalt Pavinq
Technoloqists, Vol. 40, 1971, pp. 418-437.
[i0] Anderson, D. A., Bahia, H. U., and Dongre, R.,

"Rheological Properties of Mineral Filler-Asphalt Mastics
and Their Relationship to Pavement Performance," Effects of
Aqqregates and Mineral Fillers on Asphalt Mixture
Performance; ASTM STP I147, R. C. Meininger, ed., American
Society for Testing and Materials, Philadelphia, 1992.
[ii] Bahia, H. U., Anderson, D. A., and Christensen, D. W.,

"The Bending Beam Rheomter: A Simple Device for Measuring
Low Temperature Rheology of Asphalt Binders," Paper
Presented at the 1992 AAPT Meeting, March 1992, Charleston,
South Carolina.
[12] Anderson, D. A., Christensen, D. W., Bahia, H. U.,

"Physical Properties of Asphalt Cements and the Development
of Performance-Related Specifications," Paper presented at
the 1991 AAPT Meeting in Seattle, Washington, March 2-4,
1991.
[13] Dongre, R., Sharma, M. G., and Anderson, D. A.,
"Development of Fracture Criterion for Asphalt Mixes at Low
Temperatures," Transportation Research Record 1228,
Washington, D.C.: TRB, National Research Council, 1989, pp.
103-111.
Robert P. Chapuis I and Pierre-Paul L~gar~ 2
A SIMPLE METHOD FOR DETERMINING THE SURFACE AREA OF FINE AGGREGATES AND
F I L L E R S IN B I T U M I N O U S M I X T U R E S
R E F E R E N C E : Chapuis, R. P., and L~gar~, P.-P., "A S i m p l e M e t h o d f o r

D e t e r m i n i n g t h e S u r f a c e A r e a of F i n e A g g r e g a t e s a n d F i l l e r s in
B i t u m i n o u s M i x t u r e s , " Effects of Aqgreqates and Mineral Fillers on
Asphalt M i x t u r e Performance, ASTM STP 1147, R.C. Meininger, Ed.,
A m e r i c a n Society for Testing and Materials, Philadelphia, 1992.
ABSTRACT: In bituminous mixtures the finer fraction of the sand and the
filler have the highest surface area or specific surface. Surface
related p h y s i c o - c h e m i c a l properties are known to largely influence the
p e r f o r m a n c e of asphalt mixtures. Usual methods to determine the surface
area are approximate, often based on local experience with specific
crushed rock materials. They may yield poor estimates of the true
surface area. This paper proposes a simple m e t h o d which does not involve
an interpretation of the operator to visually determine some shape
factor. Only the gradation curves of the fine aggregate and the filler,
and their apparent specific gravity are considered in the p r o p o s e d
method, w h i c h does not use a shape factor or "fudge" factor. This m e t h o d
is d e m o n s t r a t e d with five fillers for which the surface area was
m e a s u r e d independently according to ASTM C 204.
KEYWoRDS: asphalt mixture, filler, surface area
A certain amount of filler is necessary in bituminous m i x t u r e s to

obtain the required density and strength. The filler particles fill a
p o r t i o n of the space between sand and gravel particles, and thus
contribute to increase density. The filler also influences the o p t i m u m
asphalt content in bituminous mixtures by increasing the surface area
(or specific surface) of mineral particles. And simultaneously, the
surface properties of the filler particles m o d i f y s i g n i f i c a n t l y the
rheological properties of asphalt such as penetration, ductility, and
also those of the m i x t u r e such as resistance to rutting. Several studies
(Kallas and Krieger [I]; Kallas and Puzinauskas [2]; Kallas et al. [3];
Holmes and Stewart [4]; Puzinauskas [5]; Eick and Shook [6]; Dukatz and
A n d e r s o n [7]; Ishai et al. IS]; Craus et al. [9); Langlois et al. [I0])
conclude that fillers largely influence the asphalt m i x t u r e performance,
depending mostly on their m i n e r a l o g y and finer size particles.
IProfessor, Department of Mineral Engineering, Ecole Polytechnique,

P.O.Box 6079, Sta.A, Montreal (Quebec), Canada, H3C 3A7.
2Professional Engineer, Public Works Laboratory, City of Montreal,
999 L o u v a i n St. East, Montreal (Quebec), Canada, H2M 1B3.
177
Copyright by
Copyright 9 ASTM
1992 Int'l
(all rights
by ASTM
International
07:29:10 EST 2011
Several tests have been used to c h a r a c t e r i z e the fillers surface

properties: standard sieving and hydrometer tests, special s e d i m e n t a t i o n
tests (Dukatz and A n d e r s o n [7]), shape and rugosity factors (Ishai et
al. [8]), Blaine fineness tests (Huet [ii]; for d e s c r i p t i o n of this test
d e v e l o p e d for Portland cement, see: Blaine [12]; Ober and F r e d e r i c k
[13]; Duriez and A r r a m b i d e [14]), m e t h y l e n e blue tests (Huet [ii];
L a n g l o i s et al. [I0]), and p l a s t i c i t y index (Langlois et al. [i0]).
The purpose of this paper is limited to a p r e s e n t a t i o n of a simple

m e t h o d to d e t e r m i n e the surface area of fillers and fine aggregates
w h i c h appear to be a key factor for asphalt m i x t u r e performance.
BACKGROUND
W h e n an asphalt m i x t u r e is designed, it is essential t o c h e c k

w h e t h e r the solid particles are adequately coated w i t h asphalt cement to
m a i n t a i n the required performance. This v e r i f i c a t i o n is m a i n l y based on
the surface area, Ss, e x p r e s s e d in square meter per k i l o g r a m of dry
solids (m2/kg). The filler fraction has the highest c o n t r i b u t i o n to the
surface area.
Usual m e t h o d s to determine S s are approximate, and o f t e n based on

e x p e r i e n c e w i t h local crushed rock materials. Well known a p p r o x i m a t e
m e t h o d s are those of H v e e m [15], Duriez and A r r a m b i d e [14]. In the
p r o v i n c e of Quebec, the method of standard B N Q - 2 3 0 0 - 9 0 0 is c u r r e n t l y
applied. All these methods use a constant surface area for the filler
particles, i n d e p e n d e n t l y of its m i n e r a l o g y and gradation.
To overcome the inaccuracy of these a p p r o x i m a t e methods, Craus and

Ishai [16] p r o p o s e d an analytical m e t h o d using data of actual size
distribution, fineness and shape p r o p e r t i e s of the filler particles.
Their m e t h o d gave calculated S s values very close to values of S,
o b t a i n e d e x p e r i m e n t a l l y in accordance with ASTM Test M e t h o d for Fineness
of P o r t l a n d Cement by Air P e r m e a b i l i t y A p p a r a t u s (C 204-63). Craus and
Ishai [16] also compared the Ss values p r e d i c t e d by their m e t h o d for
sands w i t h those p r e d i c t e d by either H v e e m [15], or Duriez and A r r a m b i d e
[14] methods. However, they have not used in their c o m p a r i s o n the two
other methods Duriez and A r r a m b i d e [14] have p r o p o s e d for the surface
area of fines.
In practice, the m e t h o d of Craus and Ishai [16] is not simple to

use. Furthermore, it involves a visually defined shape factor w h i c h m a y
be v i e w e d as a "fudge factor" to fit p r e d i c t e d values with m e a s u r e d
values of surface area.
In this paper, after a detailed c o n s i d e r a t i o n of five a v a i l a b l e

methods, a r e l a t i v e l y simple method is p r o p o s e d and demonstrated.
METHODS FOR SANDS
The five following m e t h o d s are considered hereafter for a sand:

(i) H v e e m [15], (2) B N Q - 2 3 0 0 - 9 0 0 (Quebec), (3) Duriez and A r r a m b i d e
[14], (4) Craus and Ishai [16], and (5) the p r o p o s e d method. In the case
of a clean sand for which the g r a d a t i o n of fines is unknown, the
proposed method will be combined with an equation of Duriez and
A r r a m b i d e [14] for surface area of a fine filler.
CHAPUIS AND LEGARE ON BITUMINOUS MIXTURES 179
H v e e m [15] e s t i m a t e s S 3 in a s p h a l t m i x t u r e s w i t h s u r f a c e a r e a
f a c t o r s w h i c h are r e l a t e d to a g i v e n sieve size, a n d are to be
m u l t i p l i e d by the t o t a l p e r c e n t a g e of a g g r e g a t e p a s s i n g t h a t sieve. The
g e n e r a l f o r m u l a is:
Ss = ~ Sa~No. PNo in k g / m 2 (i)
w h e r e SaLNo" is t h e s u r f a c e a r e a f a c t o r for s i e v e n u m b e r No., and PNo is

t h e p e r c e n t a g e p a s s i n g by w e i g h t t h a t sieve. H v e e m [15] d e f i n e d v a l u e s
of Saf.No for s i e v e s No. 4 - 8 - 1 6 - 3 0 - 5 0 - 1 0 0 - 2 0 0 as b e i n g 0 . 4 1 - 0 . 8 2 - 1 . 6 4 -
2.87-6.14-12.29-32.77 m2/kg r e s p e c t i v e l y . T h e s u r f a c e a r e a of H v e e m [15]
for t h e fines (32.77 m2/kg) is v e r y s i m i l a r to that of 40 k g / m 2 as
p r o p o s e d by P i n g o n [17, 18].
The s t a n d a r d B N Q - 2 3 0 0 - 9 0 0 (Quebec) m a k e s u s e of a m e t h o d v e r y
s i m i l a r to t h a t of H v e e m [15], e x c e p t for t h e m e t r i c size of sieves. The
v a l u e s of t h e s u r f a c e area f a c t o r for s i e v e s 5 - 2 . 5 - 1 . 2 5 - 0 . 6 3 0 - 0 . 3 1 5 -
0 . 1 6 0 - 0 . 0 8 0 m m are 0 . 4 1 - 0 . 8 2 - 1 . 6 4 - 2 . 8 7 - 6 . 1 4 - 1 2 . 2 9 - 3 2 . 7 7 m2/kg
respectively.
Duriez and A r r a m b i d e [14, 19] proposed to d e t e r m i n e the surface

a r e a with:
S s = 135 A + 12 B + 2.3 C in k g / m 2 (2)
w h e r e A, B and C are t h e p e r c e n t a g e s by w e i g h t of t h e f r a c t i o n s finer

t h a n 80 pm, in t h e r a n g e 80 ~ m - 0.315 mm, and in t h e r a n g e 0 . 3 1 5 - 5.0
mm respectively. T h i s e q u a t i o n (2) is p r e s e n t l y u s e d in F r a n c e (CETE de
L y o n [20]). The s u r f a c e a r e a of D u r i e z and A r r a m b i d e [14] for t h e fines
(135 m2/kg) is s i g n i f i c a n t l y l a r g e r t h a n t h a t of 32.77 m2/kg as p r o p o s e d
by H v e e m [15]. D u r i e z and A r r a m b i d e ([14], t o m e 1, p.138) h a v e d e r i v e d
t h e i r h i g h v a l u e as an a v e r a g e for d i f f e r e n t m i n e r a l fillers. In t h e i r
c o m p a r i s o n , C r a u s and Ishai [16] h a v e c o n s i d e r e d t h i s eq. 2 only.
H o w e v e r , for e s t i m a t i n g the s u r f a c e a r e a of fillers, D u r i e z a n d
Arrambide ([14], t o m e I, p.288) h a v e p r o p o s e d t w o o t h e r e q u a t i o n s . T h e i r
first e q u a t i o n is said to a p p l y to v e r y fine fillers:
S' s (fines < 80 ~m) = 60 + 12 f,0 in k g / m 2 (3)
w h e r e fl0 is t h e p e r c e n t a g e by w e i g h t of the f i l l e r s m a l l e r t h a n i0 gm.

In t h e c a s e of a fine filler, D u r i e z and A r r a m b i d e [14] h a v e p r o p o s e d :
S" s (fines < 80 ~m) = 38 + 5.6 fJ5 in k g / m 2 (4)
where fl5 is t h e p e r c e n t a g e by w e i g h t of t h e filler smaller than 15 gm.
C r a u s and Ishai [16] h a v e p r o p o s e d a m e t h o d b a s e d on m a t h e m a t i c a l

i n t e g r a l f u n c t i o n s a l o n g t h e g r a d a t i o n curve. M o r e c a l c u l a t i o n is
r e q u i r e d t h a n in t h e p r e v i o u s m e t h o d s . T h i s m a y be a r e a s o n w h y it has
not b e e n c o n s i d e r e d in t h e u s u a l m e t h o d s or s t a n d a r d s of r e g u l a t o r y
a g e n c i e s , a l t h o u g h it seems to p r o v i d e a g o o d e v a l u a t i o n of t h e s u r f a c e
area. In our c o m p a r i s o n for a sand, p r e s e n t e d below, t h e e x a m p l e of sand
g r a d a t i o n p r o v i d e d by C r a u s and Ishai [16] w a s r e t a i n e d w i t h t h e i r o w n
c a l c u l a t i o n of s u r f a c e area.
The p r o p o s e d m e t h o d is d e r i v e d as follows. S i m p l e g e o m e t r i c a l
c o n s i d e r a t i o n s m a y be used to e s t i m a t e t h e s u r f a c e a r e a (or s p e c i f i c
surface) S, of a g g r e g a t e s . If d is the d i a m e t e r of a s p h e r e or t h e e d g e
l e n g t h of a cube, t h e s u r f a c e a r e a of a c o l l e c t i o n of s u c h s p h e r e s or
cubes is:
S s (d) = 6 / d ps in k g / m 2 (5)
w h e r e Ps is t h e d e n s i t y (kg/m 3) of s p h e r e s or cubes. M a n y t h e o r e t i c a l
d e v e l o p m e n t s h a v e b e e n p r o p o s e d to b e t t e r d e f i n e t h e Ss v a l u e of
i r r e g u l a r p a r t i c l e s as a f u n c t i o n of s h a p e factors, r u g o s i t y factors,
p r o j e c t i o n f a c t o r s ( D a l l a v a l e [21]; O r r a n d D a l l a v a l e [22]; G r e g g and
S i n g [23]). T h e p r e d i c t i o n s of t h e s e t h e o r e t i c a l d e v e l o p m e n t s h a v e b e e n
c o m p a r e d w i t h e i t h e r d i r e c t or i n d i r e c t d e t e r m i n a t i o n s of t h e s u r f a c e
area. S u c h p r e d i c t i o n s are n e v e r f u l l y s a t i s f a c t o r y , and it s e e m s t h a t
eq. 5 g i v e s t h e b e s t e s t i m a t e of S,. In t h e c a s e of a g g r e g a t e s a n d
f i l l e r s for a s p h a l t m i x t u r e s , it is p r o p o s e d h e r e to a p p l y t h e
t h e o r e t i c a l eq. 5 to a real g r a d a t i o n c u r v e as follows:
Ss = (6/Ps) ~ [(PNoD - PNod) / d] in k g / m 2 (6)
w h e r e (PNoD - PNod) is the p e r c e n t a g e by w e i g h t s m a l l e r t h a n s i z e D and

l a r g e r t h a n next size d. Eq. 6, in t h e c a s e of a c l e a n sand, w i l l be
c o m b i n e d w i t h eq. 4 for a fine filler. E q . 6 is s i m i l a r to a n o t h e r
e q u a t i o n p r o p o s e d by D u r i e z and A r r a m b i d e [14] w h e r e (D+d)/2 is u s e d
i n s t e a d of d.
T h e five m e t h o d s e x a m i n e d h e r e a b o v e h a v e b e e n u s e d to p r e d i c t t h e
s u r f a c e a r e a of t h e c l e a n sand d e s c r i b e d by C r a u s a n d I s h a i [16] in
t h e i r T a b l e i. T h e a p p a r e n t s p e c i f i c g r a v i t y is 2.65 ( d e n s i t y ps = 2 650
kg/m~). T h e s a n d g r a d a t i o n is g i v e n in o u r T a b l e 1 below, w h i c h
i l l u s t r a t e s t h e p r o p o s e d m e t h o d for t h i s sand.
TABLE 1 -- S u r f a c e a r e a for a c l e a n s a n d as p r e d i c t e d by
the proposed method.
Sieve opening Cumulative Difference X S = 6/dps Product XS

(mm) p a s s i n g (%) (PNoD - PNod) m2/kg m2/kg
4.76 i00 . . . . . . . . . . . . .
2.38 95 0.05 0.95 0.048
1.19 87 0.08 1.90 0.152
0.59 62 0.25 3.84 0.959
0.297 22 0.40 7.62 3.049
0.149 8 0.14 15.2 2.127
0.105 4 0.04 21.56 0.863
0.074 2 0.02 30.60 0.612
. . . . . . . . 0.02 38 + 5.6 fl5 0.760
Notes: i. T h e X v a l u e for p a r t i c l e s s m a l l e r t h a n 74 ~ m is 0.02,

i n d e p e n d e n t l y of t h e p e r c e n t a g e s m a l l e r t h a n 40 ~ m g i v e n by
C r a u s a n d I s h a i [16].
2. T h e f1~ v a l u e is a s s u m e d to be null for t h i s c l e a n sand.
3. T h e S, v a l u e is o b t a i n e d by s u m m a t i o n of XS v a l u e s as 8.57
m2/kg.
The five p r e d i c t i o n s of the S, value are:
(i) H v e e m [15]: 7.39 m2/kg (Craus and Ishai [16]);

(2) B N Q - 2 3 0 0 - 9 0 0 (Quebec): 7.39 m2/kg (equivalent to H v e e m [15]);
(3) Duriez and A r r a m b i d e [14]: 6.98 m2/kg (Craus and Ishai [16]);
(4) Craus and Ishai [16]: 8.34 m2/kg;
(5) the p r o p o s e d method: 8.57 m2/kg as c a l c u l a t e d w i t h Table I.
It appears that the three methods u s i n g a constant surface area

for the filler p a r t i c l e s (independently of m i n e r a l o g y and gradation),
u n d e r e s t i m a t e the surface area of this clean sand. However, the simple
m e t h o d p r o p o s e d in this paper gives a value very close to that o b t a i n e d
w i t h the more accurate method of Craus and Ishai [16].
This justifies the use of the p r o p o s e d m e t h o d for sands. N o w this

m e t h o d will be extended to be applicable also to fillers. Then it will
be d e m o n s t r a t e d w i t h five fillers for w h i c h the surface area was
m e a s u r e d i n d e p e n d e n t l y according to ASTM C 204.
METHODS FOR FILLERS
The previous example showed substantial d i s c r e p a n c i e s b e t w e e n the

p r e d i c t i o n s of the five different methods in the case of a clean sand.
In recent asphalt mixtures, the p e r c e n t a g e of filler may reach 20%, thus
creating larger d i s c r e p a n c i e s b e t w e e n predictions. In the m e t h o d s of
H v e e m [15], as followed by BNQ-2300-900, the value of S, is 32.77 m2/kg
for any filler. In the m e t h o d of equation 2 (Duriez and A r r a m b i d e [14]),
S, = 135 m2/kg for any filler. In the m e t h o d of Craus and Ishai [16], the
S, value is the result of m u l t i p l y i n g a first value, w h i c h depends only
on the g r a d a t i o n curve, with a v i s u a l l y d e f i n e d shape factor b e t w e e n 1
and 2. Thus, the e s t i m a t e d value of S, depends on the personal
a p p r e c i a t i o n of the operator e x a m i n i n g p a r t i c l e s t h r o u g h a microscope.
Craus and Ishai [16] have calculated the surface area of five
fillers by their m e t h o d and compared their results with direct determin-
ation a c c o r d i n g to A S T M C 204-63. These 5 fillers are e x a m i n e d here for
c o m p a r i s o n w i t h the proposed method, and with e q u a t i o n s 3 and 4 of
Duriez and A r r a m b i d e [14].
To be acceptable in asphalt mixtures, the fillers must have a low

p l a s t i c i t y index (Langlois et al. [i0]) w h i c h means that the clay size
p a r t i c l e s in the fillers must be e s s e n t i a l l y inactive rock dust.
Consequently, the surface area of fillers should be p r e d i c t a b l e from the
g r a d a t i o n curve only. It is p r o p o s e d here to extend the use of e q u a t i o n
6 to p a r t i c l e s smaller than 80 ~m and down to the smaller size as g i v e n
by the h y d r o m e t e r analysis. The results o b t a i n e d by this m e t h o d are
shown in Tables 2 to 6 for the five fillers of Craus and Ishai [16].
TABLE 2 -- Surface area for a limestone filler (p, = 2 8 8 0 kg/m3).
Sieve opening Cumulative Difference X S = 6/dp s Product XS

(mm) passing (%) (PNoD - PNod) m2/kg m2/kg
0 074 I00 ....

0 060 94 0.06 34.72 2.08
0 050 89 0.06 41.67 2.08
0 040 83 0.06 52.08 3.13
0 030 76 0.07 69.44 4.86
0 020 65 0.ii 104.17 11.46
0 010 45 0.20 208.33 41.67
0.005 24 0.21 416.67 87.50
eq. d = 0 . 0 0 2 9 .... 0.24 718.39 172.41
Notes: i. T h e S s v a l u e is o b t a i n e d b y s u m m a t i o n o f X S v a l u e s a s 3 2 5 . 2
m2/kg.
2. T h e m e t h o d t o d e t e r m i n e t h e e q u i v a l e n t d is g i v e n b e l o w .
TABLE 3 -- Surface area for a hydrated lime (p, = 2 3 0 0 kg/m~).
Sieve opening Cumulative Difference X S = 6/dPs Product XS

(ram) passing (%) (PNoD - PNod) m2/kg m2/kg
0.074 i00 .....

0.060 99.5 0.005 43.48 0.22
0.050 99 0.005 52.17 0.26
0.040 98.5 0.005 65.22 0.33
0.030 98 0.005 86.96 0.43
0.020 97.5 0.005 130.43 0.65
0.010 87 0.105 260.87 27.39
0.005 35 0.52 521.74 271.30
eq. d = 0 . 0 0 2 9 .... 0.35 899.55 314.84
Note: The S s value is o b t a i n e d by summation of XS values as 615.4 m2/kg.
TABLE 4 -- Surface area for glass beads (p, = 2 3 0 0 kg/m3).
Sieve opening Cumulative Difference X S = 6/dp, Product XS

(mm) passing (%) (PNoD - PNod) m2/kg m2/kg
0.074 i00 . . . . . . . . . . . . . .
0 060 70 0.30 43.48 13.04
0 050 48 0.22 52.17 11.48
0 040 27 0.21 65.22 13.70
0 030 12 0.15 86.96 13.04
0 020 7 0.05 130.43 6.52
0 010 5 0.02 260.87 5.22
0 005 2.4 0.026 521.74 13.57
eq. d = 0 . 0 0 2 9 ..... 0.024 899.55 21.59
Note: The S s value is o b t a i n e d by summation of XS values as 98.2 m2/kg.
TABLE 5 -- S u r f a c e area for a d o l o m i t e (Ps = 2 8 3 0 kg/m3).

0.074 i00 . . . . . . . .
0.060 84 0.16 35.34 5.65
0.050 73 0.ii 42.40 4.66
0.040 58 0.15 53.00 7.95
0.030 47 0.ii 70.67 7.77
0.020 35 0.12 106.01 12.72
0.010 22 0.13 212.01 27.56
0.005 15 0.07 424.03 29.68
eq. d = 0 . 0 0 2 9 --- 0.15 731.08 109.66
Note: The Ss value is o b t a i n e d by s u m m a t i o n of XS v a l u e s as 2 0 5 . 7 m2/kg.
TABLE 6 -- S u r f a c e area for a b a s a l t (p, = 2 8 8 0 kg/m3).

0.074 I00
0.060 87 0 13 34 72 4.51
0.050 78 0 09 41 67 3.75
0.040 66 0 12 52 08 6.25
0.030 55 0 ii 69 44 7.64
0.020 43 0 12 104 17 12.50
0.010 30 0 13 208 33 27.08
0.005 20 0 i0 416.67 41.67
eq. d = 0 . 0 0 2 9 .... 0.20 718.39 143.68
Note: T h e S, v a l u e is o b t a i n e d by s u m m a t i o n of XS v a l u e s as 2 4 7 . 1 m2/kg.
T h e g r a d a t i o n s p r o v i d e d b y C r a u s and Ishai [16) a r e l i m i t e d to 5

@m. F o r the p a r t i c l e s sizes s m a l l e r t h a n 5 ~m, an e q u i v a l e n t s i z e m u s t
be d e f i n e d for our m e t h o d . As t h e u s u a l s i z e is d e f i n e d by s i e v i n g , t h e
e q u i v a l e n t s i z e c o r r e s p o n d s to an a v e r a g e v a l u e b a s e d on t h e a p p a r e n t
s u r f a c e w h i c h is p r o p o r t i o n a l to d 2. Then, t h e e q u a t i o n g i v i n g t h e
e q u i v a l e n t d i a m e t e r is:
(equivalent d) 2 = 1~ in y2dy : D2/3 (7)
When D = 5 @m, the equivalent s i z e d is 2.9 ~m, as u s e d in T a b l e s

2 to 6.
The various predictions for t h e five fillers appear in T a b l e 7.
TABLE 7 -- S u m m a r y of p r e d i c t i o n s for surface areas of fillers.
V A L U E S OF S U R F A C E A R E A (m2/kg)
T y p e of Measured Craus-Ishai This method Eq. 3 Eq. 4
Filler ASTM C 204-63 [16] (1991) [14] [14]
Limestone 263 258 325 600 346

H y d r a t e d lime 869 750 615 1104 553
Glass beads 86 78 98 120 72
Dolomite 202 183 206 324 195
Basalt 247 217 247 420 240
DISCUSSION AND CONCLUSION
In T a b l e 7, it a p p e a r s t h a t t h e p r o p o s e d m e t h o d g i v e s a fair
e s t i m a t e of t h e s u r f a c e a r e a of fillers, w i t h m u c h s i m p l e r c a l c u l a t i o n s
t h a n t h e m e t h o d of C r a u s and Ishai [16], and w i t h o u t u s i n g a n y v i s u a l l y
d e f i n e d shape f a c t o r as in t h e i r method. T h e q u a l i t y of t h e p r e d i c t i o n
m a y be v i z u a l i s e d in F i g u r e i.
S i m i l a r l y , t h e e q u a t i o n 4 as p r o p o s e d by D u r i e z and A r r a m b i d e [14]
for s o - c a l l e d fine fillers, g i v e s fair e s t i m a t e s of t h e s u r f a c e area,
e x c e p t for t h e h y d r a t e d lime. T h e i r Eq. 3, however, o v e r e s t i m a t e s t h e
s u r f a c e a r e a of the five "fillers".
It is c l e a r from t h e d a t a in T a b l e s 2 to 6 t h a t t h e s m a l l e r size
p a r t i c l e s in t h e f i l l e r h a v e t h e g r e a t e s t i n f l u e n c e on t h e s u r f a c e area.
The e q u a t i o n s 3 a n d 4 of D u r i e z and A r r a m b i d e [14] t r y to t a k e s u c h
i n f l u e n c e into account. In fact, for t h e five f i l l e r s t h e s u r f a c e a r e a
is f o u n d to be u s u a l l y w e l l c o r r e l a t e d to fs, t h e p e r c e n t a g e by w e i g h t
s m a l l e r t h a n 5 @m, as shown in F i g u r e 2.
I 1 I I I | /~[ I I
600 _ ~,o~/./// HL 9
% ,~ (exception)
< - o~,o ~
UJ
< 400
o
<
ii
~0 2OO
a
uJ
0
- GB
t~ 1GB I I I , I I I I
UJ 200 400 600 800 I000
m o
n
MEASURED SURFACE AREA (m 2 / kg)
FIG. 1 -- C o r r e l a t i o n between predicted and m e a s u r e d surface areas.
! i 1 i i i 9
HL
8oo
600
<
UJ
rr
< 4O0
L
D B
~D 200
I i I i I I 1
o lO 20 30 40
P E R C E N T A G E BY W E I G H T < 5jum
FIG. 2 -- Correlation between surface area S, and fs.
Both figures indicate an excellent agreement, except for the

hydrated lime which may present problems of flocculation during a
hydrometer test. Then, it may be concluded that the proposed method has
a good potential to determine the surface area of fillers. It has also
the advantage of using only gradation parameters, without any
interpretation of the operator in relation to some visually determined
shape factor.
REFERENCES
[i] Kallas, B.F., and Krieger, H.C., 1961, "Effects of consistency of

asphalt cements and types of mineral filler on the compaction of
asphalt concrete", Proceedings Association of Asphalt Paving
Technologists, Vol. 29, pp. 152-172.
[2] Kallas, B.F., and puzinauskas, V.P., 1961, "A study of mineral
fillers in asphalt paving mixtures", Proceedings Association of
Asphalt Paving Technologists, Vol. 31, pp. 493-525.
[3] Kallas, B.F., Puzinauskas, V.P., and Krieger, H.C., 1962, "Mineral
fillers in asphalt paving mixtures", Highway Research Board,
Bulletin 329, Washington, D.C.
[4] Holmes, C.D., and Stewart, W.D., 1968, "The effect of mineral
filler on the viscosity of asphalt cement", Dept. of Civil Engng,
Owens University, OH, Report No. 44, 55 p.
[5] Puzinauskas, V.P., 1969, "Filler in asphalt mixture", The Asphalt
Institute, College Park, Md., Research Report 69-2, 31 p.
[6] Eick, J.H., and Shook, J.F., 1978, "The effects of baghouse fines
on asphalt mixtures", The Asphalt Institute, college Park Md,
Research Rept 78-3, 44 p~
[7] Dukatz, E.L., and Anderson, A.D., 1980, "The effects of various
fillers on the mechanical properties of asphalt and asphaltic
concrete", Proceedings Association of Asphalt Paving
[8] Ishai, I., Craus, J., and Sides, A., 1980, "A model for relating
filler properties to optimal behavior of bituminous mixtures",
P r o c e e d i n g s Association of Asphalt Paving Technologists, Vol. 49,
pp. 416-437.
[9] Craus, J., Ishai, I., and Sides, A., 1981, "Durability of
bituminous paving mixtures as related to filler type and
properties", P r o c e e d i n g s A s s o c i a t i o n of Asphalt P a v i n g
[i0] Langlois, R., Moreau, G., Tremblay, G., and Moreux, J.-C., 1991,
"Caract~risation des fines dans les enrob~s bitumineux",
P r o c e e d i n g s of the 26th AQTR Annual Meeting, Quebec, 1991, pp. 3-
16.
[ii] Huet, M., 1987, "~largissement des possibilit~s d'utilisation des
sables dans les enrob~s", Laboratoire Central des Ponts et
Chauss~es, Paris, Taxe parafiscale, 122, 18p.
[12] Blaine, R.L., 1941, "Studies of the measurement of specific
surface by air permeability", Bulletin of the American S o c i e t y for
T e s t i n g and Materials, Philadelphia, PA, Vol. 108, pp. 17-20.
[13] Ober, S.S., and Frederick, K.J., 1959, "A study of the Blaine
fineness tester and a determination of surface area from air
permeabilty data", Symposium on Particle Size Measurement, ASTM
STP 234, American Society for Testing and Materials, Philadelphia,
PA, pp. 279-287.
[14] Duriez, M., and Arrambide, J., 1962, Nouveau trait~ de m a t ~ r i a u x
de construction, Tomes I, 2 and 3, Dunod, Paris.
[15] Hveem, F.N., 1974. "Mix design method for asphalt concrete", MS-2,
The Asphalt Institute, College Park, Md.
[16] Craus, J., and Ishai, I., 1977, "A method for the determination of
the surface area of fine aggregate in bituminous mixtures",
Journal of Testing and Evaluation, Vol. 5(4), pp. 284-291.
[17] Pingon, A., 1948a, "Etude de la granulom~trie des sables, des
gravillons et des b~tons, partie A", Travaux, ao~t 1948, Paris,
pp. 469-472.
[18] Pingon, A., 1948b, "Etude de la granulom~trie des sabies, des
gravillons et des b~tons, parties B ~ D", Travaux, octobre 1948,
Paris, pp. 553-558.
[19] Duriez, M., and Arrambide, J., 1954, Liants h y d r o c a r b o n 4 s :
m o r t i e r s et b4tons bitumineux, Dunod, Paris.
[20] CETE de Lyon, 1984, M e m e n t o of French S p e c i f i c a t i o n s - P a v e m e n t s
(in French), SETRA et LCPC, Paris, 68p.
[21] Dallavale, J.M., 1948, "Micromeritics - The technology of fine
particles", 2nd edition, Pitman, New-York.
[22] Orr, C., and Dallavale, J.M., 1959, Fine p a r t i c l e m e a s u r e m e n t
size, surface and pore volume, MacMillan, New-York.
[23] Gregg, S.J., and Sing, K.W.S., 1967, Adsorption, Surface Area and
Porosity, Academic Press, London and New York.
1 2
Mukhlis A. Shahrour, and Bassam G. Saloukeh
EFFECT OF QUALITY A N D Q [ I K N T I T Y O F ~ Y ~ FTIJ.I~R (PASSIM8

SIEVE NO. 200) ON A S P H A L T I C ~ I N DtIBAI.
: Shahrour, A. M., and Saloukeh, G. B., "Effect of Quality

and Quantity of Locally Produced Filler (Passing SieveNo. 200) on
Asphaltic Mixtures in Duhai", Effects of Aggregates and Mineral
Fillers o_n_n Asphalt Mixture Performance, ASTM STP 1147, Richard C.
Meininger, Editor, American Society for Testing and Materials,
Philadelphia, 1992.
ABSTRACT : Failures commonly observed in the Emirate of Dubai

roads are rutting and cracking. The objective of this study is to
evaluate the influence of different fillers extracted from the
different local aggregate sources on the performance of asphaltic
mixtures. In addition, cement, hydrated lime, and a sodium
silicate based filler (Bagrat KS-300) were also considered in
the study.
A total of ten types of fillers were tested in the laboratory for

grading, particle shape, surface texture, voids when coapacted,
absorption and specific gravity.
The quality of each type of filler was studied by adding different

fillers to bitmnen at two filler to bitumen ratios ( 0.5 and 1.5
by weight) where characteristics of the bitumen cement properties
(penetration, softening point, kinematic viscosity at 135 degrees
celsius, and penetration index "P.I.") were studied.
I
Head of Research and Investigation Unit and Research
Scientist, respectively, Construction Quality Control and Research
Section, Dubai Municipality, Box 67, Dubai, 0.A.E.
2
Chemist, Construction Quality Control and Research Section,
Dubai Municipality, Box 67, Dubai, U.A.E.
187
Copyright 9 1992 by ASTM International www.astm.org
A Marshall mix design study was m d e u s l n g a l l types of fillers and

different ratios to evaluate the performance of different types and
filler quantities in the asphalt mixture .
: filler types, f i l l e r , filler ratio, filler-bitumen,

stiffening ratio, bitumen extender, bitumen modifier, filler-bitumen
system.
~ O N
Two types of asphaltic mixtures are currently used in Dubei

Asphaltic Concrete (AC) and Dense Bitumen Mac~lam (DBM).
The most c o m ~ n mode of failure in Duhei roads is plastic flow

leading to ruts in the wheel tracks or the sideways flow of the
material on bends, roundabouts and junctions. The rutting mode of
failure is generally associated with roads of asphaltic concrete
mixtures. While fatigue cracking m cd e of failure is mostly associated
with roads of dense bitumen mecadammixtures.
A combination of the relentless heat (ambient air temperature

for nearly 5months of the year can reach up to 50 degrees celsius
and pavement surface temperature can reach up to 85 decjrees
celsius), virtually uncontrolled axle loading (single axle loads can
reach up to 27000 kg, tandem axle loads of 54000 k g ~ e r e recorded),
and high tyre pressure (was measured for trucks and recordedas high
2
as 10.55 kglcm ), lead to severe load conditions on IXt~i
pavements.
The quality and quantity of filler used in

hot asphaltic mixtures ffreatly affect its performance.
The function of mineral filler has been recognized to be more
than filling voids.
Fillers usually stiffen asphalt cements variably

depending on the filler type and quantity. It has been
reported that certain field failures were related to the poor
quality of filler used [!].
Filler in asphalt mixtures m y act as an asphalt extender or it

may stiffen the binder depending on the type of filler used. The
fillers which extend the asphalt-cement effectively are increasing the
asphalt-cement voltme in the mix .
Stiffening of mixes caused by certaln types of fillers Is not

easy to demonstrate. However, stiffening becomes quite apparent
during laboratory mixirg. Some fillers do stiffen laboratory mlxes to
the point that a great effort is required to mix them physically.
SHAHROUR AND SALOUKEH ON PASSING SIEVE NO. 200 189
Stiffening in filler-bitmnen mixtures can be measured by a

decrease in penetration, an increase in softening point or an increase
in viscosity . Fillers from different sources will produce quite
different levels of stiffening. The difference in stiffening caused
by different types of fillers cannot be readily explained by filler
size, gradation or particle shape ; nor is the stiffening always
reflected in the Marshall properties of hot asphalt mixtures.
This research was undertaken to achieve the fol lowing

objectives :
1. Characterize various generic type of fi Ilers produced

locally by their physical and chemical properties.
2. Determine the effect of filler mterlal on the physical

properties of bitumen.
3. Determine the effect of quality and quantity of fillers on

the physical properties of selected asphalt mixtures.
4. Develop a suitable specification to ensure qualitative

and quantitative control of fillers to be incorporated in the
bituminous mixtures used in hot climates.
FIW.T.~ ~ Z A T I O N
Six mineral fillers (Gabbro origin) and one limestone filler

were sampled as per BS 34061part I, from different crushers. These
represent different generic types depending on the aggregate source
being processed. Additionally, three manufactured fillers were
considered in the study (namely : Ordinary Portland Cerent (OPC),
Hydrated Lime, and Sodium Silicate "Bagrat KS-300").
PiWSIC~L ~ E S
Only the minus No. 200 sieve (0.075mm) fraction of fillers was
evaluated in this study. The grading of each filler was measured with
a test that is cci,.,only used to determine the gradation of fine soil
(ASTM D422). In this test, the filler is allowed to settle in
solution and the rate of settling is related to particle size. The
rate of settling is measured with a hydrometer and hence the test
is often referred to as a hydrometer analysis test. The
gradation of each filler is given in Table I and typical gradation
curves for some fillers, and recxam,ended gradation limits for filler
TABLE 1--Properties o__ff f i l l e r s .

...............................................................................................
I I Filler Type
Seria!l Test Type i...................................................................
I I LS G[ ~A ~ WF ?A SI KS CE RL
...... I................. :.......................................................................
i Relative Density 2.71 2.89 2.76 2.9! 2.94 Z.7B 2.78 2.93 3.15 2.27
(BS Bi2 Part Z)
2 Specific Surface 332 231 165 284 164 167 180 [900 327 '**
(ASTM C-204)
2
(M Ikg)
3 Voids in Dry 0,37 0.43 0.44 0.42 0.40 0.43 0.44 ** 0.42 **
Compacted Filler
IBm - 8[2)
4 Chemical Analysis
{ASTM c-ll4)
% Si0Z 1,26 48.9[ 39.60 49.80 46.9! 46,Z7 45~6Z 98.0 Z0,48 3.74
% Ca0 52,71 14.94 32,33 16.91 !5.73 5.24 6.75 0.0 63.82 67.35
%MgO 0.68 20,32 12.49 [4,75 [5,92 43.43 40.41 0.0 1,6 1.40
%R203 45.35 15,83 [5.58 !8.54 21.44 5.06 7.22 2.0 14.[0 27.51
5 Hydrometer
Analysis
(BS-1377!
75 MIC lO0 100 lO0 100 i00 iO0 lO0 * * *
50 M[C 93 83 85 89 90 85 80 * * *
30 MIC 86 77 80 85 85 79 58 ~ * '
20 M[C 73 65 60 78 51 6i 45 ' * *
i0 ~IC 5! Z7 29 40 Z9 3~ Z8 * * *
05 MIC 32 [7 i4 [8 15 18 20 * * *
03 MIC 20 12 12 12 !3 13 !5 * * *
O[ MIC 8 i[ II 11 12 11 12 * * *
6 Fineness Indicator 3.4 4,1 4.1 3.7 4,1 4.0 4.4 * * *
IASTM D-422}
7 Liquid Limit 25 32 32 3Z 30 30 31 * * *
{BS-1377}
B Plastic Limit 18 Nil Nil Nil Nil Nil Nil * * *
(BS-1377)
9 Plasticity Index 7 NP NP NP NP NP NP * * *
IBS-13771
10 pH Value t3 8.4 8.5 8.5 8,8 9.1 8.7 6.3 12.5 12.3
(BS-[377)
...........................................................................................
LS : Lime Stone, WF : Wadi Filly, KS : Hagret, * : can not be determined as per HS-1377
GI : Gibca , ?h : Tawoon, CE : O.P.Cement, **: can not be determined as per BS-BI2
Mk : Manama, SI : Siji , KA : Kadra , ***: can not be determined as per ASTM C204
HL : Hydrated Lime,
by the U.S. Army Waterways Experiment Station [4], based on i00

percent of the material passing a No. 200 sieve (0.075 micron) are
shown on graphs given in FiGures I, 2, 3 and 4.
In this study several calculated values were used to

characterize the gradation of fillers;
Fineness Indicator, U n i f o r m i t y C o - E f f i c i e n t (Cu), and Co-

Efficient of C u r v a t u r e (Cc). The f i n e n e s s i n d i c a t o r i s c a l c u l a t e d
by d i v i d i n g by 100 t h e sum o f t h e p e r c e n t a q e o f f i l l e r r e t a i n e d on a
s t a n d a r d s e r i e s of s i e v e s (75, 50, 30, 20, 10, 5, 3 and 1 micron)
[ ! ] . The f i n e r t h e f i l l e r i s , t h e s m a l l e r t h e f i n e n e s s i n d i c a t o r .
D e t a i l s a r e g i v e n in T a b l e 1.
The Co-efficient of curvature (Cc) and the uniformity co-

efficient ( C u ) are determined by plotting the grading curve and by
using such curve particle sizes that corresponding to 10%, 30% and 60%
passing (DI0, D30 and D60 respectively). The Co-efficients are then
calculated as follows [!] :
D60
Cu = (I)
DI0
2
D 30
Cc = (2)
(DI0) (D60)
Values of DI0, D30, and D60 are obtained from qrading curves.
All mineral fillers except limestone have shown a higher value than
grading above 10% passing as explained by Figures I, 2 and 3. This
makes the calculations for (Cu) and (Cc) to be indetermined. The
uniformity co-efficient is a measurement of the range of particle
sizes that are present, the larger the ratio of (D60) to (DI0), the
wider the range of sizes. The co-efficient of curvature is a
measurement of the shape of the gradation curve which relates to its
uniformity of particle distribution.
Atterberq L i m i t s
Liquid limit, plastic limit and plasticity index for

each filler were determined as per British Standard (BS) 1377. All
fillers under study were found to be non-plastic except the
limestone filler which was found to be plastic with a plasticity index
of 7 (noting that the American Association of State Hiqhway and
Transportation Officials "AASHTO" specifies maximum allowable limit
for plasticity index of 4). Atterberg limits for each filler are
given in Table i.
100-- ~ , ,~ ~ l ;:.
90 ...... r ........:.-; ........7:i .................... ............. t ........ ;......... / ; ~ I i ............. ~. i .....
P
a
. . . . . . . . " i i .
6
8
i
ri
g
% . . . . l k ! l . . : t '
~o,
,o .... ...... i ,
..... ).: :;:f ..... t I
lO- ; ;7 ~~7<7; ; ; I
0 _L i i iiii l.-" i i i J__Lmi_.LI i i i i iiiiI____X_ i I i iii
1.000E-04 1.000E-03 0.01 0.1 1
81eve ( m m )
Wadi Filly
FIG. 1 - - W a d i F i l l y & G r a d a t i o n L i m i t [4_..]
100!
9o}-
80 . . . . . . . . :" "'1
i //1~
P 70
a ~)I ~t]t - "
s 60
8 ............ : I t"fi 't
i 50
%
n
g 4o F
]. 9 .7;
I .-~J"' I ./"[ :
=~f
10
0
1.000E-04
'
1.O00E-03 0.01
I tl
0.1
S i e v e {ram)
x Manama
FIG. 2 - - M a n a m a Filler & Gradation Limit
100
90
..
80
P 7 o
i ........... . . . . . . . ;..... /!l~, " ........... :...........
a 7 : : : r ; ;; ! ~ / :7 : ; : : I
8 60 ......... ~ .............~ ................................... ; ! / / ..........! - ~--,?-~ . . . . . . . . . . . . . . . . : ..............
8
i
n
g 40 ........... " ' ' ................. .............. ' / i ........... , i ............ . . . . . . . : . . . . . . . . . . . ' ........
I : : ' : ' ' : .,l t ) : : ' " :
% 3 0 1 - ...... ; ; .... i ..iw/;;~)/.: ; ; :; . . . . .
~oI-"....: : .....> 5 ~ .;;r" ....'

...... ..<i~i :
0L~_-L i i ~• ~'/ i i i iiiil__~ i i i iii,t i i..I i i i i ~ l
1.000E-04 1.000E-03 0.01 0.1 1
Sieve (ram)
Lime Stone
FIG. 3--L.Stone Filler & Gradation Limit
1~176 i
eov ................. ;.--; ......
/ til~
r #:./.........
/ 9 ".....
#
80 t ................................
;........L ..................................... ................. 7. .......... i-, ......
P 70 ..... '- ...... ;. . . . . . . . . . . . . .

+U :i ......
a
ll 8o t ....... { ............ :-; ....................... I / ............. ; . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8 I
/ .
i / " .if .......... ........

n
g 40 - 9
./
20 - ; ............ :l . . . . . . . . .;:--~_~ :!z-
10 : " H ~ " >" / "
0 __L t i iiJ_ilL---~, i L ilii . ~ l ~ I I I 1 t J~L_LLL!_.
1.000E-04 1.000E-03 0.01 0.1 1

Sieve (mm)
* Kadra
FIG. 4--Kadrar Filler & Gradation Limit
Surface Area
Surface areawas determined using the Blain air p e ~ i l i t y

apparatus (ASTM C204). The sample of hydrated lime filler could not
be tested because a proper bed could not be prepared and the san~le
of Bagrat (KS-300) was too bulky to carryout the test. The surface
area for each filler is given in Table i.
Specific Gravity and Voids in Conpacted Filler
The specific gravity ( relative density ) of fillers was

determined using the test procedure described in BS 812. Also, the
void characteristics of each filler was measured using BS 812.
Specific gravity, voids in conpacted filler and surface area for
fillers in this study are given in Table i.
pHValue
The pHvalues of fillers were determined after mixing with

water in the ratio of water/filler = 2.5 as per BS 1 3 7 7 . Results
are also given in Table 1 .
Micro-structural Analysis
Slide s ~ p l e s of different fillers, and samples of filler mixed

with bitumen were prepared as per ~ C295 for micro-structural
analysis. The shape and form of the fra~nents for each filler were
observed under a polarized microscope to ascertain the extent of
differences in the mineral content. The minerals were not identified
except for a calcareous classifcation. Fillers 1 to 7 as listed in
Table 2 are natural crushed rock where isotrophic minerals were
absent, while 8, 9 and i0 are manufactured chemical types of filler.
Observations which gave the particles shape (crushed/uncrushed),
texture and mineral classification were then recorded and are
shown in Table 2.
C~4ICAL~ITION
The chemical con~position of the fillers used in this study

were as tested and analyzed as per ASTMCII4 and are shown in Table I.
P R O P I ~ I E S OF BINDII~
A single asphalt cement binder (AC 60/70) was used to prepare

filler-bitumen mixtures and was utilized for preparation of asphaltic
mixtures. Properties of the 60/70 binder are given in Table 3.
SHAHROUR AND SALOUKEH ON PASSING SIEVE NO. 2 ~ 195
TABLE 2 - - M i c r o s t r u c t u r a l analysis of fillers (AS.TM C - 2 9 5 ) .
Serial THe of Calcareous s Columnar Fleochrortsm

Humber Filler Shape Sock Frafaent Fra~eut tathlike %
t t Fragment |
1 Line Stone Crushed tO0 lil ill lil

2 Gibca Crushed t2 9g 2 90
3 Manana Crushed 23 75 2 85
4 [adra Crushed 2 65 35 99
5 Wudi Filly Crushed 15 83 17 97
6 kl Yawoon Crushed i 92 ! 96
7 5iji Crushed 28 ~0 2 IO
8 (5-300 Uucrushed Nil Ill Hit Ill
9 O,P,Ceaent Crushed Ill ill Nil Ill
10 Hydrated gncrushed iil Nil Jil all
Lime
TABLE 3--Propertles of asphalt cement binder (AC 60/70)
Serial Name of Test Test Test

No. Method Result
1 Softening Polnt,oc A S T M D36 51.7

(R and B)
2 Kinematic Viscosity, cSt A S T M D2170 468
at 135~
3 Ductility, cm A S T M DII3 >100
(25~ 5cm/mln)
4 Penetration, .lmm A S T M D5 57
(100g, 25~ 5sec)
5 Specific Gravity A S T M D70 1.034
6 Mixing Temperature,6C A S T M D2493 154-160
(viscosity 170+I-30 cSt)
7 Compaction T e m p e r a t u r e , ~ ASTM D2493 142.5-147.5
(viscosity 280+I-30 cSt)
PROPEI~rIES OF FIT.I.RR-BIND~
As is well known, a relationship exists between the deformation

resistance of an asphalt mix and the cc~oosition of the mortar. A
higher stiffening effect of the mortar produces a road surfacing with
a higher resistance to deformation. According to Velske [5], in
order to achieve good practical properties, the binding agent film
must remain so thin that it offers only minimal possibilities for
deformation and the stresses in a binder of low viscosity are
transmitted over the smallest distance possible. The exceptional
importance of the filler is demostrated here, where as the term
(filler) points out to the participation in the filling of voids in
the grain structure. Filler also contributes to the conversion of the
pure straight-run binder into a mortar.
At higher field temperatures, the deformation behavior of

asphalt pavement becomes critical. For this reason, the highest
possible viscosity for the filler-bitumen mixture is desirable
because this has a favourable influence upon the deformation
resistance of the pavement. Thus, at a certain temperature an
increase in the viscosity can be achieved either with more filler or
with the use of effective filler. For all asphalt mixtures, the
viscosity of a filler-bitumen mixture cannot be increased unlimitedly
by increasing the concentration of filler amount. Such a high
concentration will cause problems in achieving the required
compaction and will cause problems with more aging and cracking.
Additionally, more binder is required to cover extra filler amounts
added to the mix.
The proportioning of ccalx)nents for filler-bitumen mixtures was

based on a weight basis. Filler-bitumen ratios considered by weight
which are used in this study are : 0.5 and 1.5 ( 0.25 and 0.8 for some
fillers were also investigated). The 0.5 ratio corresponds to a
typical Dubai wearing course mix proportion obtained from the
job-mix formula. The filler-bitumen ratio can be computed, on a
volume basis, from the apparent specific gravity value of the
fillers as follows:
F/B ratio (by wt) x sp gr of bitumen

F/B ratio = (3)
(by vol) Apparent sp qr of filler
The filler-bitumen ratio used in this study is on a weight

basis unless otherwise stated. Before mixing, both AC 60/70 and
filler were preheated in an oven to 155 degrees celsius (mixing
temperature of AC 60/70). The mixing process was accomplished in an
isomantal thermostatically controlled to maintain higher temperature.
The fillers were added in increments and mixing took place until a
homogeneous mix was obtained. After mixing, the mixtures were poured
into small metal containers and stored for future tests.
The stiffening effect of the filler was determined by adding

different percentages of each type of filler to AC 60/70 and
measuring the penetration ( 100g , 25 degrees celsius, 5 sec) of the
mixtures as per ASTM D5, and softening point using Ring and Ball
(R & B) test as per ASTM D36. Also, the kinematic viscosity (as per
ASTM D2170) of the mixture at 135 degrees celsius was determined.
Figure 5 shows the difference in softening point for selected

filler-bitumen using different filler/bitumen ratios.
A wide range in stiffening was shown by the different fillers

and ratios used in this study and is listed in Table 4. The
viscosity of the filler-bitumen mixtures divided by the viscosity of
asphalt c~nent AC 60/70 is calculated to obtain a stiffening ratio
based on the viscosity for each filler - bitumen mixture. The
hydrated lime filler had shown a superior stiffening ratio (4.4)
compared to all other filler types (ranges from 1.4 to 1.9).
A study by Mcleod has shown that the addition of a suitable

additive, can change dramatically the temperature susceptibility of
paving mixtures [6].
Fillers are considered as additives, and therefore the filler

type must be selected carefully to achieve the change required [6].
For an understanding of paving temperature susceptibility,

Figure 6 illustrates that the telrperature susceptibility of a filler
bitumen is the change in its consistency (penetration or viscosity)
for a given change in temperature [6]. In Figure 6 three filler-
bitumen mixtures, each with the same consistency at 25 degrees
celsius but with quite different consistencies above 25 degrees
celsius are illustrated .
Filler-bitumen 3 has the steepest slope of the three filler-

bitumen mixes, its consistency changes the most over a given range of
t~nperatures, and it is considered here to have a high t~perature
susceptibility. The smallest change in consistency for a given change
in temperature is shown by filler-bitumen i, therefore it provides
greater pavement stability at all temperatures. The slope of filler-
bitumen 2 in Figure 6 is intermediate between filler-bitumen 3 and
filler-bitumen i, and it is said to have intermediate medium
temperature susceptibility.
The temperature susceptibility for filler-bitumen mixtures is

measured in temns of their penetration at different temperatures (25,
30 and 35 degrees celsius). By measuring penetration at different
tenperatures, and plotting the results, a penetration index (P.I.) can
be calculated from the equations shown as follows [!] :
leg penetration (at Tx - at Ty)

Slope A = (4)
Tx - Ty
Log/xSP(~162
100 ~ .................................................................................................................................................. : ........... . .............. ....
PZ:IZ::ZZI i:iii17 i:::171:i/::i':iii::::i::: I:I+I::II:::::III:IZ:ZTZZZI:::::III::TI::II:I: i::5: ::::::5:iii::ii:i::::i i:ii:ii5:ii1511 ::ii:ii:iii:::iii:ii15::i:::::iii:i:i:ii:::i:i:::
~iiii:ili:iii:,::i:i: il:
I ......................................... 5~: +/. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 : ~ . ~ . - + + - . ~ .................................................
10 }
....... .
-/
. . . z ~
- / : /
..~---
~
..............
+
E .... :: :: + + ~ ' - ~ ...... L.:~ :: : .... " : ........... :: .......: : : ........
F .................................................................
0 0.5 1 1.5 2
F i l l e r / B i t . Ratio
HYDRATED LIME -+-- MANAMA

WADI FILLY -m-- O.P,CEMENT
FIG. 5 - - D i f f e r a n c e In S o f t e n i n g Point Of
S e l e c t e d F i l l e r - B i t . V s F / B Ratio
Consistency
1
l- .......... :~.-.~::'+:,,~+. ..... : + ~ : .... : ..... : .......................... +.................. +.................. I
: : -~
.......... .................i .............................................................................................. ,.....................................i...................:............
i i J ~ i t ~ i i i i
20 30 40 50 60 70 80 90 t00 110 120 130 140

Temperature ~
Filler-Bit.1 i Filler-Bit.2 ~ Filler-Bit.3
FIG. 6 - - D i f f e r e n t Temp.Sueceptibility
TABLE 4--Properties of filler-bitumen mixtures.
{a) (b)
F/B Penetration
(O.lmm,5sec,!OOg) P.l. Softening Kinematic Stiffeninl
Seria[ AC 60/70 ratio ............................ Point Viscosity Ratio
No. + 25~ 30~ 35'C based on
Type of By (~ (cSt) Vis. at
Filier Neight R & B at 135'C 135'C
....... iiJ JJ .... fT6 ....... 7i ...... ii .... ...... ...... ......... Z:; .....
!,5 27 42 66 0.00 67,4 *
Gibca 0.5 43 7! !18 - 0.60 55.8 810 1.7

!,5 26 38 56 + 1.11 64.1 *
3 Manama 0,5 44 68 107 + 0.24 56.9 841 1,8

!.5 23 34 51 + 0.91 72.9 *
4 Kadra 0.5 46 72 113 + 0,16 52.4 773 1.7

1,5 25 38 58 + 0,53 67.3 *
5 Wadi Fiiiy 0,5 42 72 123 - 0,70 54.3 789 1.7

!.5 27 40 60 + 0.91 64.4 *
6 A1Tawoon 0.5 42 70 !17 - 0.70 55.4 817 1,7
!.5 25 37 55 + 1.11 69.3 *
7 Siji 0,5 45 69 109 + 0.25 52,2 852 I.B

1,5 24 34 52 § 0,94 66.5 *
8 KS - 300 0,25 ** ** ** ** 59.0 * **
9 O.P. Cement 0.5 44 71 i15 - 0.28 56.3 676 1.4

1,5 26 40 52 0,00 66.1 *
s Hydrated Lime 0,5 34 52 80 + 0.50 60.9 2066 4.4

0,8 23 33 46 ~ 2.00 79,4 *
................................................................................................
9 Can not be determined as per ASTM D2170.
k*
The mix was exceeding the fluid property, and therefore can not be tested as per related
ASTM standards.
(a) F/B = Filler-Bitumen Ratio
(b) P.[. = Penetration index (EquaEion 5)
Where, the values of log penetration at Tx and at Ty are taken

from the graph for any two different temperatures (Tx, Ty).
20 - 500i%
Penetration Index (P.I.) = (5)
1+50A
A P.I. of > + 2 represents filler-bitumen mixture of low

temperature susceptibility, while a P.I. of < - 1.0 indicates a filler
bitumen mixture of high temperature susceptibility [8].
Penetration indices for all filler-bitumen mixtures used in

this study are given in Table 4.
MARSHALL MIX DESI(~ STtK)Y
A selected grading was considered to carry out a Marshall mix

design study using Wadi Filly aggregate. The bulk specific
gravity and water absorption of the combined aggregate used were
obtained as per ASTM C127 and C128 (bulk spcific gravity = 2.90, water
absorption = 0.6%). The bitumen absorption of the combined aggregate
was calculated (bitumen absorption = 0.3%) as per the Asphalt
Institute Manual Series No. 2 (MS-2)[9]. The gradation adjustments
shown in Table 5 were made in the gradation of the mixes when the
filler content was increased from zero to 2, 4, and 6%. In order to
accon~3date the increased quantity of filler, an equal volume of fine
aggregate (minus No. 16 )was removed and replaced by an equal volume
of filler. Four inch M~rshall specimens were prepared and tested for
each filler type using different filler contents (0%, 2%, 4%, and
6%). For each percentage of filler content, 15 Marshall specimens were
prepared at different binder contents (3.5%, 4%, 4.5%, 5% and 5.5%)
except for some fillers (hydrated Iime and Bagrat"KS-300") where
the use of low bitumen content was not sufficient to provide proper
coating for the aggregate. Mixing and conpacting temperatures of
asphaltic mixtures were obtained as per ASTM D2493. The Marshall
specimen were compacted using 75 blows for each face. The specimen
were then tested for bulk specific gravity as per ASTM D2726, void
content (voids in mix "%VIM", voids in mineral aggregate "%VMA ", and
voids filled with bitumen "%VFB "), stability and flow.
Calculations of (%VIM) and (%VMA) were obtained as per MS-2, (%VFB)
was calculated as follows :
% VFB = [ (VMA - VIM) / VMA ] x i00 (6)
TABLE 5--Gradation For Mixtures Containinq 0, 2, 4 and 6% Filler
Percent Filler in Mix 0 % 2 % 4 % 6 %
Sieve Size (ram) Total Percentage Passing
25 100 I00 I00 I00

19.5 95 95 95 95
12.5 84 84 84 84
9.5 78 78 78 78
4.75 48 48 48 48
2.36 30 30 30 30
1.18 20 20 20 21
0.60 13 14 14 15
0.30 9 10 11 12
0.15 5 6 7 8
0.075 0 2 4 6
A selected binder content (4%) was considered for crmRmarison

study from the range(3.5-4.5) which represents the bittanen content
range applied by different specifications used in Dubai. This range
has been found to be more suitable to Dubai conditions where agcirecjate
absorption of bitumen is very low (0.2 to 0.5% for the combined
aggregate used in the mix) and very high ambient temperatures are
encountered. Under such conditions, the optimum binder content
calculated as per MS-2 can not be applied, since it will lead to very
high bitumen contents which provoke bleeding.
The selected optimum binder content applied in IAthai which

satisfies different specifications in current use can be as much as
20% lower than the optimum bitumen content calculated as per MS-2.
Figures 7, 8, 9 and I0 represent Marshall parameters (%VFB,

%VIM, %VMA and Bulk Specific Gravity) for mixes using selected binder
and fillers (4 fillers were chosen based on different specific gravity
of filler: Wadi filly and OPC representing a higher specific gravity
range, Manama representing a medium specific gravity range, and
hydrated lime representing a lower specific gravity range). Test
results are shown in Table 6.
Test results of asphaltic mixtures using eight fillers and

different filler contents are shown in Tables 7, and 8.
The Marshall parameters were not significantly affected by

changing the type of filler at specific filler contents, lamever, the
type of filler significantly affected the filler-bitumen properties
by stiffening the binder to different degrees (~hich is clearly shown
in filler-bitumen properties in Table 4.). The stiffening of bitumen
% V.F.B.
68 . . . . . . . . .
66
64
62 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .-~--.
5 8 . . . . . . . ~, z 9
/ ./ -
56 ..................................... : - ~ : ~ t J ~:. . . . . . . . . . . . . . . . . .
.............................~ ~ ...............................................................
54
52
L I I
50
0 0.5 1 1.5 2
Filler/Bit. Ratio.
Filler Type
H.Lime ~ Manama D O.P.Cement • Wadi Filly
FIG. 7 - - % V.FR. VS F/B Ratio
% V,I,M.
9
8.5 ..............................................................................................................................................
8 ................... ~ ..................................................
7.5 . . . . . . . . . . ~.~ \ . . . . . . . . . . . . . . .
7 ............................................... ~ ' ~ " ~ " ~ . ................................................................

e.5 .............................................................................. ~ : : ~ ~ : ~ . ...............................................................................................
..............................................................~ ................ .~'~,,~.: ~ . ..............~ ......................................................
5.5 ..................................................................................................
5 I I I
0 0.5 1 1.5 2
Filler/Bit.
Filler Type
HJ_ime ~ Manama o O.P.Cement ~ Wadi Filly
FIG. 8 - - % V.I.M VS F/B Ratio
% V.M.A.
17
15.8t- ...........................................................................................................................................................
1 6 . 6 I- ................................... ~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1 6 . 4 ~- ......... ~ . . . . . . . . . . . . .
16.2 i- .........................................~ ' \ ~ ..............................................................................
1 5 . 6 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ::-:::: .............................................................
15.4 . . . . . . . . . . . . . . . . . . . ~~+~'-..~ ' ~ . . . . . . . . . . . . . .
15.2 . . . . . . . . . . . . . . . . . ~ ............ : x ~ .....................
15 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ~ ................... ~ . ~ ..................
14.8 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . "~. .......... ~. ..........
14.6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . :': . . . . . . . . . . . . . . . . . :'~"~ . . . . . . . . . . .
14.4 .........................................................................................................................................~ .....................................
14.2 ................................................................................................
14 i i i
0 0.5 1 t5 2
Filler/Bit.
Filler Type
9 H.Lime ~ Manama o O.P.Cement x Wadi Filly
FIG. 9 - - % V . M . A . V S F I B R a t i o
B u l k 8p g r
2.6
2.59 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
2.58
2.57
2.56
2.55
2.54
2.53
2.52
2.51
.... I I I
2.5
0 0.5 1 t5 2
Filler/Bit.Ratio
9 RDI FILLY i MANAMA

-~ O.P.CEMENT o HYDRATED LIME
FIG, l O - - B u l k ap gr V8 F/B Ratio.
204 A S P H A ~ MIXTURE PERFORMANCE
TABLE 6--Marshall Parameters For SeLected FiIlerB

At Different Filler-Bitumen Ratio.
Type of FIB Bulk | t ~ Stability Flow

Filler Batio 5p Gravity VIM VI~ VFB (Kgf) (n)
Manama 0.5 2,529 7.3 16.1 54.6 t690 3.6
i.O 2,552 5.8 15.3 62.0 t654 3.2
i.5 2.576 6.0 [4.4 58,6 i315 3.2
Wadi Filly 0.5 2,519 7.7 16.6 53.8 t607 3.t
1,0 2.564 6.1 15.1 59.5 1819 3.3
1,5 2.567 5.7 15.0 62.0 1560 2.6
KS - 300 0.4 2.515 6.2 t6.4 62,2 1288 3.3
O.P. Cesent 0,5 2.522 7.9 16.6 52.1 1098 2.9
i.O 2.563 6.3 15.4 58.9 1583 2.8
1.5 2.579 6.0 15.0 59.8 1339 3,0
Hydrated 0.5 2.544 6,3 15.3 58.5 1693 3.0

Lime
i.O 2.549 5.7 14.6 60.8 t560 3.4
1.5 * * * * * *
Without 0 2.473 9,8 17.7 44.6 1016 2.6

Filler
* : Can not achieve proper coating of aggregate.
SHAHROUR AND SALOUKEH ON PASSING SIEVE NO. 2 ~ 205
TABLE 7--M~arshall Study Using Selected Oradinq
Filler Type and Percentage Used
Marshalll t Hydrated Lime Lime Stone gadl Filty

lest IBinder Without Filler 21 41 6| 21 41 61 21 41 6t
13,5 2,473 2.5i2 * * 2,522 2.546 2,566 2.504 2.531 2.546

Bulk 14.0 2.495 2.544 2.549 * 2.530 2.564 2,581 2.519 2.562 2.568
Specific) 4.5 2,511 2.555 2,582 * 2.554 2.587 2.613 2.541 2.572 2.594
Gravity 5.0 2.518 2.572 2.602 * 2.575 2.590 2.616 2.563 2.590 2.615
5.5 2.532 2.581 2,595 * 2.585 2.605 2.614 2.577 2.609 2.615
3.5 9.8 0.3 i i 0.1 7.1 6.4 9,1 0.I 7.4

4.0 0,3 6.3 5.7 * 6.6 5,7 4,7 7.6 6,2 5.0
! 4.5 6.9 5.i 3.7 * 5.2 4.1 2.6 6.1 4,9 4.1
VIN 5.0 5,9 3.6 2,2 * 3.7 2,8 i.7 4.4 3.3 3.4
5.5 4.5 2.5 i.6 * 2.5 i.7 0.0 3.1 2.1 1.6
3.5 17,7 15.9 * * 15.9 15.I 14.6 16.6 15,7 15,3

4,0 17.3 15.3 14.6 * 15.0 14.0 i4.2 16.0 15.1 15.1
I 4,5 17.3 15.3 14.0 * 15.7 14,5 16.7 16.3 15.2 14.5
VHA 5.0 17,5 15,2 13,7 * 15.5 i4.6 13.0 16.1 14,9 14.3
5.5 17.4 15.4 14,4 * 15,6 14,8 14,2 10.1 15,1 14.0
3.5 44.6 48,2 * * 49.7 54.3 56.1 45.0 48.1 51,9

4.0 52,1 58,1 61.5 * 50.2 62.2 66.4 54,2 60.i 61.3
I 4.5 60.1 66.3 74.3 * 67.1 73.i 81,3 62.6 61.2 73.1
VFH 5.0 66,3 76.4 85.1 * 75.9 81.9 87.9 72.5 78.3 83,0
5.5 74.1 83.9 89.2 * 83.9 09.2 94.1 00.6 87.2 08.9
............................
3.5 1016 1500 * * 1339 1500 1250 1441 1640 1580

Stabi- 4.0 1292 1680 1560 * I537 1620 1290 1607 1820 1560
iity 4.5 i401 1600 1630 * 1568 1590 1240 1520 1740 1500
(Kgf) 5.0 1367 1560 1599 * 1479 1500 1350 1630 1650 1550
5,5 1311 1550 1503 * 1609 1470 i290 1571 1670 1510
3,5 2.6 2.7 9 * 2.9 3.1 3.3 2.9 3.1 2.0

4,0 2,9 3,1 3.4 * 2,9 3.1 3.5 3.1 3.3 2.6
Flow 4,5 3.4 3.1 3.9 * 2.9 3.5 3.2 3.1 3.3 3.1
(mm) 5.0 3.1 3.4 4.0 * 3.4 3,6 3,6 3.3 3.3 3.1
5.5 3,2 3.1 4.4 * 3.3 3.6 3.6 3.3 3.5 3.2
* : Cannot achieve proper coating of aggregate.
206 A S P H A ~ MIXTURE PERFORMANCE
TABLE 8--Marshall Study Osinq Selected Gradlnq
Filler Type and Percentage Osed
MarshaLll t : O,P, Cement ) Kadra : Yavoon Gibca

Criteria~Binder) 21 41 6| : 2~ 41 68 ~ 2| 41 68 21 41 6~
,, ,,
I 3,5 2,494 2,539 2,564:2,512 2.541 2.536',2.522 2,546 2.566 2.504 2.531 2.546
Belk ', 4.0 2.524 2.562 2.57712.534 2.568 2.562:2.5[9 2.558 2.559 2.522 2.554 2.555
Specific'. 4.5 2.532 2.5"/8 2.60512.552 2.584 2.59612.531 2.578 2.588 2.538 2.584 2.586
Gravity 5.0 2.554 2.586 2.615', 2.572 2.622 2.60212.558 2.60[ 2.62[ 2.560 2.605 2.60[
5.5 2.572 2.61 2.62512.584 2.622 2.60212.570 2.601 2.622 2.572 2.605 2.611
t i
i i
t i
3.5 9,7 8.[ 7.4 I 8.5 8.3 7.3 : 9.1 7.5 7.3 8.8 7.9 7.0
4.0 8.[ 6.4 6.3 :6.8 6.4 5.8 :7.7 5.9 5.0 7.6 6.4 6.5
t 4.5 6,8 4.9 4.3 15.5 5.[ 3.7 ~ 6.5 4,4 4.[ 6.[ 4.0 4.l
VIM 5.0 5.2 3.8 3.1 I 3.8 3.3 2.2 : 4.7 2.0 L9 4.3 2.9 3.2
5.5 3,6 2.2 2.[ :2.6 L.7 3.2 : [.8 [.L 3.4 2.t 2.t 2.t
i I
8
3.5 17.1 15.7 15.2 I [6.3 [4.8 15.3 I 16,7 15.4 15.2 16.5 15,6 15,4
4.0 16.6 t5.4 15.[ : 16,[ [4.2 [5.2 : [6.5 [5.3 t5.[ [6.5 [5.4 15,3
I 4.5 16.7 [5.3 14.2 I [5.9 14.6 14.4 : 16.5 14.8 14.5 16.3 14.8 14.7
V~ 5.0 [6.4 [5,5 H.3 : [5,7 [4.6 14.4 I [6.1 [4.8 13.8 15.9 14,6 [4.7
5.5 16,4 15.1 14.7 I 15.7 15.2 14.9 I I6.I 15.1 14.3 16.2 15.1 14.0
. . . . . . . . . . . . . . . . . . . . . . . . . . . . .
3.5 43.3 49.8 51.1 I 48.2 45.4 52.1 ', 45.7 50.9 52.2 45.8 48.0 49.9
4,0 53.2 58.9 59.9 ', 56.8 56.8 62.7 ', 53.2 60.8 60.9 53.8 50.8 57.5
t 4,5 59.[ 67.7 70.8 '. 64,9 66.2 75.3 ~ 6[.3 70.7 73.8 62.3 70.3 70.[
VFH 5,0 68.4 75.5 79.9 I 73.8 77.2 88.9 '. 71.2 81.2 85.8 73.4 80.1 77.9
5.5 76.9 86.2 86.0 ', 82.7 82.[ 92.4 : 79.9 87.8 92.7 79.8 87.2 86.7
3,5 1180 1320 [460 I 1340 [440 1450 I 1240 1490 1350 [460 1540 1380
4.0 [[00 1560 [340 : [430 [580 [500 : 1520 [700 1425 [560 [690 [380
Stabi-I 4,5 1490 1540 1300 : [520 [5[0 L490 I [5[0 [700 [475 [740 [ 7 3 0 1420
[ity I 5,0 [450 [450 1340 ~ [570 1480 [360 : 1500 1620 [520 [450 [720 [350
(Kgf)) 5,5 1390 1480 1450 : 1420 [450 [380 I 1620 1580 1310 1450 [520 1400
I
I 3,5 2+9 2+8 3.3 I 2.8 3.1 3.1 t 2+9 2.7 2.8 2.0 2.8 LL
I 4,0 2.9 2.8 3.1 : 3.1 3.1 3.3 :2.9 3.1 3.1 3,1 3.2 3.5
Flow I 4.5 3.2 3.3 2.8 12.9 3.2 3.5 ~2.9 3.2 3.5 2.9 3.2 3.S
(mm) l 5,0 3.7 3.2 2.9 13.4 3.2 3.3 13.2 3.3 3.2 3.3 3,2 4.2
5.5 3.3 3.4 2.8 13.4 3.6 3.5 : 3.3 3.2 4.[ 3.2 3.4 4.3
........ I
will reflect on the stiffening of the asphaltic mixture. To evaluate

such contribution, advanced mechanical testing, such as :creep,
indirect tensile and resilient modulus are required to be carried out.
I. Test results show that all types of mineral fillers used in

Dubai ~nirate play the role of an extender to the binder with minimal
stiffening effect.
However, the selection of a filler type to improve the

binder stiffening property under extremely high temperatures
p r e v a i l i n g in Dubai a r e a is extremely c r i t i c a l . F i l l e r s which show
more o f a s t i f f e n i n g e f f e c t (by modifying the b i n d e r p r o p e r t y
s i g n i f i c a n t l y under high temperatures) should be c o n s i d e r e d to
a d d r e s s the permanent deformation problems on IAtbai r o a ~ . Further
r e s e a r c h i s , however, r e q u i r e d .
2. Hydrated lime filler has shown superior stiffening

properties when mixed with the binder compared to all other
filler types. For an addition of 0.8 ratio of hydrated lime to
bitumen, the softening point increases from 51.7 (for AC 60/70) to
79.4, the penetration (0.1mm) decreases from 57 to 23 at 25 degrees
celsius, the PI increases from - 0.78 to + 2.0.
I. It is r~nded to use hydrated lime filler in

current asphaltic mixtures in a ratio of 0.5 to 0.8 of the bitumen
content as a partial substitute of the mineral filler.
2. Two mineral fi llets (Wadi Fi 1 ly and Manama) and hydrated

lime are to be further studied in the next phase of the research.The
authors will prepare asphalt samples using a typical grading used
in Dul~i and perform advanced testing such as indirect tensile, creep
and resilient modulus. The purpose of such testing is to study the
mix susceptibility to temperature using different fillers, and the
fillers contribution to the rutting problem encountered in Dubai
roads. Finally, the study is expected to produce with criteria for
filler specifications which will hopefully contribute to minimize the
rutt ing problems.
I~EF~mqCES
[I] Warden, W.B., Hudson, S.B. and Howell, H.C., "Evaluation of

Mineral Filler in Terms of Practical Pavement Performance".
[2] Prithvi S. Kandhal, "EAraluation of Baghouse Fines in Bituminous

Paving Mixtures," Proceedinq of AAPT, Vol.50, 1981, 10p 150-210.
[3] Dr. David A. Anderson, " Guidelines on the use of B a ~ o u s e

Fines", National Asphalt Payment Association, IS101-11 , 1987,
pp 1-6.
[4] U.S. Army, "Investigation of the Design and Control of Asphalt

Paving Mixtures, Technical Memorandum # 3-254, Waterways
Experiment Station, May 1948.
(5] Velske,"Baustofflehre - Bituminous Stoffe", Wener Ingenieur -

Texte 25, 1976, pp 8-30.
[6] Norman W. Mcleod, "tluployirg Paving Asphalt Temperature

Susceptibility in the Structural Design of Asphalt Paveme, ts",
1987, pp 240-258.
[7) John M. Edward, "Introduction to Bituminous Materials", lecture

note, 24th to 28 March, 1980.
[8] John M. Edward, "Analytical Design of Bituminous Pavement" ,

University of Nottingham, lecture note, 24th to 28th March,
1980.
[9] Asphalt Institute, Asphalt Institute Manual Series No.2 (I~3-2),

Chapter VI, 4th. Edition, August 1988, pp 67-79.
Mixture Evaluation
Tien F. Fwa I and Slew A. Tan I
L A B O R A T O R Y EVALUATION OF RUTTING POTENTIAL OF A S P H A L T M I X T U R E S
REFERENCE: Fwa, T. F. and Tan, S. A., "Laboratory Evaluation of

Rutting Potential of Asphalt Mixtures," Effects of AHHregates and
Mineral Fillers on Asphalt Mixture Performance, ASTM STP 1147,
Richard C. Meininger, editor, American Society for Testing and
Materials, Philadelphia, 1992.
ABSTRACT: This paper describes a laboratory procedure to evaluate

the rutting potential of asphalt mixtures using a wheel tracking
machine. Moving loads were applied through solid rubber tired
wheels that travel with a reciprocating motion on slab specimens of
asphalt mixtures. The specimens were prepared in parallepipedal
steel molds by means of static compression to achieve fairly uniform
density distribution within individual specimens. The machine
allows rut tests to be conducted at different temperatures,
magnitudes of applied loads, and speeds of loading. Results of a
test program are reported to illustrate the application of the
procedure to study the effects of temperature and loading speed on
rut development in an asphalt mixture.
KEYWORDS: asphalt mixtures, rutting deformation, rut depth, rutting

rate, wheel tracking machine.
Rutting of asphalt pavements is a major form of distress

observed on the roads of Singapore where daily pavement temperature
(except on rainy days) rises above 50~ throughout the year.
Rutting deformations are commonly found in the wheel tracks of the
slow lanes of major roads, and the approaches to traffic-light
junctions in the city and industrial areas. Due to the
inconvenience and high costs of frequent repairs in these areas, the
road authorities of Singapore are currently experimenting with
different designs of asphalt mixture to combat the problem of
rutting.
As part of the effort to identify rutting-resistant asphalt

mixtures suitable for use in the hot tropical climate of Singapore,
the present study was undertaken to develop a laboratory means to
evaluate the rutting potential of asphalt mixtures. This paper
describes the capability of the rut test apparatus, and presents the
Isenior Lecturers in the Department of Civil Engineering, and

Principal Researchers in the Centre for Transportation Research,
National University of Singapore, Kent Ridge, Singapore, 0511.
211

Dec 22 07:29:10 EST 2011
OataAcquisition ~
rigid coveting ~T 1 /9
IIt' II..'t
- I r'
, ,1 ~ w,,,
25f,mm carriage~ ~J'rubber tyred wheels
~-2L-~ //~< 5% C~weight,~ ~ o.sskwc~s,a~ torque
~ ~ : ~ ,, - , , - : : ~ ~ - II Motor driving crank at
.....
wei""~ ~ ~--------~/ / - [[ / conlroUedspeed
heater ~ ---------A---2-2-_-_ w
w Iar t t ' ' ' "'"
I c ~ l ] ~ - - ~ -~i~ bmperature
]r ~drain va{ue n\ '
jj n ~ournn ~ spro.ybar
FIG. I -- Schematic of wheel tracking apparatus.
results of a test program conducted to demonstrate its applications

to the evaluation of rutting potential.
WHEEL TRACKING APPARATUS
Various forms of full scale track tests if, 2] and laboratory

simulated wheel tracking models [3, 4] have been used by researchers
to study rutting in asphalt pavements. Full scale test tracks are
costly to construct and maintain. They are beyond the reach of most
highway agencies and researchers. Laboratory wheel tracking models
are cheaper, with the advantage that specimen preparation and
tracking tests can usually be completed within days. The rut test
apparatus adopted in the present study belongs to this category. It
was a modification of a three-wheel immersion tracking machine used
in the United Kingdom [5]. The major modification is in the rut
depth measurement method where an automated electronic system was
installed to replace the chart drum and pen recorder system in the
original apparatus. Figure 1 is a representation of the apparatus
whose main features include control of test temperature, wheel
speed, and wheel loads.
Test specimens were prepared in one of the following three

sizes: 407mm by 137mm by 90mm, 407mm by 214mm by 90mm, or 407mm by
443mm by 90mm. The apparatus can test three specimens of the
smallest size concurrently. When specimens of the other two sizes
are chosen, only one specimen could be accommodated per test. The
wider specimens are useful in circumstances where the influence of
test mold on rut formation becomes important.
Test temperature was maintained by means of a water bath heated

with an electric heating element. Water bath temperature could be
adjusted within a range from room temperature to 70~ by an
electronic thermostat with a differential setting of 0.2~ An
FWA AND TAN ON RUTTING POTENTIAL OF ASPHALT MIXTURES 213
adjustable weir kept the water level at about 5 mm above the top
surface of test specimens. Control of wheel speed was achieved by a
constant torque geared motor with electronic speed control. The
free-rotating wheels, each measured 45 mm wide and had a diameter of
200 mm, covered a path of 254 mm in a reciprocating motion. A
constant test speed of between 0 and 80 passes per minute could be
selected. A test speed of n passes per minute means that the test
wheel will pass a given point of a specimen n times every minute.
Each cycle of the reciprocating motion of the wheel produces two
passes for all the points on the wheel tracks. Wheel loads could be
varied from 18 to 54 kg to give approximate equivalent tire pressure
from 175 to 530 kPa.
EXPERIMENTAL PROCEDURE
Specimen Preparation
The mix proportions and aggregate gradations for the two

asphalt mixtures tested, designated as W3 and WI, are shown in Table
I. Asphalt cement of 60/70 penetration grade and granite aggregates
were used. The specimen size selected was 407mm by 137mm by 90mm
(16.0 in. by 5.4 in. by 3.5 in.). Two methods of specimen
compaction were tried out. The first method followed the procedure
laid out in ASTM Standard Practice for Preparation of Bituminous
Mixture Beam Specimens by Means of the California Kneading Compactor
(D 3202-83). The mixture was compacted in two equal layers. Each
layer received 40 tamping blows at a compactor pressure of 75 psi
(517 kPa), followed by 40 blows at I00 psi (689 kPa) and 40 blows at
200 psi (1,379 kPa) foot pressure. The entire specimen was then
compressed under a static load at the Fate 6.4 mm/min until an
applied pressure of 400 psi (2,758 kPa) was reached. In the second
method, the entire mass of the mix was compressed directly under a
static load of 75 t, giving an average compaction pressure equal to
13,450 kPa. The constant pressure was maintained for a duration of
i0 min. Both methods produced specimen density in the order of 2.20
g/cm 3 for the two mix types considered. Higher density could be
achieved if desired by applying higher pressure.
It was desirable to be able to produce specimens of the same

density and having similar density distribution within each
specimen. A twin-probe gamma-ray nuclear gauge [6] was used to
determine non-destructively the cross-sectional density distribution
of each specimen. The second method of compaction was adopted for
the study because of the better uniformity of density obtained
TABLE 1 -- Mix proportions and aggregate gradations.

Mix Binder Aggregate size distribution (% passing)
Type Content 19mm 13mm 9.5mm 6.4mm 3.2mm 1.2mm 0.3mm O.075mm
W3 5.8% i00 95 90 77 58 37 19 6
W1 6.1% i00 I00 I00 95 74 47 27 6
within each specimen, and its better repeatability in terms of the

level of compacted density achieved. Furthermore, a relationship
between the magnitude of applied load and compacted density can be
established for a given mix if one is interested in studying the
effect of compacted density on the mix's rutting potential.
Temperature Control
The rut test apparatus had a water pump which circulated water
in the bath containing test specimens. The stability of temperature
within test specimens were checked by monitoring specimen
temperature in actual test conditions. Temperatures were measured
at the top and bottom surfaces, and the mid-point of three specimens
during a trial test. For the rut test condition where the top
surface of specimens was about 5 mm below water level, it took about
2.5 h for all three specimens to reach the bath temperature of 60C ~
Therefore, three hours of preheating of test specimens was allowed
in the thermal bath of the apparatus before all rut tests.
Subsequent temperature measurements during rut tests indicated the
top, bottom, and mid-point temperatures of test specimens were
maintained within I~ of the desired test temperature.
Rut Depth Measurements
Rut depths developed in test specimens were computed from

measurements of the vertical distance of the axle of the wheel of
interest from a pre-determined reference level. This reference
level was provided by a horizontal gantry which slid along the two
sides of the rut test apparatus. The vertical distance was measured
with an electronic LVDT, one for each of the three wheels.
By means of an electronic triggering system installed along one

side of the apparatus, vertical distance measurements could be made
at designated points along "the horizontal path of travel. The data
acquisition unit allowed for automatic measurements to be made with
the wheel in motion. However, this proved to be difficult because
of errors induced by machine vibration. Instead, static
measurements were made by stopping the test at about 1 h intervals.
This was not a serious limitation at all because each set of static
rut profile measurements of five points along the wheel tracks of
three specimens took less than two minutes. All measurements were
made with the resolution of LVDT to O.Ol mm.
DISCUSSIONS OF TEST RESULTS
For the W3 mix, rutting tests were conducted for two

temperatures ( 45~ and 60~ ) and two test speeds ( 20 and 60
passes/min ). The test speeds of 20 and 60 passes/min correspond to
average wheel speeds of 0.3 and 0.9 km/h respectively. To study the
effect of mix type on rutting, specimens of WI mix were tested at
60~ for the same two test speeds.
I_ 2.25x _1
APPLIED
[ LOAO
Jllllll,,,,ll/ , t t t t
,ttt
FIG. 2 -- Pattern of deformation by finite element analysis.
Deformation Pattern
Analytical solutions indicating the likely deformation pattern

of asphaltic materials under simulated loading can be obtained by
means of finite element analysis or other suitable techniques.
Figure 2 shows the pattern of permanent displacements derived from a
finite element solution [7] for a problem with loading and boundary
conditions similar to the rutting test conducted in this study.

Each arrow identifies the direction of deformation, and the length
of the arrow represents the magnitude of the movement. The solution
indicates that downward compression would occur directly under the
applied load, lateral movements take place in materials near to the
loaded area, and upward movements are found near the side walls
because of the constraints of the rigid boundary.
To enable deformation trends to be identified in the rutting

test, all specimens were cut into four sections of equal length
prior to the test. Each cut face was vertical and perpendicular to
the longitudinal axis of the specimen. Horizontal and vertical
lines forming rectangular grids were drawn on each cut face (see
Fig. 3(a)- ) so that the pattern of deformation could be traced.
Density of the material within each rectangular grid was determined
by means of the twin probe gamma-ray density gauge.
Typical test results are shown in Figs. 3(b) and 4. The

general deformation characteristics highlighted earlier by the
analytical model are verified by the experimental results.
Practically all the vertical deformation under the loaded area took
FIG. 3 -- C r o s s - s e c t i o n a l views of W3 specimen before as
after test
--" 29
I
I
, .~ -'g-.~ ,.-. ,. ,..,
I
I
|
o ,C~ Q C~
C~ -original position of aggregate grid line before rutting test

:"'~-.., -displaced position after rutting test -- grid line after rutting test
FIG. 4 -- D e f o r m a t i o n pattern of W3 s p e c i m e n in r u t t i n g test.
Test Speed = 60 passes/min

Temperature = 50%
E -2 Humber of
E Wheet Passes
c:, 1200
~ - ~. 3600
u-I
Q -6
5000
z 8LO0
p.-. 10800
13200
15600
-I0 i ~ ~ I i I l
0 50 100 150 200 250 300 350 ~00
DISTANEE ALONGLONGITUDINALAXIS OF SPEEIMEN
FIG. 5 -- Rut p r o f i l e s of W3 s p e c i m e n at d i f f e r e n t s t a g e s o f test
place within a depth of 60 mm from the top surface. Away from the
center-line of the loaded area, the tendency of lateral movement can
be noted. This led to rotations of the stones in this region and
heaving on the two sides of the loaded area, as can be seen from
F i g . 4.
LonAitudinal Rut Profile
Typical longitudinal rut profiles at different stages of the

rutting test are shown in Fig. 5. It can be seen that rut depth was
not uniform along the wheel track. The deepest rut depths were
found to occur at the two ends of the wheel track, while the
shallowest was located near to the mid-point of the track. It is
believed that the following two features of the rutting test were
the causes for the differences between rut depths at the ends and
near the mid-point of a wheel track: (i) The speed of a wheel
decreased towards zero as it approaches an end of the wheel track,
became zero momentarily at the end of the wheel track before the
wheel moved again in the reverse direction. The loading time for
the material at the end of a wheel track was therefore longer than
near the mid-point where the speed of the wheel was higher. (ii) The
described changes of speed at the end of a wheel track also brought
about deceleration and acceleration forces on the test specimen.
The corresponding surface forces were less near the mid-point where
more gradual speed changes were involved.
Figure 6 plots the speed and acceleration profiles of the

cyclic wheel tracking movement during the rutting test at a test
speed of 60 passes/min. Figure 6(a) traces the time variations of
speed and acceleration, while Fig. 6(b) shows the speed and
acceleration variations along the wheel track. Similar results were
also observed for the test speed of 20 passes/min. The constant
torque machine provided a relatively uniform speed and acceleration
for the major portion of the travel. The central portion of the
travel path was subjected to a repetitive pattern of speed and
acceleration reversals. At the two ends of the travel path, the
wheels were brought to an abrupt stop, as can be seen from Fig.
6(b), before they started to move in the opposite direction. The
asphaltic materials at these two points experienced additional
surface forces as they had to resist the rotational inertia of the
wheels. These data confirm the differences in the nature of loading
at the end points and near the mid-point of the wheel track, as
identified in the preceding paragraph.
Rate of Rutting
The r a t e o f r u t t i n g with traffic loading of a design mix is a

major concern of pavement engineers. For example, an asphalt
mixture that develops a great portion of its total rut depth in the
early stage is undesirable as it requires frequent maintenance and
tends to provide relatively low s e r v i c e a b i l i t y during most of its
service life. Depicted in Fig. 7 is an example showing the
development of rut of W3 specimens as the test progressed. A sharp
increase in rut depth occurred initially, possibly due to
E
(Q) VQriafions with time /Acceler~ion
1.0
C~
"< 0.5
u.J
.._...J
~. I .,% ,~ i%. t /% /
ttl
~ 0 ~. !/ ~x I./ ~x. ,.,.ix ;

N -0.5
~
E
-1.0
I/ \ J I~, ~ ,:
m
m -1.5 I _ I I I
2 t, 6 8
TIME (sec)
~-- 1.5
(bl Variations wffh wheel position
E 9 . / AccelerQtion
1.0
< 0.5
-...,./
~ -0,5
Speed A ~"
N -1.o -- A
~ -0.5 I l .I I I I
0 50 100 150 200 250
POSITION OF WHEEL (mm)
FIG. 6 -- Variations of wheel speed and acceleration
5
E
:=1::
~3
~2
:::I:=
~ / Speed 20 passes/min
ij_l ol V l"emper0turel,~l;
i-.-- z
oO I I I I I I
2.5 5 7.5 10 12.5 15 17.5
WHEEL PASSES ( x 103)
FIG. 7 -- Rut development of W3 specimens in a ruttinE test.
50
~o/~k,k Speed 20 passes/min o Wl~eel]
E lemperature 1,5~ ~ W~ee:~
E
,~ 30
• 20 ~
-< 10
\d
~ : -t0 I I I I I f
0 2.5 5 Z5 10 12.5 15 17.5
WHEEL PASSES(x I03}
FIG. 8 -- Rate of rutting of W3 specimens in a r u t t i n g test.
d e n s i f i c a t i o n of the mixture. The increase tended to stabilize as

more wheel passes were applied. This general trend of changes in
r u t t i n g rate with the number of wheel passes, as illustrated in Fig.
8, was observed in all the tests conducted in this study.
Currently no field data have been collected to compare with the

pattern of rut development obtained from the laboratory tests. Due
to differences in the loading conditions and rate of loading, it is
unlikely that the rate of rutting observed in the current tests can
be used to predict field rutting behavior directly. Further work is
n e e d e d t o s t u d y how t h e l a b o r a t o r y observed rutting rates could be
related to actual field performance.
Indian Institute of Technology Kharagpur pursuant to License Agreement. No further reproductions auth
TABLE 2 -- Rut depths of W3 specimens after lO,gO0 passes

Test Test Rut depth after 10,800 passes (mm)
Speed Temperature At mid-point At end point
20 45~ 3,89mm 5.93 mm
passes/min
60~ 12.95 mm 14.08 mm
60 I 45~ 2.74 mm 3 . 6 9 mm
passes/min [ 60~ 8.03 mm 7.86 mm
Effect of Temperature
Table 2 records the rut depths after 10,800 wheel passes for
different test conditions. These results are plotted in Fig. 9
which shows that deeper rut depths were measured at the higher test
temperature. This is within expectation because the asphalt binder
would be less viscous at higher temperature, and hence more prone to
plastic deformation under loading.
The effect of temperature became increasingly pronounced when

the test speed was reduced. This has a significant implication on
the rutting behavior of asphalt pavements at traffic-light junctions
where slow moving and static traffic loads are of major concern. An
illustration of the implication can be made using the results for
the test speed of 20 passes/min (about 0.3 km/h). Assumin~ 10,800
passes of load were applied respectively at 45~ and 60-C, the
corresponding rut depths (from Table 2) were 3.89mm and 12.95 mm.
That is, about 77% of the rut depth would be developed at 60~ It
is easy to see from this example that in the tropical climate of
Singapore, where pavement surface temperature stays above 60~ for 3
to 4 h on a typical hot sunny day [8], the rutting performance of
asphalt mixtures at 60~ is a critical design consideration for
pavements at traffic-light junctions and slow lanes that carry heavy
traffic.
Effect of Loading Speed
In general, as illustrated in Fig. 9(a), shallower rut depths

were obtained when higher test speeds were adopted. The difference
in rut depths obtained with different test speed selections were
especially significant at 60~ the temperature at which rutting
deformation should be evaluated for the Singapore climatic
conditions. Although no guidelines are currently available as to
what test speed should be used to study the rutting problem at
traffic-light junctions, 20 passes/min appears to be an acceptable
choice judging from the test results based on the Singapore W3 mix.
This test speed produced sufficient depths of rut for comparison
purposes at the end of 10,800 passes, requiring a total running time
of 9 h which could be accommodated within a day.
Figure 9(b) suggests that the effect of test temperature tended
E
E
(a) Effect of Temperature ~-.o ~Test Speed
12
(Mid-Point} I*" ~ / )20 pQsses
per min
(End P0inl}
~ 10
~- "~ ~ ~ '~ Test Speed
~ 8 t/" "~ ~ J60 passes
~ g o.1 / .-. ~ per m,n
I..-.-
u_ / .......-. ~..-"-
--,r"-
ELI
~ 0 I ! I I
/,5 50 55 60 65
TEST TEMPERATURE(~
E
E
I/., Tested ~ A ~ . . . (b) Effect of Test Speed
r
at 60~ ! , , ~ ~ ~ ~ (Mid- Point)
o
12
I0
8
~ End Point)
Tested Io----~ ...........

L, at LS~ to------_
"'Jr-"
I.---
O
.-,
i,I
2
0 1 I I I I
,r",r-- 10 2O 30 ~0 50 60 70
TEST SPEED(passes/mini
FIG. 9 -- Rut depth as a function of test speed and temperature
TABLE 3 -- Rut depths of W3 and W1 specimens tested at 60~

Rut Depth of W3 Specimens Rut Depth of W1 Specimens
Test Speed
Mid-Point End Point Mid-Point End Point
20 passes/min 12.95mm 14.08mm ll.54mm ll.58mm
60 passes/min 8.03mm 7.86mm 4.16mm 4.71mm
to diminish as the test speed was increased. The difference in rut

depths at 60~ and 45~ fell from about 9 mm when tested 20
passes/min (0.3 km/h) to about 5 mm when tested at 60 passes/min
(0.9 km/h). Unfortunately, due to the limitation of the apparatus,
it was not possible to verify this trend for the range of traveling
speeds of actual traffic. Assuming that the trend in Fig. 9(b) is
valid, one would expect that under Singapore conditions where
pavement temperature varies within the range of 25~ to 70~ the
influence of temperature on rutting behavior of a through road
section would be very different from one at a traffic-light
junction.
Effect of Mix Type
The rut depths of W3 and WI specimens after 10,800 wheel passes

at 60~ for two different test speeds are compared in Table 3. The
results show that the W3 specimens rutted more than the W1
specimens. Although it is difficult to identify the actual reasons
for the different rutting behavior observed, differences in mix
composition, aggregate gradation and compacted density are believed
to be the major contributing factors. The binder content was 5.S%
for the W3 mix and 6.1% for the W1 mix. While the W3 mix had bigger
top size aggregates, the W1 mix had much higher proportions of
sand-size particles which could contribute to the strength of binder
mortar. The average bulk densit~ for the W3 and Wl specimens was
respectively 2.Z26 and 2.198 g/cm-, and their corresponding air void
contents were 7.8% and 7.6%.
CONCLUSIONS
A relatively simple apparatus for laboratory evaluation of

rutting potential of asphalt mixes was described. The testing
characteristics of the apparatus and the rutting behavior of test
specimens were examined. The test program conducted in this study
illustrated the ability of the described test procedure to identify
the effects of temperature, loading speed and mix type on the
rutting potential of asphalt mixtures. Based on the test results of
two Singapore mixes, it appears that rutting potential comparison of
different mixes could be made by conducting the test at 60~ with a
test speed of Z0 passes/min.
The main limitation of the test procedure is that the test

results cannot be related directly to field rutting behavior of
in-service pavements. The rate of rutting of an in-service pavement
is determined by the actual traffic loading which cannot be easily
simulated by an accelerated laboratory test such as the one adopted

in this study. The effect of reciprocating moving loads as
opposed to unidirectional traffic loads is also not known.
Nevertheless, the findings of this study indicate that the procedure
described can be useful as a practical tool for comparing the
rutting potentials of different asphalt mixtures.
REFERENCES
[I] Sharma, M.G., Kenis, W.J. and Mirdamadi, L, "Evaluation of

Mechanical Properties of Inservice Pavements from Field Data,"
Proc. 6th Int. Conf. on Structural Design of Asphalt Pavements,
University of Michigan, Vol. I, 1987, pp. 550-561.
[2] Autret, P., deBoissoudy, B. and Gramsammer, J.C., "The Circular
Test Track of the Laboratoire Central Des Ponte et Chaussees
Nantes -- First Results," Proc. 6th Int. Conf. on Structural
Design of Asphalt Pavements, University of Michigan, Vol. I,
1987, pp. 710-724.
[3] Valkering, C.P., Lancon, D.J., deHilster, E. and Stoker, D.A.,
"Rutting Resistance of Asphalt Mixes Containing Non-
Conventional and Polymer-Modified Binders," Proceedings Assoc.
of Asphalt Paving Technologists, Vol. 59, 1990, pp. 560-609.
[4] Nievelt, G. and Thamfald, H., "Evaluation of the Resistance to
Deformation of Different Road Structures and Asphalt Mixtures
Determined in the Pavement-Rutting Tester," Proceedings Assoc.
of Asphalt Paving Technologists, Vol. 57, 1988, pp. 320-345.
[5] Road Research Laboratory, Bituminous Materials in Road
Construction, Her Majesty's Stationery Office, London, 1962,
pp. 138-140.
[6] Tan, S.A. and Fwa, T.F., "Nondestructive Density Measurements
of Cylindrical Specimens by Gamma-Ray Attenuation," 3ournal of
Testing and Evaluation, ASTM, Vol. 19, No.2, 1991, pp. 155-160.
[7] Goacolou, H, "Calculation of the Rutting of Structures --
CASTOR Program," Proc. 6th Int. Conf. on Structural Design of
Asphalt Pavements, University of Michigan, Vol. I, 1987, pp.
191-199.
[8] Tan, S.A. and Fwa, T.F., "Influence of Pavement Materials on
the Thermal Environment of Outdoor Spaces," Accepted for
publication in Building and Environment, Int. J. of Building
Science and its Application, 1991.
G e r a l d A. Huber and T. Scott Shuler
PROVIDING SUFFICIENT VOID SPACE FOR ASPHALT CEMENT:

RELATIONSHIP OF MINERAL AGGREGATE VOIDS AND AGGREGATE
GRADATION
REFERENCE: Huber, G.A., and Shuler, T.S., " P r o v i d i n g

Sufficient Void Space for Asphalt Cement: R e l a t i o n s h i p of
M i n e r a l Aggregate Voids and Aggregate Gradation",
AgQreQates and Mineral Fillers on AsPhalt Mixture Performance.
ASTM STP 1147, R i c h a r d C. Meininger, editor, American Society
for T e s t i n g and Materials, Philadelphia, 1992.
ABSTRACT: Since the 1960's, the Asphalt Institute has

r e c o m m e n d e d voids in the mineral aggregate (VMA) to control
m i n i m u m asphalt content in asphalt aggregate mixtures. VMA
criteria are b a s e d on aggregate nominal m a x i m u m sieve size.
Use of V M A criteria requires a clear u n d e r s t a n d i n g of the
r e l a t i o n s h i p of V M A to aggregate gradation. Nominal m a x i m u m
size and m a x i m u m density lines must also be understood.
In the asphalt industry today, significant confusion exists

c o n c e r n i n g different methods used to draw "maximum" d e n s i t y
lines. A need exists to determine which m e t h o d represents the
m a x i m u m p a c k i n g of aggregate materials. M a x i m u m d e n s i t y lines
are required to m o d i f y gradations and change VMA. Closely
related to m a x i m u m density lines is the d e f i n i t i o n of aggregate
nominal m a x i m u m size which also requires clear definition.
This paper reports the relationship of VMA to aggregate

gradation and particle characteristics for a c o n t r o l l e d
experiment. Using results from this experiment plus three
other databases of mix designs, different methods of drawing
m a x i m u m density lines are evaluated. A definition of nominal
m a x i m u m sieve size which is more specific than the current ASTM
d e f i n i t i o n is proposed.
KEYWORDS: m a x i m u m density line, nominal m a x i m u m sieve size,

voids in mineral aggregate
Mr. Huber and Dr. Shuler are research engineers at the Asphalt
Institute, Box 14052, Lexington, KY 40512
225
Copyright by
Copyright 9 ASTM
1992 Int'l
(all rights
by ASTM
International
07:29:10 EST 2011
INTRODUCTION
F r o m the earliest days of asphalt mixture design the important role of

aggregate gradation was recognized [I]. Various methods of analyzing
gradations were evaluated searching for an aggregate gradation ~'law"
which c o u l d be used in the selection of gradation for m i x t u r e design.
Today, aggregate gradations are commonly e v a l u a t e d using a "0.45 power

chart". Despite the chart'S usefulness some confusion exists regarding
its use in practical applications. One use of the 0.45 power chart is
to estimate available voids in the m i n e r a l aggregate (VMA~ of c o m p a c t e d
mixtures. Increased V M A is obtained by m o v i n g further from the m a x i m u m
d e n s i t y line but several methods exist for d r a w i n g m a x i m u m density
lines.
This paper provides some history of gradation evaluation and current

use of the "0.45 power chart". The main question to be a n s w e r e d is "How
should the m a x i m u m density line be drawn?" A review of commonly used
m e t h o d s for d r a w i n g m a x i m u ~ density lines is made and several databases
of m i x designs are evaluated to determine which m e t h o d more
a p p r o p r i a t e l y relates to change in mixture V M A with change in gradation.
Specific recommendations are made regarding use of the 0.45 chart to

evaluate gradations.
HISTORY OF GRADATION ANALYSIS
The link between aggregate gradation and asphalt m i x t u r e performance

was recognized early in the development of mix design methods. Prior to
the establishment of formal mixture design methods, gradation was
e v a l u a t e d to determine asphalt demand. Formulas were applied to the
gradation and asphalt requirements were c a l c u l a t e d which recognized the
need for a m i n i m u m amount of asphalt binder to provide s a t i s f a c t o r y
durability.
By the 1920's during development of the Hubbard Field m e t h o d of m i x

design the importance of air voids was recognized as a key parameter
which c o n t r o l l e d field performance of mixtures [2, 3]. The H u b b a r d
F i e l d mix design is based on the need for air voids and for a m i n i m u m
amount of asphalt binder. Voids in total mix and voids in aggregate
mass were both specified.
Separate from the work of Hubbard and Field, the M i c h i g a n State

Highway Department developed a mixture design m e t h o d in the early 1930's
in which the shape of the gradation curve was e v a l u a t e d and e x p r e s s e d as
a "Gradation Modulus" from which the "Bituminous C a p a c i t y of Aggregate"
was d e t e r m i n e d [4].
All of these early mixture design methods were b a s e d on a belief that

a "gradation law" e x i s t e d which controlled asphalt m i x t u r e properties.
HUBER AND SHULER ON VOID SPACE FOR ASPHALT CEMENT 227
C o n s i d e r a b l e research effort was e x p e n d e d to d i s c o v e r this law which

c o n t r o l l e d aggregate packing. A s s o c i a t e d with the g r a d a t i o n law was the
belief that an "ideal" gradation e x i s t e d which w o u l d p r o v i d e adequate
space for a m i n i m u m amount of asphalt and air voids as well as provide
adequate stability under traffic.
By the 1940's, several researchers including N i j b o e r [5], W e y m o u t h [6]

and F u l l e r [6] p r o p o s e d g r a d a t i o n curves which they b e l i e v e d r e p r e s e n t e d
the ideal g r a d a t i o n for m a x i m u m p a c k i n g of aggregate solids. Nijboer
showed e x p e r i m e n t a l l y that the densest p a c k i n g of a g g r e g a t e p a r t i c l e s
was o b t a i n e d if the gradation, as p l o t t e d on a log-log chart of percent
p a s s i n g vs sieve opening, had a slope of 0.45. In the late 1950's Goode
and Lufsey [7] d e v e l o p e d a chart to evaluate aggregate gradations in
which the vertical axis was an arithmetic scale of percent p a s s i n g by
weight and the horizontal axis was sieve o p e n i n g raised to the power
0.45. This is the "0.45 chart" commonly used today to evaluate
aggregate gradations.
A l t h o u g h the 0.45 chart has p r o v e d useful for g r a d a t i o n analysis, it

does not represent the "gradation law" d e s i r e d by e a r l i e r researchers.
Nijboer, Goode and Lufsey and Hudson and Davis all c o n f i r m e d that the
densest aggregate p a c k i n g was obtained for gradations, which when
p l o t t e d on a log log chart of percent p a s s i n g by weight versus sieve
size opening, had a straight line with a slope of 0.45 [6]. However, in
practice, few if any actual m i x designs meet this straight line
r e l a t i o n s h i p and agreement on how to evaluate more typical gradations is
not unanimous.
In their original p u b l i c a t i o n Goode and Lufsey o m i t t e d a m e t h o d of

d r a w i n g a m a x i m u m d e n s i t y line. In response to a written d i s c u s s i o n
they r e c o m m e n d e d a m e t h o d to select a m a x i m u m size d e p e n d i n g on shape of
the g r a d a t i o n curve. Their m e t h o d has not g a i n e d acceptance. Instead
several m e t h o d s of d r a w i n g m a x i m u m d e n s i t y lines exist t o d a y both in the
literature and in methods routinely used. There are significant
d i f f e r e n c e s among the methods and no one m e t h o d appears more correct
than another.
GRADATION AND VOIDS IN MINERAL AGGREGATE
The p a c k i n g c h a r a c t e r i s t i c s of coated aggregate p a r t i c l e s in an

asphalt m i x t u r e are related to aggregate surface c h a r a c t e r i s t i c s and
gradation. A g g r e g a t e surface c h a r a c t e r i s t i c s of the p a r t i c l e s include
a n g u l a r i t y and surface texture. Gradation is the size d i s t r i b u t i o n of
the particles. When selecting aggregate for a project, surface
c h a r a c t e r i s t i c s are not selected to obtain VMA. Conversely, V M A of a
m i x t u r e is o b t a i n e d by default. If additional VMA is required, changes
are u s u a l l y made to the aggregate gradation. In practice, changes are
often made by adding natural sands which are p r e d o m i n a n t l y minus 600 ~m
sieve material.
Natural sands have been identified as a cause of d e c r e a s e d resistance

to p e r m a n e n t d e f o r m a t i o n and tender m i x problems d u r i n g construction.
As a result, limits have been p l a c e d on sand content and increases in
V M A must be a c h i e v e d by overall adjustment of gradation. U n l i k e natural
sand addition, g r a d a t i o n adjustment can sometimes p r o d u c e c o n f u s i n g
results. M o v i n g away from the "maximum d e n s i t y line" sometimes causes

decreases in VMA.
AsPhalt Institute Experiment
To evaluate the effect of aggregate surface c h a r a c t e r i s t i c s and

gradation an experiment was p e r f o r m e d at the Asphalt Institute. The
experiment was d e s i g n e d to evaluate gradation without c o n f o u n d i n g by the
effects of different angularity and surface texture. The experiment
objective were
i. To compare VMA of mixtures which had identical gradations but

different aggregate surface characteristics.
2. To compare VMA of mixtures which had the same aggregate type

from the smallest to largest aggregate size but have different
gradations.
Two aggregate sources were used. One aggregate was a c r u s h e d

limestone of low absorption quarried near Lexington, Kentucky. The
other was an u n c r u s h e d river gravel also of m e d i u m absorption from the
Ohio River in northern Kentucky. C h a r a c t e r i s t i c s of each aggregate
stockpile are listed in Table i.
TABLE 1 - - Characteristics of aggregate stockpiles in gradation

experiment
CRUSHED LIMESTONE UNCRUSHED GRAVEL
PROPERTY #68a I #4 1 #8 I #I0 #68 I #89 I Sand
Bulk S.G. 2.713 2.717 2.704 2.589 2. 633 2.626 2.624
Apparent S.G. 2.762 2.762 2.761 2.753 2.770 2.762 2.718
% Water A b s o r b e d 0.65% 0.51% 0.77% 2.3% 1.87% 1.88% 1.32%
% Crushed Faces I00.0 I00.0 100.0 I00.0 0.0 0.0 0.0

Stockpile #4, etc. refer to American Society for Testing and
Standards Standard Classification for Sizes of
Aggregate for Road and Bridge C o n s t r u c t i o n (D448).
Percent crushed faces b a s e d on particles larger than 4.75 m m
sieve.
M i x design gradations were selected independent of a stockpile

b l e n d i n g analysis. Gradations were selected to meet d e s i r e d gradation
curve characteristics. Hence, rather than using percentages of each
stockpile, the stockpiles were used to supply particles of specific
sizes.
Six gradations were used in the experiment as listed in Table 2 and

shown in Figures 1 and 2. All the gradations c o n t a i n e d five percent
p a s s i n g the 75 ~ m sieve. Gradations A, B and F are d e f i n e d by a power
r e l a t i o n s h i p e x c l u d i n g the percent p a s s i n g the 75 ~ m sieve as follows:
(I00-P75) (dn-75 n)
p = + P75
(Dn-75 n)
where
P = percent passing
P75 = p e r c e n t p a s s i n g 75 ~ m sieve, 5%
d = s i e v e size opening, ~rn
D = m a x i m u m s i e v e size opening, ~ m
n = exponent
F o r an e x p o n e n t of 0.45 w h e n p l o t t e d on a "0.45 p o w e r c h a r t " this

e x p r e s s i o n w i l l be as a s t r a i g h t line j o i n i n g t h e p e r c e n t p a s s i n g 75 ~ m
s i e v e w i t h the m a x i m u m size. For p o w e r s of n g r e a t e r t h a n 0.45, the
line w i l l p l o t as a c o n c a v e d o w n w a r d c u r v e j o i n i n g t h e m i n u s 75 ~ m sieve
w i t h the m a x i m u m size. G r a d a t i o n s A, B a n d F h a v e e x p o n e n t s of 0.55,
0.70 a n d 1.00 r e s p e c t i v e l y a n d p r o d u c e the g r a d a t i o n c u r v e s s h o w n in
F i g u r e I.
TABLE 2 - - Gradations used in V M A e x p e r i m e n t
GRADATIONf % Passinq
Sieve A I B o I E I F
18.75 mm i00.0 I00.0 I00.0 I00.0 I00.0 i00.0

12.50 mm 79.0 75.0 88.0 96.0 95.0 67.4
9.50 m m 68.0 63.0 74.0 79.0 83.0 52.3

4.75 m m 47.0 40.0 47.0 47.0 59.0 28.5
2.36 m m 32.0 26.0 35.0 37.0 43.0 16.5
1.18 m m 22.0 17.0 26.0 30.0 35.5 10.5
600 ~ m 15.0 12.0 20.0 25.0 26.5 7.6
300 ~m i0.0 8.0 13.0 15.0 12.2 6.1
150 ~rn 7.0 6.0 8.0 9.0 5.3 5.4
75 ~ m 5.0 5.0 5.0 5.0 4.6 5.0
G r a d a t i o n C a n d D are b a s e d on G r a d a t i o n A w i t h h u m p s on t h e 12.50 m m
a n d the 600 ~ m sieve. G r a d a t i o n E is a fine g r a d a t i o n m o d e l e d on an
a c t u a l g r a d a t i o n e n c o u n t e r e d in the field. G r a d a t i o n s C, D a n d E s h o w n
in F i g u r e 2 are all S - s h a p e d g r a d a t i o n s m e a n t to s i m u l a t e g r a d a t i o n s
w h i c h o f t e n o c c u r in p r a c t i c e .
A l l m i x t u r e s u s e d in this e x p e r i m e n t w e r e a r t i f i c i a l l y c o m b i n e d from
individually sized aggregate particles. Aggregate particles were sieved
into i n d i v i d u a l sizes, w a s h e d to r e m o v e c l i n g i n g d u s t a n d t h e n
r e c o m b i n e d in r e q u i r e d a m o u n t s to f o r m the d e s i r e d g r a d a t i o n .
00.0
90.0 9 Gradation A
80.0
9 Gradation B
70.0
(5
_z 9 Gradation F
60.0
(/)
50.0
Q.
40.0
30.0
20.0
10.0
0.0
75 300 1.18 2.36 4.75 9.5 12.5 19.0 mm
150 600
SIEVE OPENING (0.45 power)
FIGURE 1 - - Gradations A, B and F used in V M A e x p e r i m e n t
100.0
90.0
80.0
70.0
z 60.0
(fJ
50.0 9 Gradation C
D.
40.0
, Gradation D
30.0
9 Gradation E
20.0
10.0
0.0
75 300 1.18 2.36 4.75 9.5 12.5 19.0 mm
150 600
SIEVE OPENING (0.45 power)
FIGURE 2 - - Gradations C, D a n d E u s e d in V M A e x p e r i m e n t
D i r e c t c o m p a r i s o n s c a n b e m a d e b e t w e e n m i x d e s i g n p r o p e r t i e s of
crushed and uncrushed aggregate mixtures. A l l t h e p a r t i c l e s in e a c h m i x
design are from the same source. Uncrushed, natural sand was not used
w i t h t h e l i m e s t o n e b l e n d n o r w e r e l i m e s t o n e s c r e e n i n g s u s e d in t h e
u n c r u s h e d gravel mixture. Therefore, since aggregate g r a d a t i o n was

m a n u f a c t u r e d to identical curves, differences b e t w e e n c r u s h e d and
u n c r u s h e d m i x t u r e s are b a s e d on d i f f e r e n c e s in surface texture and
angularity.
A single asphalt cement used for this experiment was an AC-20 with the
c h a r a c t e r i s t i c s listed in Table 3. Mixing and compaction temperatures
were d e t e r m i n e d at 170 • 20 cSt and 280 • 30 cSt a c c o r d i n g to A S T M
S t a n d a r d Test M e t h o d for Resistance to Plastic Flow of Bituminous
M i x t u r e s U s i n g M a r s h a l l Apparatus (D1559).
Asphalt Institute ExPeriment Test ~xotocol
The t e s t i n g protocol of the e x p e r i m e n t a l plan c o n s i s t e d of p e r f o r m i n g

a complete M a r s h a l l m i x design for each aggregate g r a d a t i o n and type. A
total of 12 m i x designs were done (2 aggregate types times 6
gradations). A mechanical, flat-faced, fixed-base single M a r s h a l l
hammer was used for compaction. The hammer was m o u n t e d on an oak
p e d e s t a l m e e t i n g the requirements of A S T M D1559.
Figure 3 s h o w s a typical m i x design result for G r a d a t i o n E of the

u n c r u s h e d gravel. Two summary tables of m i x design results are listed
in Table 4 and 5. Table 4 lists the m i x t u r e properties at four percent
total asphalt content b y weight of mix. Table 5 lists m i x t u r e
p r o p e r t i e s at four percent air voids.
TABLE 3 - - Properties of asphalt cement used in g r a d a t i o n experiment
Property I Value
Viscosity (60~ 2150 poises
Viscosity (135~ 427 cSt

Specific Gravity 1.03
Mixing Temperature 150 - 156 ~
Compaction Temperature 140 - 145 ~
Asphalt Institute Experiment Comparison of A a g r e a a t e Type
M i x t u r e s made with u n c r u s h e d aggregates compact into a denser

arrangement than do u n c r u s h e d aggregates which have m o r e macro texture
(sharp edges) and m i c r o texture (freshly fractured faces). Aggregate
relative d e n s i t y can be e x p r e s s e d as the amount of voids in the mineral
aggregate. Hence, percent V M A is used to compare d e n s i t i e s of c r u s h e d
and u n c r u s h e d mixtures.
Air Voids, % VMA, %
14.0 I
15.0
4.0 "~, 13.0 m--m~m.....on/m
2.0
o.o
3.0
f% 4.0
u
~'~.
w
5.0
!
6.0
12.0
11.0
10.0
3.0
m
4.0
9
5.0
9
6.0
Asphalt Content, % AsphaltContent,%
Marshall Stability, N Density, kg/m3
8000 I
6000
m ~ ' m ~ m ~ ml,,.,
~
--m
2410.0
2390.0
./ m ~ m
4000
2000 2370.0
./'/
0 ! ! ' 2350.0
| 9 9
3.0 4.0 5.0 6.o 3.0 4.0 5.0 6.0

Asphalt Content, % Asphalt Content, %
FIGURE 3 Gradation E, uncrushed gravel aggregate Marshall design

curves
TABLE 4 - - Mixture properties at four percent asphalt content,

by weight of mix
Crushed Limestone Uncrushed Gravel

Air Marshall Air Marshall
Gradation Voids VMA Stabilit3 voids VMA Stabilit~
(%) (%) (N) (%) (%) (N)
A 2.3 11.5 12,770 1.7 10.2 7,030
B 3.1 12.1 10,940 1.9 11.9 5,780
C 2.5 11.6 13,880 1.8 10.5 7,610
D 3.8 12.8 14,460 2.1 Ii.I 8,360
E 5.7 14.4 14,190 4.5 13.1 6,940
F 6.4 15.2 8,410 3.6 12.4 3r870
TABLE 5 Mixture properties at four p e r c e n t air voids
Crushed Limestone Uncrushed Gravel

Air Marshall Air Marshall
Gradation Voids VMA Stability Voids VMA Stabilit
(%) (%) (N) (%) (%) (N)
b b
3.3 ii.0 a 3.1 II.2 a
b b
3.7 12.1 a 3.1 I0.7 a
b b
3.4 ii.4 a 3.2 10.5 a o . .
3.9 12.8 14,230 3.4 11.4 8,850

4.4 14.1 14,680 4.2 13.1 6,890
5.1 15.5 7,110 4.0 12.6 3r870

a E x t r a p o l a t e d f r o m d a t a at 4.0 t o 6.0% A C
b
M a r s h a l l s t a b i l i t y not m e a s u r e d
V M A of the u n c r u s h e d m i x t u r e s is t y p i c a l l y 1 to 2% less t h a n V M A of
the crushed aggregates. A i r v o i d s for the u n c r u s h e d a g g r e g a t e are
a l w a y s l e s s t h a n for c r u s h e d a g g r e g a t e , a l s o t y p i c a l l y 1 to 2% less.
The d i f f e r e n c e in air v o i d s b e t w e e n u n c r u s h e d a n d c r u s h e d a g g r e g a t e is
s h o w n o v e r t h e r a n g e of a s p h a l t c o n t e n t s in F i g u r e 4 for G r a d a t i o n E.
O v e r the e n t i r e range t h e a v e r a g e d i f f e r e n c e in a i r v o i d s is about one
percent.
80 r , ~ 9 Crushed
~~" 9 Uncrushed
~ 4.0 ~
2.0 9
0.0 I I I I !
30 35 40 45 50 55
Effective Asphalt Content, '
FIGURE 4 - - Gradation E, air v o i d c o m p a r i s o n for c r u s h e d and uncrushed

aggregates
Asphalt Institute Experiment Relationship of V M A a n d G r a d a t i o n
TO e v a l u a t e a g g r e g a t e m i x t u r e b l e n d s g r a d a t i o n s are p l o t t e d on a semi-
p o w e r f i g u r e in w h i c h p e r c e n t p a s s i n g is p l o t t e d on the v e r t i c a l axis
a n d t h e s i e v e size o p e n i n g r a i s e d to the 0.45 p o w e r is p l o t t e d on the
h o r i z o n t a l axis. T h e s e p l o t s of g r a d a t i o n are t y p i c a l l y m a d e as p a r t of
an a s p h a l t m i x t u r e design, not o n l y to q u a l i t a t i v e l y a s s e s s the shape of
t h e g r a d a t i o n , b u t also to m a k e c h a n g e s to the g r a d a t i o n d u r i n g design.
The most common reason to select a new g r a d a t i o n d u r i n g design is a

d e f i c i e n c y in voids in the mineral aggregate (VMA). In m a n y areas of
N o r t h A m e r i c a aggregate p a c k i n g tendencies result in very dense mixtures
with low VMA. P u b l i s h e d guidance to m a k i n g g r a d a t i o n changes is not
specific enough to be useful to mixture designers. Often vague
statements such as "Move further away from the m a x i m u m d e n s i t y line" is
the only advice available.
D A
100.0
0 S ....1~6ns'%;"Re~fe e r- - 6 6 6 ~ ] ~ [ R ~ e " ~ '~'';;~-~'~~
90.0 ......................................................
. . . .;.--.~;. j ; .~. ~. .,.,...,... ;~. . . . . . . . . . . . . . . . . . . ,,
80.0
70.0
Z 60.0
9 E .................................2 " ~ ; 9~' ~ ~J~ ' ~' ' ~ "~' ~ /.........................................................
m 50.0
13.
40.0
30.0 ..............................
20.0 ....................................
...,-Z:
10.0
0.0
75 300 1.18 2.36 4.75 9.5 12.5 19.0 m m
150 600
S I E V E O P E N I N G (0.45 power)
FIGURE 5 - - Location of gradations from d e n s i t y reference line
The gradations used in this experiment are shown in Figure 5 along

with a d e n s i t y reference line referred to as the m a x i m u m d e n s i t y line as
c u r r e n t l y d e f i n e d by the Asphalt Institute [8]. By looking at Figure 5,
a q u a l i t a t i v e ranking of distance from the d e n s i t y reference line can be
made. All gradations are b e l o w the line with G r a d a t i o n F b e i n g the
furthest away. Other gradations p r o g r e s s i v e l y closer to the m a x i m u m
d e n s i t y line are Gradations B, A, C and D with G r a d a t i o n E b e i n g the
closest.
M i n i m u m V M A for each of the m i x designs is listed in Table 6. If the

advice of "move away from the m a x i m u m d e n s i t y line" is correct, then VMA
should p r o g r e s s i v e l y increase m o v i n g from G r a d a t i o n E tO G r a d a t i o n F.
To the contrary the VMA decreases and then begins to increase. This
t r e n d is the same for c r u s h e d and u n c r u s h e d aggregate. V M A does not
seem to relate to distance from the m a x i m u m d e n s i t y line when drawn
t h r o u g h the nominal m a x i m u m sieve size.
By d e f i n i t i o n a m a x i m u m d e n s i t y line should be the g r a d a t i o n which

provides the densest p a c k i n g of aggregate particles. The closer a
g r a d a t i o n is to the m a x i m u m d e n s i t y line, the denser should be the
p a c k i n g of aggregate particles. Therefore, the d a s h e d line in Figure 5
does not meet the definition of a m a x i m u m d e n s i t y line. Indeed judging
by the V M A of each gradation, a m a x i m u m d e n s i t y line seems to run

somewhere t h r o u g h the middle of these gradations.
TABLE 6 - - V M A related to distance from m a x i m u m density line
Minimum VMAf %
Increasing
Distance From Crushed Unorushed
M a x i m u m D e n s i t y Lin6 Aggregate Aggregate
E 13.9 12.8
D 12.6 Ii
C 11.6 10.4
A 11.5 10.8
B 12.1 10.4
F 14.4 12.4
NOMINAL MAXIMUM SIEVE SIZE
Nominal m a x i m u m size of aggregate gradations and the d r a w i n g of

m a x i m u m d e n s i t y lines are two technical points which cause confusion in
the asphalt industry today. At least three different d e f i n i t i o n s of
nominal m a x i m u m sieve size are routinely u s e d by h i g h w a y agencies in
N o r t h A m e r i c a today, A S T M [9], the Asphalt Institute [8] and M c L e o d
[i0]. D e p e n d i n g on gradation, the nominal m a x i m u m size m a y be the same
by all d e f i n i t i o n s or they m a y be different.
The A s p h a l t Institute defines nominal m a x i m u m size as the first sieve

in the s p e c i f i c a t i o n to retain some material. M a x i m u m size is the
smallest sieve through which all aggregate will pass.
A m e r i c a n Society for Testing and M a t e r i a l s S t a n d a r d T e r m i n o l o g y

R e l a t i n g to M a t e r i a l s for Roads and Pavements, (D8), defines nominal
m a x i m u m size as "the smallest sieve opening through which the entire
amount of the aggregate is p e r m i t t e d to pass" M a x i m u m size is "the
smallest sieve opening through which the entire amount of the aggregate
is r e q u i r e d to pass" [9].
M c L e o d s u g g e s t e d nominal m a x i m u m size be d e f i n e d as "the next larger

s t a n d a r d sieve size than the sieve size on which at least i0 percent of
the total aggregate is retained" [i0]. M a x i m u m sieve size w o u l d be one
sieve larger than nominal m a x i m u m size and by d e f i n i t i o n w o u l d be the
smallest sieve through which the entire amount of a g g r e g a t e must pass.
Three aggregate gradations are listed in Table 7 and shown in Figure

6. Also listed in Table 7 is aggregate g r a d a t i o n b a n d s a s d e f i n e d by
A S T M D3515, Hot M i x e d Hot Laid Bituminous Asphalt M i x t u r e s [i0]. This
g r a d a t i o n b a n d is labeled as "1/2 inch nominal gradation" limits. By
inference this g r a d a t i o n band has a nominal m a x i m u m size of 12.5 m m
since 12.5 m m is the smallest sieve through which all m a t e r i a l is
p e r m i t t e d to pass. M a x i m u m size of this g r a d a t i o n is 19.0 m m s i n c e t h e

19.0 m m s i e v e is t h e s m a l l e s t s i e v e t h r o u g h w h i c h all t h e m a t e r i a l is
r e q u i r e d to pass.
N o w c o n s i d e r the n o m i n a l m a x i m u m size of G r a d a t i o n s I, II a n d III

u s i n g t h e i n f e r r e d A S T M d e f i n i t i o n , the A s p h a l t I n s t i t u t e d e f i n i t i o n and
McLeod's definition.
TABLE 7 Example gradations for m a x i m u m sieve size
ASTM Limits, % Aggregate Gradation, %
Passing Passing
Sieve Minimum Maximum II III
19.0 m m i00 i00.0 i00.0 100.0

12.5 90 100 95.0 100.0 75.0
9.5 74 92 83.0 79.0 63.0
4.75 44 74 59.0 47.0 40.0
2.36 28 58 43.0 37.0 26.0

1.18 18 41 35.5 30.0 17.0
600 ~ m I0 30 26.5 25.0 12.0
3O0 5 21 12.2 15.0 8.O

150 3 15 5.3 9.0 6.0
75 2 I0 4.6 5.0 5.0
ASTM
N o m i n a l Size 12.5 12.5 19.0
M a x i m u m Size 19.0 19.0 25.0
Asphalt Institute
N o m i n a l Size 12.5 9.5 12.5
M a x i m u m Size 19.0 12.5 19.0
Nominal Size 12.5 12.5 19.0

Maximum Size 19.0 19.0 25.0
U s i n g i n f e r r e d A S T M d e f i n i t i o n s , G r a d a t i o n I has a n o m i n a l m a x i m u m
s i z e of 12.5 m m a n d a m a x i m u m size of 19.0 mm. Using the Asphalt
I n s t i t u t e d e f i n i t i o n , G r a d a t i o n I has a n o m i n a l m a x i m u m size of 12.5 mm,
t h a t is, t h e first sieve to r e t a i n some m a t e r i a l is the 12.5 m m sieve
a n d a m a x i m u m s i e v e size of 19.0 mm.
W i t h the M c L e o d d e f i n i t i o n the m i x t u r e has a n o m i n a l m a x i m u m size of

12.5 m m s i n c e t h e first s i e v e w h i c h r e t a i n s m o r e t h a n 10% is 9.5 m m a n d
n o m i n a l m a x i m u m size is one sieve larger, that is 12.5 mm.
........ Min
100
........ Max ........................................................ ,.,~.- % ~ . . . . , . ~ . : _ ~ ...........
90
-'" .. '
, - .. . . . . . . f . j . ........ . / . . . . .........................
80
"' ~ ...........................................
5:; ............7 ; " /
70 ............................. .;,...~.. ...................... ;.,~ ......................................................
o
z_ 60
v~ 50
" 40
3O
20
," >.-'L..-.'" _ ...................
10
75 300 1.18 2.36 4.75 9.5 12.5 19.0 rnm
150 600 SIEVE (0.45 power)
FIGURE 6 - - Gradations illustrating definition of n o m i n a l maximum size
G r a d a t i o n II fits w i t h the A S T M 12.5 m m n o m i n a l m a x i m u m g r a d a t i o n

l i m i t s a n d c o u l d be c l a s s i f i e d as h a v i n g a n o m i n a l m a x i m u m size of 12.5
mm. U s i n g t h e A s p h a l t I n s t i t u t e d e f i n i t i o n , t h e m i x t u r e has a n o m i n a l
m a x i m u m s i z e of 9.5 m m s i n c e t h a t is the first s i e v e to r e t a i n s o m e
material. W i t h M c L e o d ' s d e f i n i t i o n the m i x t u r e is c l a s s i f i e d as h a v i n g
a n o m i n a l m a x i m u m size of 12.5 m m s i n c e t h a t is one s i e v e l a r g e r t h a n
t h e 9.5 m m s i e v e w h i c h is the first s i e v e to r e t a i n m o r e t h a n 10%
material.
G r a d a t i o n III by i n f e r r e d A S T M d e f i n i t i o n has a n o m i n a l m a x i m u m s i z e
of 19.0 mm. The g r a d a t i o n f a l l s o u t s i d e the 12.5 m m g r a d a t i o n b a n d but
w o u l d fit w i t h i n t h e 19.0 m m n o m i n a l g r a d a t i o n b a n d (not shown). On the
o t h e r h a n d the A s p h a l t I n s t i t u t e d e f i n e s G r a d a t i o n III as h a v i n g a
n o m i n a l m a x i m u m s i z e of 12.5 m m s i n c e it is s t i l l the first s i e v e to
retain some material.
G r a d a t i o n III is d e f i n e d b y M c L e o d as h a v i n g a n o m i n a l m a x i m u m size of
19.0 m m s i n c e the first s i e v e to r e t a i n m o r e t h a n 10% m a t e r i a l is t h e
12.5 m m sieve. M a x i m u m size is one s i e v e larger, t h a t is, 25.0 mm.
In summary, M c L e o d ' s d e f i n i t i o n c o n s i s t e n t l y c a t e g o r i z e s m i x t u r e s
c o r r e c t l y a c c o r d i n g to the A S T M g r a d a t i o n b a n d w i t h i n w h i c h t h e y fall,
t h e r e b y m a t c h i n g t h e i n f e r r e d A S T M n o m i n a l m a x i m u m a n d m a x i m u m sizes.
M c L e o d ' s d e f i n i t i o n of n o m i n a l m a x i m u m size is m o r e s p e c i f i c t h a n the
ASTM definition.
Within existing ASTM definitions nominal maximum and maximum sieve

s i z e m u s t be i n f e r r e d f r o m the A S T M g r a d a t i o n bands, o t h e r w i s e m o r e t h a n
one s i e v e w i l l m e e t the d e f i n i t i o n . C o n s i d e r G r a d a t i o n I. Without
c o m p a r i s o n to A S T M g r a d a t i o n b a n d s in D3515, t h e n o m i n a l m a x i m u m s i z e
c o u l d be 12.5 m m or 19.0 mm. B o t h s i e v e s c o u l d be d e f i n e d as t h e
s m a l l e s t s i e v e t h r o u g h w h i c h 100% is p e r m i t t e d to pass. M a x i m u m size
could 19.0 m m or larger depending on which sieve is d e s i g n a t e d as the

smallest sieve through which 100% of the m a t e r i a l is required to pass.
By themselves, the definitions of nominal m a x i m u m and m a x i m u m sieve size
are not specific enough to be useful. Only by applying the definitions
to the gradation bands in Table 1 of ASTM D3515, Standard Specification
for Hot-Mixed, H o t - L a i d Bituminous Paving Mixtures, can a m e a n i n g f u l
e v a l u a t i o n be done.
The d e f i n i t i o n as p r o p o s e d by McLeod is specific without requiring

c o m p a r i s o n to gradation bands. The resulting sieves specified by
McLeod's definition match those of ASTM when used with the gradation
limits in D3515.
MAXIMUM DENSITY LINES
The history of m a x i m u m density lines as used in the asphalt industry

is b a s e d on the work of L.W. Nijboer p u b l i s h e d in 1948 [5] and the work
of J. F. Goode and L.A. Lufsey published in 1962 [7].
Nijboer e v a l u a t e d gradations which p l o t t e d as straight lines on a log-

log chart of percent p a s s i n g and sieve opening in microns. Example
gradations shown in Figure 7 have the same m a x i m u m size with different
slopes, n. N i j b o e r compacted asphalt specimens for each gradation and
d e t e r m i n e d voids in mineral aggregate for each m i x t u r e which were
p l o t t e d against slope of the gradation line as shown in Figure 8.
N i j b o e r repeated his experiment for both crushed stone and gravel
m i x t u r e s with the same results. M i n i m u m aggregate voids occur when the
gradation slope on a log-log chart is 0.45.
In their 1962 publication Goode and Lufsey v e r i f i e d Nijboer's work

with gravel aggregates confirming that a straight line log-log gradation
with a slope of 0.45 p r o v i d e d the least voids in mineral aggregate. In
their paper they describe the mathematics to d e v e l o p a semi-power chart
for gradation. The vertical axis for percent p a s s i n g is an arithmetic
axis. The horizontal axis for sieve size opening e x p r e s s e d in microns
is the opening raised to a power of 0.45. This is the chart commonly
referred to as a "0.45 power" chart. Any straight drawn through the
origin to a m a x i m u m size will plot as a straight line with a slope of
0.45 on a log-log chart.
Goode and Lufsey used the 0.45 power chart they had d e v e l o p e d to
evaluate gradations which were not straight lines. They p e r f o r m e d a
series of experiments using contrived gradations which had different
levels of specific curves and bumps. For example they e v a l u a t e d the
gradations in Figure 9 which had different levels of a "hump" on the 600
~rn sieve. They e v a l u a t e d three groups of humped gradations. One group
was high in total sand, another m e d i u m and the third low in total sand.
r-
100.0
9p 1
,/
:~j n, VMA
. / . , I ~ , ~ 1 - 0.31, 14.3%
9 3 .I.,<:/,e/./~- 2 - 0.38, 12.9%
[] 4 3 - 0.45, 12.7%
10.0 '
4 - 0.52, 13.5%
O O 5 5 - 0.59, 14.9%
O. J~ 6 6 - 0.66, 16.5%
1.0 L
I 10 100 1000 10000 100000
Sieve Opening (microns
FIGURE 7 - - Gradations used to evaluate aggregate packing

(after Goode and Lufsey)
<
"'
zw"
" ~(.9
17.0
16.0
15.0
14.0
13.o
tii!iiiiiiiii
_Z~ 12.0
a~ 11.0
> 10.0
9.0
| ! | I
8.0
0.3 0.4 0.5 0.6 0.7
S L O P E OF G R A D A T I O N ON LOG LOG CHART
FIGURE 8 - - Relation of mixture VNA t o g r a d a t i o n slope

(after Goode and Lufsey)
100.0
90.0 .......................... VMAo ............................................................... .......................
80.0
70.0 .................. 9 - 1 5 . 3 % ........................................... J ........... 9 7

(.9
z 60.0 .............................10 - 1 5 . 0 % . . . . . . . . . . . . . . . . . . ~ ..................
(n 11 - 1 5 . 4 % ................~ ................................. ' ~ 8
(n 50.0
o.. ................................................... . ~ . . . . . . . . . . . . . . . . . 9 9
40.0
30.0 .................../ : ~ ,o
20.0
....................~ ................................................................................................................... 0 11
10.0
0.0
75 300 1.18 2.36 4.75 9.5 12.5 mm
150 600
SIEVE (0.45 power)
FIGURE 9 - - Simulated "humped" gradations (after Goode and Lufsey)
They also s t u d i e d "skip" gradations which were deficient in m a t e r i a l

b e t w e e n the 4.75 and 2.36 m m sieves. Goode and Lufsey o b s e r v e d that
"skip" gradations d i d not react the same way as hump gradations. They
s u g g e s t e d that "skip" gradations c o u l d p r o d u c e a denser m i x t u r e than a
straight line g r a d a t i o n which p l o t t e d nearly on top of the " m a x i m u m
d e n s i t y line". Figure 10 shows Goode and Lufsey g r a d a t i o n numbers 12,
22, 23 and 24 with the c o r r e s p o n d i n g voids in m i n e r a l aggregate. The
m a x i m u m d e n s i t y line as s u g g e s t e d by Goode and Lufsey is drawn through
the origin to the largest sieve size. G r a d a t i o n 12 which is n e a r l y on
top of their m a x i m u m density line has a much higher V M A than G r a d a t i o n
23 or 24, despite the fact that G r a d a t i o n 24 is d e f i n i t e l y the furthest
away from the m a x i m u m d e n s i t y line.
Goode and Lufsey c o n c l u d e d that "for skip gradations, low m i n e r a l

voids are a s s o c i a t e d with curves that stay a p p r e c i a b l y b e l o w the m a x i m u m
d e n s i t y line". Indeed they warn that "this fact must be taken into
account when using the m a x i m u m d e n s i t y line as a reference for a d j u s t i n g
skip gradations"[7].
Goode and Lufsey's work d e v e l o p i n g the "0.45-power" chart has g a i n e d

wide acceptance as a tool. One aspect though which has not been readily
a c c e p t e d is their recommendations for d r a w i n g a m a x i m u m d e n s i t y
reference line. In a written d i s c u s s i o n to their 1962 AAPT paper, Goode
and Lufsey s u g g e s t e d a procedure, but it has not been used in practice.
Several methods to draw m a x i m u m d e n s i t y lines are in use though no
p u b l i s h e d m a t e r i a l is available to suggest one m e t h o d is b e t t e r than
another [ii].
100.0
............ ,,, ,1 12 o~:1
...................................................... ~ i i ~ . ... .;. . ~.
90.0
80.0 9 22 ~
. . . . . . . . . . .-. . . . . . . . . . . . . . . . . . . . .-. . . . . . . . . . . . . . . . . . . . .-. . . . . . . . . . . . . . , ................ ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
jo . .............................................
70.0
Z 60.0 . . . . . . . . . . . . . . . . . a,..., m ......................................
In
In 50.0
~/..2".~ ~ .............. 12-15.0 Vo
"~ 40.0 ........................................................
.~ ~ ............ 22 -15.7% ...................
30.0 .................................i / ' ~ c a ' ................................................................. "" " ..... "...............

, . 2 ~ ' . . . - .... 24 -1:3.4% . . . , ~
20.0
10.0
0.0
75 300 1.18 2.36 4.75 9.5 12.5 19.0 mm
1 50 6 0 0
SIEVE (0.45 power)
FIGURE i0 - - Simulated "skip" gradations (after Goode a n d Lufsey)
EVALUATION OF EXISTING MIX DESIGN DATABASES
The q u e s t i o n of w h e r e to d r a w a m a x i m u m d e n s i t y line m u s t be a n s w e r e d
if the full u s e f u l n e s s of the 0.45 p o w e r chart is to be r e a l i z e d . An
e v a l u a t i o n was d o n e of t h r e e d a t a b a s e s u s i n g two m e t h o d s of d r a w i n g t h e
m a x i m u m d e n s i t y line as follows:
I. From the o r i g i n to 100% p a s s i n g the n o m i n a l maximum sieve size

2. From the o r i g i n to 100% p a s s i n g the maximum s i e v e size.
D e f i n i t i o n s of m a x i m u m a n d n o m i n a l m a x i m u m s i e v e size to be u s e d in
the a n a l y s i s are t h o s e p r o p o s e d b y M c L e o d [i0]. N o m i n a l m a x i m u m sieve
size is "one sieve l a r g e r t h a n the first s i e v e to r e t a i n m o r e t h a n 10%
of t h e a g g r e g a t e " . M a x i m u m s i e v e size w o u l d be one s i e v e l a r g e r t h a n
n o m i n a l m a x i m u m size.
D i s t a n c e f r o m the m a x i m u m d e n s i t y line is q u a n t i f i e d as t h e s u m of
a b s o l u t e d i f f e r e n c e b e t w e e n the m a x i m u m d e n s i t y line a n d t h e g r a d a t i o n
at e a c h of the s i e v e sizes. An e x a m p l e is s h o w n in T a b l e 8 for
G r a d a t i o n N u m b e r 22 w h i c h has a n o m i n a l m a x i m u m size of one h a l f inch.
The a b s o l u t e v a l u e of the d i s t a n c e f r o m t h e " m a x i m u m d e n s i t y line" is
c a l c u l a t e d a n d summed.
Difference = Absolute (Px - pmx)

where
Px = percent passing sieve "x" in e v a l u a t e d gradation
pm x = percent passing sieve "x" on m a x i m u m density line
TABLE 8 - - Example calculation of d i s t a n c e from density line
D r a w n to 100% N o m i n a l D r a w n to 100%
Maximum Sieve Maximum Sieve
Sieve Gradation, Density Density
Size % passing Line Difference Line Difference
19.0 m m I00.0 i00.0 0.0 I00.0 0.0
12.5 m m 98.0 I00.0 2.0 82.8 15.2
9.5 m m 84.0 88.4 4.4 73.2 10.8
4.75 m m 58.0 64.7 6.7 53.6 4.4
2.36 m m 52.0 47.2 4.8 39.1 12.9
1.18 m m 36.0 34.6 1.4 28.6 7.4
600 ~m 24.1 25.5 1.4 21.1 3.0
300 ~m 15.6 18.7 3.1 15.5 0.1
150 ~ m 9.3 13.7 4.4 11.3 2.0
75 ~m 4.7 i0.0 5.3 8.3 3.6

S u m of
Absolute 33.5 59.4
Differenc~
Re E v a l u a t i o n of G o o d e - Lufsev Data
A l l of the G o o d e a n d L u f s e y m i x t u r e s are c l a s s i f i e d as 12.5 m m n o m i n a l

size mixtures. D e n s i t y lines for t h e s e m i x t u r e s are d r a w n t h r o u g h f r o m
t h e o r i g i n to 100% p a s s i n g the 12.5 m m s i z e a n d f r o m the o r i g i n to the
19.0 m m size. T h e s e two lines are d e m o n s t r a t e d in F i g u r e II w h i c h shows
the "skip" gradations. C o n s i d e r the d i s t a n c e f r o m e a c h g r a d a t i o n to the
12.5 m m d e n s i t y line. G r a d a t i o n 12 is c l o s e s t to the line, yet V M A is
not the lowest. G r a d a t i o n 22 a p p e a r s to be a b o u t the s a m e a v e r a g e
d i s t a n c e a w a y but c r o s s e s the line. V M A for G r a d a t i o n 22 is h i g h e r .
G r a d a t i o n 23 is d e f i n i t e l y f u r t h e r a w a y f r o m the l i n e a n d G r a d a t i o n 24
is f u r t h e s t a w a y f r o m the 12.5 m m d e n s i t y line a n d the V M A is
progressively decreasing.
N o w c o n s i d e r the d i s t a n c e of e a c h g r a d a t i o n a w a y f r o m t h e 19.0 m m
line. G r a d a t i o n 22 is the f u r t h e s t a w a y a n d has the h i g h e s t VMA.
G r a d a t i o n 12, 23 a n d 24 are p r o g r e s s i v e l y c l o s e r a n d V M A b e c o m e s
p r o g r e s s i v e l y lower. If d i s t a n c e a w a y f r o m a d e n s i t y l i n e p r o v i d e s an
i n d i c a t i o n of d e n s i t y a c h i e v e d , t h e n the d e n s i t y l i n e d r a w n to the 19.0
m m s i e v e a p p e a r s to be the m a x i m u m d e n s i t y line for t h e s e m i x t u r e s .
D i s t a n c e a w a y f r o m the 12.5 m m a n d f r o m the 19.0 m m d e n s i t y line w e r e

c a l c u l a t e d for all h u m p e d a n d "skip" g r a d a t i o n s , s p e c i f i c a l l y G r a d a t i o n s
7 to 24. A r e g r e s s i o n a n a l y s i s was d o n e to r e l a t e V M A to d i s t a n c e f r o m
t h e m a x i m u m d e n s i t y lines as d r a w n to the n o m i n a l m a x i m u m s i z e a n d to
the m a x i m u m size. S c a t t e r p l o t s of d i s t a n c e f r o m the d e n s i t y l i n e d r a w n
t h r o u g h the n o m i n a l m a x i m u m size a n d d i s t a n c e f r o m the d e n s i t y l i n e
d r a w n t h r o u g h the m a x i m u m size are s h o w n in F i g u r e s 12 a n d 13 w i t h the
l i n e of b e s t fit i n d i c a t e d . C o r r e l a t i o n c o e f f i c i e n t s of R s q u a r e d are
0.004 a n d 0.915 r e s p e c t i v e l y for the two r e l a t i o n s h i p s .
100.0
90.0
80.0
oB ~o."
70.0
o
z 60.0
cn
< 50.0 ........| ~....~,:. ~ .............. VMA ................
........................... ................ 1 2 - i s 0yo
a. 40.0
..........." ' ~ ~ ............... 22 -15.7% ..............
30.0
20.0
10.0
0.0
75 300 1.18 2.36 4.75 9.5 12.5 19.0 mm
150 600 SIEVE (0.45 power)
FIGURE II Goode and Lufsey s i m u l a t e d "skip" gradations with trial

d e n s i t y reference lines.
19
W
O
LU 18
o
< 17
.-I
n-
'" 16 m
Z
B g
z_ 15
13 I I I I
20 30 40 50 60 70
Distance From Density Line
T h r o u g h N o m i n a l M a x i m u m Size
FIGURE 12 - Relation of V M A to distance from d e n s i t y
- line drawn
t h r o u g h the nominal m a x i m u m size for Goode and Lufsey database
244 ASPHALT MIXTURE P E R F O R M A N C E
19
.'I
0
UJ
n,, 18
O
< 17
J
I.u
Z 16
X
-- 15
> 14
13 I I I
20 40 60 80 100 120
Through Maximum Size
FIGURE 13 - - R e l a t i o n of V M A to d i s t a n c e f r o m d e n s i t y l i n e d r a w n
through the m a x i m u m s i z e for G o o d e a n d L u f s e y d a t a b a s e
LeFebvre Data Set E v a l u a t i o n
A set of m i x d e s i g n s was d o n e in the 1 9 5 0 ' s by J. L e F e b v r e to e v a l u a t e

t h e e f f e c t of g r a d a t i o n a n d a g g r e g a t e t y p e on m i x t u r e p r o p e r t i e s [12].
A t o t a l of 41 m i x d e s i g n s w e r e d o n e u s i n g c r u s h e d g r a v e l a n d c r u s h e d
q u a r r i e d a g g r e g a t e w i t h the same g r a d a t i o n , n a t u r a l s a n d a n d c r u s h e r
screenings with the same gradation, fine dune sand and limestone mineral
filler. A v a r i e t y of p r o p o r t i o n s of the d i f f e r e n t a g g r e g a t e s w e r e
b l e n d e d to p r o v i d e d i f f e r e n t g r a d a t i o n s on w h i c h a c o m p l e t e 75 b l o w m i x
d e s i g n was done. Of t h e s e d e s i g n s , 26 m i x t u r e s h a d a n o m i n a l m a x i m u m
s i z e of 19.0 mm, 10 h a d a n o m i n a l m a x i m u m s i z e of 12.5 m m a n d 5 h a d a
n o m i n a l m a x i m u m size of 4.75 mm. The d e s i g n a t i o n of n o m i n a l m a x i m u m
size is b a s e d on t h e d e f i n i t i o n p r o p o s e d in t h i s p a p e r , t h a t is one
s i e v e l a r g e r t h a n the first s i e v e to r e t a i n m o r e t h a n 10% m a t e r i a l .
F r o m e a c h of t h e m i x d e s i g n s m i n i m u m V M A was s e l e c t e d . D i s t a n c e was
c a l c u l a t e d f r o m d e n s i t y l i n e s d r a w n t h r o u g h the m a x i m u m s i e v e size a n d
the n o m i n a l m a x i m u m s i e v e size. T h i s m e t h o d is t h e same as s h o w n for
t h e G o o d e a n d L u f s e y g r a d a t i o n s a l t h o u g h t h e set of s i e v e s u s e d by
L e F e b v r e was d i f f e r e n t . L e F e b v r e u s e d a s i e v e set w h i c h i n c l u d e d 2.00
mm, 850 ~m, 4252 ~m, 180 ~ m a n d 75 ~ m s i e v e s u n l i k e G o o d e a n d L u f s e y who
u s e d a set w h i c h i n c l u d e d 2 . 3 6 mm, 1.18 mm, 600 ~m, 300 ~m, 150 ~ m a n d
75 ~ m sieves. F o r the l a r g e r s i e v e s b o t h u s e d 4.75 mm, 9.5 mm, 12.5 m m
a n d 19.0 mm.
R e s u l t s of a r e g r e s s i o n a n a l y s i s for the 19.0 m m n o m i n a l m a x i m u m size

g r a d a t i o n s are s h o w n in F i g u r e s 14 a n d 15. V M A c o r r e l a t e d to d i s t a n c e
f r o m the d e n s i t y l i n e t h r o u g h t h e n o m i n a l m a x i m u m s i z e w i t h an R - s q u a r e d
of 0.232 a n d to d i s t a n c e f r o m the d e n s i t y line t h r o u g h the m a x i m u m size
w i t h an R - s q u a r e d of 0.815.
22
UJ
t-.
0
W 20 i'"
n-
O
0
'~ 18
.J
n-
m 15
Z
9 .- l_
8 I I I I I I I
30 40 50 60 70 80 90 100 1 0
Through Nominal Maximum Size
FIGURE 14 - - R e l a t i o n of V M A to d i s t a n c e f r o m d e n s i t y l i n e d r a w n
t h r o u g h the n o m i n a l m a x i m u m size for L e F e b v r e d a t a b a s e
la 22
I,-
'"
n- 20 s I
18
.J 9 ~'" . 2"
n- 9 9 .%
LU
Z 15
Z
-- 12
r
5
> 10 .J
8 I I I I
25 50 75 100 125 150
FIGURE 15 R e l a t i o n of V M A to d i s t a n c e f r o m d e n s i t y l i n e d r a w n
t h r o u g h t h e m a x i m u m s i z e for L e F e b v r e d a t a b a s e
F~WA Demonstration ProSect. Field Manaqement of Asphalt Mixes
Data from Goode and Lufsey and from LeFebvre show a r e l a t i o n s h i p

exists b e t w e e n V M A and a d e n s i t y reference line d r a w n through the
m a x i m u m sieve size. If a r e l a t i o n s h i p exists b e t w e e n distance from a
d e n s i t y reference line and VMA, would q u a l i t y control i n f o r m a t i o n
p r o v i d e a suitable data base? Q u a l i t y control data that includes
g r a d a t i o n as well as volumetric p r o p e r t i e s of c o m p a c t e d specimens w o u l d
p r o v i d e a database.
The Federal H i g h w a y A d m i n i s t r a t i o n ' s D e m o n s t r a t i o n Project 74 has been

c o l l e c t i n g q u a l i t y control data on p a v i n g projects [13]. A sampling of
projects each with a significant number of data points was evaluated.
The results showed a poor c o r r e l a t i o n b e t w e e n V M A and d i s t a n c e from
d e n s i t y reference line drawn through both nominal m a x i m u m and m a x i m u m
sieve size. Figures 16 and 17 show scatter plots of d i s t a n c e from the
line drawn through the nominal m a x i m u m size and m a x i m u m size for a
p r o j e c t with 37 data points. The regression c o e f f i c i e n t s for each are
0.001 a n d 0.208 r e s p e c t i v e l y for the two d e n s i t y lines.
A r e l a t i o n s h i p b e t w e e n distance from the d e n s i t y line t h r o u g h the

m a x i m u m size does not seem as firm as for the Goode and L u f s e y or the
LeFebvre database. C o n s i d e r the data shown in Figure 16 and 17 compared
to either the G o o d e / L u f s e y or the LeFebvre data. The range of VMA's is
very small. Most data points fall b e t w e e n 12.5 to 13.5%. Also the
range of distances from the d e n s i t y reference lines is quite narrow.
Hence scatter which is inherent in m e a s u r e m e n t of V M A b e c o m e s
significant and reduces the c o r r e l a t i o n coefficient.
14.5
uJ
I-
(.9
I,,g
""
(3 14.0
(3
,,r
...J
,,,r
,',- 13.5 9 9 -..,. - ~
I,r
Z
z 13.0
I I i I a
CO
,"t
12.5
15 20 25 30 35
Through Nominal Maximum Size
FIGURE 16 Relation of VMA to distance from d e n s i t y line drawn
t h r o u g h the nominal m a x i m u m size for FHWA D e m o n s t r a t i o n Project 74
database
14.5
I,M
I--
<~
I,,U
14.0
s
.,..I
13.5
I,M
_e
Zm
13.0
U)
r 2-';" ""
5
12.5
I ' I ' I ' ~ ~ I
30 40 50 60 70

FIGURE 17 - - R e l a t i o n of V M A to d i s t a n c e f r o m d e n s i t y line d r a w n
through the m a x i m u m size for F H W A D e m o n s t r a t i o n P r o j e c t 74 d a t a b a s e
A n o t h e r s t a t i s t i c a l m e a s u r e can be u s e d to d e t e r m i n e if the c h a n g e in
d i s t a n c e f r o m a d e n s i t y r e f e r e n c e line c o r r e s p o n d s to a c h a n g e in VMA.
U s i n g an F-test, an 87% p r o b a b i l i t y e x i s t s that v a r i a t i o n of V M A as
c o m p a r e d to d i s t a n c e f r o m the d e n s i t y r e f e r e n c e line t h r o u g h n o m i n a l
m a x i m u m size is r a n d o m error. F - t e s t r e s u l t s for d i s t a n c e f r o m a
d e n s i t y r e f e r e n c e line t h r o u g h the m a x i m u m size i n d i c a t e s that o n l y 0.5%
p r o b a b i l i t y e x i s t s that the c h a n g e in V M A is r e l a t e d to r a n d o m error.
So a l t h o u g h the c o r r e l a t i o n is low, p r o b a b i l i t y is s t r o n g that a
r e l a t i o n s h i p e x i s t s b e t w e e n d i s t a n c e f r o m a d e n s i t y line d r a w n t h r o u g h
the m a x i m u m size a n d v o i d s in the m i n e r a l a g g r e g a t e .
EXAMPLE USE OF MAXIMUM DENSITY REFERENCE LINE
The p u r p o s e of d e t e r m i n i n g if V M A can be r e l a t e d to d i s t a n c e f r o m a
d e n s i t y r e f e r e n c e line is to p r o v i d e a tool w h i c h c a n be u s e d d u r i n g m i x
d e s i g n to a d j u s t g r a d a t i o n w i t h a v i e w to i n c r e a s i n g VMA. An a c t u a l
e x a m p l e is d i s c u s s e d below.
In the e a r l y s t a g e s of a m i x t u r e d e s i g n s t o c k p i l e s of a g g r e g a t e s a n d a
m a s t e r g r a d a t i o n are selected. The q u e s t i o n f a c e d b y the d e s i g n e r is
" W h e r e s h o u l d the g r a d a t i o n be p l a c e d w i t h i n the m a s t e r b a n d to o b t a i n
a c c e p t a b l e V M A f r o m the d e s i g n ? " . A i r v o i d s in the d e s i g n m i x t u r e can
be e a s i l y c h a n g e d b y a d j u s t i n g a s p h a l t content. V M A on t h e o t h e r h a n d
is r e l a t i v e l y i n s e n s i t i v e to a s p h a l t c o n t e n t a n d m u s t be c o n t r o l l e d
p r e d o m i n a n t l y b y gradation. If a careful s e l e c t i o n of g r a d a t i o n is not

made initially, a m i x design can be done, with specimens c o m p a c t e d at
five asphalt contents and volumetric properties d e t e r m i n e d only to find
the V M A can not be achieved. The only course of action then is to
d i s r e g a r d several days of work, change the g r a d a t i o n and redo the
design.
A v a i l a b l e p u b l i s h e d guidance to s e l e c t i n g initial g r a d a t i o n is
t y p i c a l l y vague advice such as "choose a g r a d a t i o n somewhere near the
center of the gradation band". A p r e f e r a b l e a p p r o a c h is to compact
t r i p l i c a t e specimens at one asphalt content, m e a s u r e V M A a n d evaluate
the results u s i n g the m a x i m u m d e n s i t y line. Several gradations can be
s e l e c t e d from a b l e n d i n g analysis and e v a l u a t e d at the same time.
The m a i n purpose of the g r a d a t i o n e v a l u a t i o n is to d e t e r m i n e whether a

g r a d a t i o n which meets s p e c i f i e d limits will meet the voids in m i n e r a l
a g g r e g a t e requirements without doing an entire m i x t u r e design. Using
stockpiles selected by the designer, a b l e n d i n g analysis is done to
d e t e r m i n e which gradation(s) will be used in the volumetric
p r o p o r t i o n i n g phase. Results of a b l e n d i n g analysis done at the Asphalt
Institute are p l o t t e d on a 0.45 power g r a d a t i o n analysis chart shown in
Figure 18. For each g r a d a t i o n triplicate specimens are m i x e d and
c o m p a c t e d at a single asphalt content. A t h e o r e t i c a l m a x i m u m specific
g r a v i t y test is done to obtain the aggregate e f f e c t i v e specific gravity.
A g g r e g a t e bulk specific g r a v i t y has been d e t e r m i n e d previously. Then
air voids and voids in the mineral aggregate are calculated.
9 Grad A
100.0
" Grad B ~,~...,~..s......~~r ' /
90.0
80.0 9 .. 9 Grad C ............................................ .'../.~..:-../,,~
o . B er J ~B,
...................................
7 0 . 0
..... - . . . . . . . Limits ................ . - ~ ...... .................................
Z
60.0
(n . . . . . . . . Limits o" ~:r ~ ~.,,.,, - M axim u m
(n 50.0
a. 40.0 ,, ..~:.
30.0
20.0 9................................
- ~".~".~"
fA~o ~ . 9~..~ .....................................................................................................
10.0
0.0
75 300 1.182.36 4.75 9.5 12.5 19 25 m m
150600
SIEVE (0.45 power)
FIGURE 18 - - Example gradations for aggregate VMA e v a l u a t i o n
R e s u l t s of the g r a d a t i o n e v a l u a t i o n are t a b u l a t e d in Table 9. The

estimate of V M A at 4% air voids is made b a s e d on w h e t h e r additional
asphalt is n e e d e d or whether the asphalt content needs to be reduced to
achieve 4% air voids. For example, G r a d a t i o n A has 3.1% air voids. To
obtain 4% air voids with this m i x the asphalt content must be slightly
reduced. The reduction is small enough that V M A will likely not be
effected. G r a d a t i o n B will require a larger reduction in asphalt
content to achieve 4% air voids which will increase the V M A slightly.
G r a d a t i o n C requires additional asphalt b i n d e r to reduce air voids to 4%
which m a y slightly reduce the VMA.
TABLE 9 - - Void properties of trial g r a d a t i o n s
Distance Air Void.9 VMA Estimated Required
Gradation From MDL % % V M A @ 4% VMA@ 4%
62.1 3.1 12.2 12.2 13.0
44.0 2.6 11.7 11.9 13.0
81.9 4.7 13.7 13.6 13.0
This m i x t u r e which has a 19.0 m m nominal m a x i m u m size will require a

V M A of 13% at 4% air voids. Of the trial gradations only G r a d a t i o n C
will be acceptable for c o n s i d e r a t i o n in the full v o l u m e t r i c
p r o p o r t i o n i n g analysis. Therefore, the m i x t u r e d e s i g n e r can continue
with G r a d a t i o n C k n o w i n g that a successful design is likely. Note that
in Figure 18, G r a d a t i o n C is further from the d e n s i t y line than
Gradations A and B.
If none of the trial gradations provide adequate VMA, another one or

two trial gradations can be selected which are further from the d e n s i t y
lineo Specimens can be c o m p a c t e d for these additional gradations to
estimate V M A at 4% air voids.
CONCLUS IONS
C o n c l u s i o n s which can be made b a s e d on the c o n s i d e r a t i o n s given in

this p a p e r are as follows:
i. The m a x i m u m d e n s i t y line on a 0.45-power chart should be drawn

t h r o u g h the origin to 100% p a s s i n g m a x i m u m sieve size.
2. The d e f i n i t i o n of nominal m a x i m u m size should be "in a standard

set of sieves, nominal m a x i m u m size is one size larger than the first
sieve to retain more than ten percent of the total aggregate".
3. M a x i m u m size should be d e f i n e d as "the smallest sieve o p e n i n g

t h r o u g h which the entire amount of aggregate is r e q u i r e d to pass, one
sieve size larger than nominal m a x i m u m size".
4. For a given gradation, aggregate surface c h a r a c t e r i s t i c s of

a n g u l a r i t y and surface texture influence the p a c k i n g of aggregate
particles. R o u n d e d gravels p r o d u c e mixtures with a p p r o x i m a t e l y one
percent lower voids in mineral aggregate than c r u s h e d limestone.
5. C o m p a r i s o n of gradations to the m a x i m u m d e n s i t y line can be done

only for the same aggregates. Comparison to a m a x i m u m d e n s i t y line is
d e s i g n e d to evaluate gradation. The effect of surface c h a r a c t e r i s t i c s
on V M A must be d e t e r m i n e d by c o m p a r i n g mixtures with the same gradation.
6. C o m p a r i s o n of aggregates with large differences in minus 75 ~ m

m a t e r i a l m a y not be valid. Each m i x t u r e database c o n t a i n e d gradations
with similar percent p a s s i n g the 75 ~m sieve. C o m p a r i s o n s of mixtures
with large differences in the minus 75 ~m sieve u s i n g d i s t a n c e f r o m the
d e n s i t y line m a y not be possible.
ACKNOWLEDGMENTS
The authors wish to acknowledge the support of the m e m b e r companies of

the Asphalt Institute. Without their support and interest, the research
and e x p e r i m e n t a t i o n p r e s e n t e d here could not have been done.
REFERENCES
[i] Goetz, W.H., "The Evolution of Asphalt C o n c r e t e M i x Design",

Asphalt Concrete M i x DesiQn: Development of More Rational
Approaches, American Society for Testing and M a t e r i a l s STP 1041,
Philadelphia, P e n n s y l v a n i a 1989.
[2] Hubbard, P. and Field, F.C., "Stability and R e l a t e d Tests for

Asphalt P a v i n g Mixtures", The Asphalt Institute. R e s e a r c h Series
N u m b e r I, O c t o b e r 1935.
[3] Hubbard, P. and Field, F.C., "Adaptation of the S t a b i l i t y Test to

Include Coarse Aggregate Asphalt Paving M i x t u r e s , " P r o c e e d i n g s of
the Technical Sessions of the A s s o c i a t i o n of AsPhalt P a v i n g
Technoloaists, pp. 109 - 114, January, 1932, Detroit, Michigan.
[4] Vokac, R., "Studies in the P r o p o r t i o n i n g of Low Cost Bituminous

Mixtures of Dense Graded A g g r e g a t e Type," Proceedings of the
Technical Sessions The A s s o c i a t i o ~ of Asphalt Paving Technologists,
pp. 84 - 102, December, 1932, N e w Orleans, Louisiana.
[5] Nijboer, L. W., "Plasticity as a Factor in the Design of Dense

Bituminous Road Carpets," Elsevier, 1948.
[6] Hudson, S. B. and Davis, R. L., " R e l a t i o n s h i p of A g g r e g a t e Voidage

to Gradation," Proceedings of The A s s o c i a t i o n of AsDhalt Paving
Technologists, pp. 574 - 593, Vol. 34, 1965, Philadelphia,
Pennsylvania.
[7] Goode, J. F. and Lufsey, L. A., "A new G r a p h i c a l Chart for

E v a l u a t i n g A g g r e g a t e Gradations," proceedings of The A s s o c i a t i o n of
Asphalt Pavin~ TeGhDologist~, pp. 176 - 207, Vol. 31, 1962, N e w
Orleans, Louisiana.
[8] Mix DesiQn Methods for AsPhalt Concrete and Other Hot Mix Types,
Manual Series No. 2, 1984 Edition, The Asphalt Institute
[9] Storer, D. A., Ed., Compilation of ASTM Standard Definitions,

Seventh Edition, 1990, American Society for T e s t i n g and Materials,
Philadelphia, Pennsylvania.
[10] McLeod, N. W., "Design of Dense Graded Asphalt Concrete Pavements,"

Canadian Technical Asphalt Association Proceedinas, pp. 352 - 404,
Vol. XXXII, November 1987, Toronto, Ontario.
[11] Turnham, N., "Gradation and Use of 0.45 Power Paper," Technical
Ouarterly, Texas State Department of Highways and Public
Transportation, Vol. 6, Issue 2, March 1991.
[12] LeFebvre, J., "Recent Investigations of Design of Asphalt Paving

Mixtures," Proceedings of The Association of Asphalt Paving
Technologists, pp. 321 - 394, Vol. 26, 1957, Atlanta, Georgia.
[13] D'Angelo, J. A. and Ferragut, T., "Summary of Simulation Studies

F r o m D e m o n s t r a t i o n Project NO. 74: Field Management of Asphalt
Mixes", pxoceedings of The Association of Asphalt PavinQ
Technologists~ Vol. 60, 1991, Seattle, Washington.
Byron E. Ruth l, Xuezhen Shen 2, and L i - H a n g Wang 3
GYRATORY EVALUATION OF AGGREGATE R[.RNDS T O D E T E R M I N E THEIR EFFECT ON

SHEAR RESISTANCE AND SENSITIVITY TO ASPHALT CONTENT
REFERENCE: Ruth, B.E., Shen, Xuezhen, and Wang, L-H, "Gyratory Evalua-
tion of A g g r e g a t e Blends to Determine Their Effect on Shear R e s i s t a n c e
and S e n s i t i v i t y to Asphalt Content," Effects of A q q r e q a t e s and Mineral
Fillers on Asphalt Mixture Performance, A S T M STP 1147, R i c h a r d C.
Meininger, Editor; A m e r i c a n Society for Testing and Materials, Phila-
delphia, 1992.
ABSTRACT: Transverse and longitudinal shoving of asphalt p a v i n g mix-

tures is p r i m a r i l y related to the inability of current test m e t h o d s to
identify the properties of mixtures at different levels of traffic
d e n s i f i c a t i o n and the influence that aggregate p r o p e r t i e s have on the
mixtures' resistance to plastic d e f o r m a t i o n (shoving). R e s e a r c h was
conducted to investigate the use of the air roller equipped G y r a t o r y
T e s t i n g Machine for evaluation of asphalt mixtures containing aggregates
from the same sources with varying amounts of crushed and natural sand
aggregates (different blend proportions) to produce both identical gra-
dations as well as different gradations. Analyses p e r f o r m e d on the test
results indicated overall excellent correlation b e t w e e n G y r a t o r y shear
(Gs) after 200 revolutions of d e n s i f i c a t i o n and asphalt content (A/C),
voids in mineral aggregates (VMA), and air void content (AV) at as-com-
pacted densities. Similarly, statistical regression analyses indicated
that asphalt, mineral filler, fine aggregate, and coarse a g g r e g a t e con-
tents have a significant effect on the Gs value for the different mix-
tures. As a result it is concluded that V M A and AV contents are not ab-
solutely essential for m i x t u r e design or m i x t u r e evaluation. It is sug-
gested that the GTM air roller procedure for testing of asphalt mixtures
is p r o b a b l y all that is needed for evaluation of a m i x t u r e ' s r e s i s t a n c e
to plastic deformation.
KEYWORDS: asphalt mixtures, aggregate blends, plastic deformation, gyra-

tory t e s t i n g machine, rutting
STATEMENT OF THE PROBLEM
Historically, test methods for flexible pavement m a t e r i a l s have

been developed with the intent of p r o v i d i n g an assessment of the m a t e -
I Dr. Ruth is a p r o f e s s o r in the Department of Civil Engineering, Univer-

sity of Florida, 345 Well Hall, Gainesville, FL 32611.
2:3 Ms. Shen and Mr. Wang are graduate research assistants in the Department
of Civll Engineering, U n i v e r s i t y of Florida, 345 weil Hall, Gainesville,
FL 32611.
252
Copyright by ASTM Int'l (all rights reserved); Thu Dec 22www.astm.org

07:29:10 EST 2011
RUTH ET AL. ON GYRATORY EVALUATION OF AGGREGATE BLENDS 253
rial's properties as related to its a s - c o n s t r u c t e d or insitu perfor-

mance. Some of these tests have been adopted or d e v e l o p e d for the ex-
p r e s s e d p u r p o s e of asphalt m i x t u r e design. Basically, m i x t u r e design
m e t h o d s such as the California (Hveem) p r o c e d u r e and the Marshall method
have p r o v i d e d reasonably good estimates of design asphalt contents.
However, it is g e n e r a l l y accepted that n e i t h e r m e t h o d is t o t a l l y ade-
quate from the standpoint of p r e v e n t i n g plastic d e f o r m a t i o n of the pave-
ment's surface w h e n subjected to heavy wheel loads and high tempera-
tures. This p r o b l e m is further aggravated by poor q u a l i t y control and
lack of knowledge regarding the properties of aggregate blends.
The reason there has not been a greater o c c u r r e n c e of plastic de-
formation properties is that each State D.O.T. or highway agency has ex-
tensive experience in the use of their aggregates. Consequently, this
knowledge has t r a n s l a t e d into specifications, test procedures, and mix-
ture design criteria that generally provide good p e r f o r m a n c e of asphalt
paving mixtures. Unfortunately, transverse and longitudinal s h o v i n g
(plastic d e f o r m a t i o n rutting) p r o b l e m s are still e n c o u n t e r e d b e c a u s e of
insufficient knowledge regarding the behavior and p r o p e r t i e s of mixtures
w i t h v a r i a b l e aggregate blends (angularity, surface texture, gradation,
etc.) throughout their d e n s i f i c a t i o n history from a s - c o m p a c t e d to fully
traffic densified. Current and more recent efforts to solve the plastic
d e f o r m a t i o n p r o b l e m by using m o d i f i e d asphalts or a supposedly new con-
cept (e.g., stone mastic mixtures) skirts the real problem.
In summary, the somewhat archaic approach to the selection of ag-
gregate blends for asphalt paving mixtures, the lack of suitable test
m e t h o d s and procedures for mixture design and for evaluation, and the
t e n d e n c y to change asphalt content rather than aggregate b l e n d in prob-
lem mixtures has contributed s u b s t a n t i a l l y to the p r e v a l e n c e of exces-
sive plastic deformation. What is needed to correct this problem? Ob-
viously, a change in some of our concepts, including greater concern
over aggregate blend properties, the development and adoption of im-
p r o v e d test methods, and greater emphasis on the formulation of mixtures
(Job Mix Formulas) that are not sensitive to minor changes in gradation,
blending, or asphalt content that c o n v e n t i o n a l l y occur during p r o d u c t i o n
of the hot-mix.
INTRODUCTION
A search for improved test methods for e v a l u a t i o n of asphalt mix-

tures and d e t e r m i n a t i o n of the effects of aggregates and mineral fillers
on the p e r f o r m a n c e of asphalt paving mixtures would reveal that both
gyratory and rolling wheel load test machines have potential applica-
tion. There are a v a r i e t y of rolling wheel load devices w h i c h may or
may not produce results that relate directly to the rutting resistance
(performance) of asphalt pavements. Assuming there is good correlation
between pavement rut depth and that achieved by a rolling wheel load
t e s t i n g device, it w o u l d provide the means for e v a l u a t i o n of mix de-
signs. However, other than the rut depth measurement, it does not yield
any other useful parameters that w o u l d assist in how to redesign a poor
mix. Conversely, gyratory compaction devices, in particular, the Corps
of Engineers Gyratory Testing Machine (GTM), as m a n u f a c t u r e d by the En-
gineering D e v e l o p m e n t s C o r p o r a t i o n in Vicksburg, Mississippi, has numer-
ous features w h i c h may be b e n e f i c i a l l y used for the purpose of evalua-
tion and/or design of asphalt mixtures either in the laboratory or in
the field at the hot-mix plant.
O r i g i n a l l y the GTM was used p r i m a r i l y with the fixed roller al-
though eventually procedures evolved for the compaction and shear prop-
erties of bituminous mixtures using the oil-filled roller (ASTM D3387-
83) w i t h the capability of m e a s u r i n g roller pressure during testing at a
given angle of gyration (fixed shear strain). R e s e a r c h using the air
roller w h i c h provides a variable stress and variable shear strain test-
ing capability has been very limited. The air-roller e q u i p p e d model 4C
G T M has been used to simulate field compaction of asphalt concrete and
of s o i l s i n c l u d i n g clay, silt, gravel, a n d c r u s h e d s t o n e [1,2]. An air-

r o l l e r d e s i g n p r o c e d u r e for a s p h a l t c o n c r e t e m i x t u r e s w a s d e v e l o p e d to
s i m u l a t e t r a f f i c d e n s i f i c a t i o n [3].
M o r e r e c e n t l y t h e a i r - r o l l e r e q u i p p e d m o d e l 6 B - 4 C G T M has b e e n
u s e d to e v a l u a t e a s p h a l t m i x t u r e s u s e d in F l o r i d a [4,5,6] and m i x t u r e s
in t h e A s p h a l t - A g g r e g a t e M i x t u r e A n a l y s i s S y s t e m (AAMAS) w h i c h w e r e c o m -
p a r e d to t e s t r e s u l t s of t h e o i l - f i l l e d r o l l e r [7]. These investiga-
t i o n s r e s u l t e d in t h e d e v e l o p m e n t of s t a n d a r d G T M s e t t i n g s for b o t h i00
m m and 150 m m d i a m e t e r t e s t s p e c i m e n s (62 and i00 m m high, r e s p e c -
tively):
A. I00 m m d i a m e t e r x 62 m m h i q h specimens
I n i t i a l a n g l e of g y r a t i o n = 3 d e g r e e s
R a m p r e s s u r e = 690 kPa
I n i t i a l air r o l l e r p r e s s u r e = 62 kPa
N u m b e r of r e v o l u t i o n s for:
c o m p a c t i o n at h o t - m i x t e m p e r a t u r e s = 18
d e n s i f i c a t i o n at 60C = 250 to 300
B. 150 m m d i a m e t e r x I00 m m h i q h specimens
I n i t i a l a n g l e of g y r a t i o n = 3 d e g r e e s
R a m p r e s s u r e = 690 k P a
I n i t i a l a i r - r o l l e r p r e s s u r e = 276 k P a
N u m b e r of r e v o l u t i o n s : s a m e as for I00 mm dia. specimens
T h i s t e c h n i c a l p a p e r p r e s e n t s t h e r e s u l t s of G T M e v a l u a t i o n of
m i x t u r e s c o n t a i n i n g d i f f e r e n t b l e n d s of a g g r e g a t e s f r o m t h e s a m e
sources. R e g r e s s i o n a n a l y s e s are u s e d to s h o w t h e r e l a t i o n s h i p s b e t w e e n
t h e G y r a t o r y s h e a r (Gs) v a l u e and p h y s i c a l p r o p e r t i e s of t h e m i x t u r e s .
MATERIALS AND TEST PROCEDURES
T w o d i f f e r e n t s o u r c e s of a g g r e g a t e w e r e s e l e c t e d for p r e p a r a t i o n
of a g g r e g a t e b l e n d s to be u s e d in the a s p h a l t m i x t u r e for t h i s i n v e s t i -
gation. A F l o r i d a l i m e s t o n e c o n s i s t i n g of t w o c o a r s e a g g r e g a t e sizes,
S - I - A and S - I - B ( s i m i l a r to No. 67 and No. 89 a g g r e g a t e s ) a n d s c r e e n i n g s
w e r e u s e d in d i f f e r e n t c o m b i n a t i o n s to p r e p a r e a g g r e g a t e b l e n d s for
F l o r i d a D.O.T. s t r u c t u r a l m i x t u r e s d e s i g n a t e d as S-I and S - I I I w h i c h
h a v e 19 m m and 12.5 m m m a x i m u m a g g r e g a t e sizes, r e s p e c t i v e l y . T h e s e ag-
g r e g a t e b l e n d s w e r e a d j u s t e d u s i n g 20 p e r c e n t n a t u r a l s a n d r e p l a c e m e n t
for t h e S-I m i x t u r e s and i0 to 20 p e r c e n t n a t u r a l s a n d in t h e S - I I I m i x -
tures. A p p r o x i m a t e l y 90 to i00 p e r c e n t of t h e n a t u r a l s a n d p a s s e d t h e
No. 40 s i e v e (0.42 mm). Consequently, an e f f o r t w a s m a d e to m a i n t a i n
the g r a d a t i o n of t h e p l u s No. i0 (2.00 mm) f r a c t i o n as c l o s e as p o s s i b l e
to t h o s e m i x t u r e s w i t h o u t n a t u r a l sand. In all cases, t h e m i x t u r e s c o n -
f o r m e d to a g g r e g a t e g r a d a t i o n s u s e d in a c t u a l c o n s t r u c t i o n p r o j e c t s ex-
cept for c h a n g e s in m i n e r a l f i l l e r c o n t e n t (percent p a s s i n g 0 . 0 7 4 m m
s i e v e o p e n i n g ) and a s p h a l t content.
T a b l e 1 g i v e s t h e a g g r e g a t e b l e n d s t h a t w e r e u s e d in p r e p a r i n g t h e
h o t - m i x samples.
T h e g r a d a t i o n s for the S-I m i x t u r e s are p r e s e n t e d in T a b l e 2 f o r
the different mixtures. M i x t u r e s w i t h 1.5 p e r c e n t a d d i t i o n a l m i n e r a l
f i l l e r (coded 3 and 4) r e s u l t e d in m i n e r a l f i l l e r c o n t e n t s r a n g i n g f r o m
3.3 to 4.0 p e r c e n t .
G r a d a t i o n s for t h e S-III m i x t u r e s are g i v e n in T a b l e s 3 and 4 for
a g g r e g a t e b l e n d s w i t h o u t and w i t h n a t u r a l sand, r e s p e c t i v e l y . The major
d i f f e r e n c e in t h e s e b l e n d s o t h e r t h a n n a t u r a l s a n d c o n t e n t is t h e a m o u n t
of m i n e r a l filler, c o a r s e a g g r e g a t e (S-I-B), and f i n e a g g r e g a t e p a s s i n g
t h e 2 . 0 0 m m (No. I0) sieve.
TABLE 1--Aggregate blends to m e e t JMF* (S-I)
% Mixtures Mixtures Mixtures

Content A1 A2 A3 A4 B1 B2 B3 B4 Cl C2 C3 C4
S-I-A 20 20 20 20 12 12 12 12 15 15 15 15
S-I-B 45 45 45 45 43 43 43 43 35 35 35 35
Asphalt 35 15 35 15 45 25 45 25 50 30 50 30
Screen-
ing
Nat. 0 20 0 20 0 20 0 20 0 20 0 20
Sand
MF** 0 0 1.5 1.5 0 0 1.5 1.5 0 0 1.5 1.5
* JMF = Job Mix Formula

** MF = Mineral Filler
TABLE 2--Wet sieve aggregate gradations for S-I m i x t u r e s
Percent Passing
Sieve S-I M i x . - 0 % Nat. Sand S-I M i x . - 2 0 % Nat. S a n d ~')

mm
A1 B1 Cl A2 B2 C2
19.0 m m (3/4") i00 i00 i00 i00 I00 i00

12.5 m m (1/2") 98.8 i00 97 98.8 i00 97
9.50 mm(3/8") 87.0 93.7 90 85.5 94 90
4.76 (#4) 54.4 62.5 65 55.1 65 65

2.00 (#i0) 33.1 43.7 42 36.3 48 46
0.42 (#40) 13.0 25.4 26.6 25.4 34 36

0.177 (#80) 6.1 11.7 11.2 10.8 12 17
0.074 (#200)** 2.5 2.0 1.9 2.5 1.8 2.0
(*) A p p r o x i m a t e l y 90% to 100% Passing No. 40 Sieve, ** M i n e r a l Filler
A l l m i x t u r e s w e r e p r e p a r e d u s i n g an A C - 3 0 w h i c h c o n f o r m e d to F D O T
specifications. A s p h a l t c o n t e n t s w e r e s e l e c t e d on t h e b a s i s of c o n -
s t r u c t i o n p r o j e c t JMF (based on 50 b l o w M a r s h a l l ) by u s i n g 0.5 p e r c e n t
b e l o w and at least one 0.5 p e r c e n t i n c r e m e n t a b o v e t h e o r i g i n a l d e s i g n
asphalt content.
T e s t p r o c e d u r e s for g r a d a t i o n i n v o l v e d c h e c k i n g a g g r e g a t e b l e n d s
by wet s i e v e a n a l y s e s , by e x t r a c t i o n b e f o r e or a f t e r c o m p a c t i o n , and by
extraction after densification. Data from the wet sieve analyses were
u s e d for S-I m i x t u r e s since e x t r a c t i o n d a t a for a s - c o m p a c t e d s a m p l e s
w e r e not a v a i l a b l e . T h e m a x i m u m d e n s i t y of t h e m i x t u r e s w a s o b t a i n e d
a c c o r d i n g to A S T M s t a n d a r d t e s t m e t h o d D 2 0 4 1 for u s e in c a l c u l a t i o n of
air v o i d c o n t e n t s u s i n g b u l k d e n s i t y v a l u e s c o m p u t e d f r o m t h e G T M s a m p l e
h e i g h t r e a d i n g s and the m e a s u r e d b u l k d e n s i t y a f t e r c o m p l e t i o n of d e n s i -
fication.
TABLE 3--Extracted aggregate gradations for S-III mixtures

without natural sand - after compaction
Percent Passing
MIX. A1 A2 A3 D2 E2
Sieve 0% Nat. 0% Nat. 0% Nat. 0% Nat. 0% Nat.

mm Sand Sand Sand Sand Sand
12.5 mm (1/2") i00 100 i00 i00 i00
9.5 m m (3/8") 99.5 99.0 99.6 96.4 98.4
4.76 (#4) 78.8 79.7 78.2 70.3 72.2
2.00 (#I0) 56.5 54.1 56.0 52.4 52.4
0.42 (#40) 29.6 23.4 26.9 24.8 24.9
0.177 (#80) 15.1 11.4 13.0 12.7 13.4
0.074 (#200)** 7.0 4.5 5.2 5.4 6.4
** M i n e r a l Filler
TABLE 4--Extracted aggregate gradations with natural sand - after

compaction
Percent Passing
MIX. B1 B2 D1 E1
Sieve 20% Nat. 10% Nat. 1 3 . 5 % Nat. 1 6 . 5 % Nat.

mm Sand Sand Sand Sand
12.5 m m (1/2") i00 i00 i00 i00
9.5 m m (3/8") 98.4 98.8 95.7 97.3
4.76 (#4) 70.0 71.5 68.2 72.3
2.00 (#i0) 46.9 45.5 52.1 51.7
0.42 (#40) 26.8 27.6 28.3 31.7
0.177 (#80) 13.1 13.3 13.5 15.1
0.074 (#200)** 4.4 4.4 4.4 4.9
** M i n e r a l Filler
T h e m o d e l G B - 4 C G T M s e t t i n g s u s e d for c o m p a c t i o n and densification

of I00 m m d i a m e t e r t e s t s p e c i m e n s a r e l i s t e d b e l o w :
Compaction Densification
I n i t i a l A n g l e of G y r a t o r = 3 degrees 3 degrees
Ram Pressure = 690 kPa 690 kPa
Initial Air Roller Pressure = 62 k P a 276 kPa
N u m b e r of R e v o l u t i o n s = 18 250 to 300
Mix Temperature = 135 t o 145 C 60 C
C o m p a c t e d hot-mix specimens were allowed to cool to room temperature,

u s u a l l y overnight, then placed in 60 C oven for 3 to 4 hours followed by
d e n s i f i c a t i o n t e s t i n g in the G T M w i t h the chuck heater set at 60 C.
A f t e r completion of d e n s i f i c a t i o n the samples were tested for bulk den-
sity and selected specimens used for extraction tests to check asphalt
content and gradation. Although not reported here, the amount of degra-
dation p r o d u c e d by d e n s i f i c a t i o n and the Marshall p r o p e r t i e s of the re-
m a i n i n g specimens were determined for use in future analyses.
ANALYSIS AND INTERPRETATION OF TEST RESULTS
The data generated from the 60 C d e n s i f i c a t i o n tests in the G T M

were reduced to conventional test parameters in the following manner.
First the Gyratory shear (Gs) values were plotted against the number of
r e v o l u t i o n s w h i c h served to identify the m i x t u r e ' s sensitivity to as-
phalt content. Figure 1 illustrates a typical plot p r e p a r e d u s i n g the
mean of three tests at each asphalt content. Also the design criteria
(Gs = 372 kPa at 200 revolutions) is shown w h i c h is thought to be ap-
p l i c a b l e for light to m e d i u m traffic conditions although it m a y apply
equally well for the design of pavements subjected to heavy traffic.
---m- 7.5% A/C t 7.0% N C ~ 6.5% N C

r7 6.0% N C x 5.5% N C ~ DES. CRITERION
Q..
O3
n-
<
uJ
I
n,-
>.
o go ibo igo 200

NUMBEROF REVOLUTIONS, N
FIG. 1--Typical GTM d e n s i f i c a t i o n results
Similarly m i x t u r e density curves were g e n e r a t e d using the m e a s u r e d

bulk density after d e n s i f i c a t i o n (e.g., after 300 revolution) with the
G T M sample height readings during d e n s i f i c a t i o n to calculate densities
at different increments of revolutions. Also, VMA and AV curves were
p r e p a r e d based on conventional test methods (e.g., Rice test) and calcu-
lation procedures.
S i n c e the m a i n o b j e c t i v e of t h i s r e s e a r c h was to d e m o n s t r a t e h o w
the G T M air r o l l e r t e s t i n g p r o c e d u r e c o u l d be u s e d to e v a l u a t e a s p h a l t
m i x t u r e s and to i d e n t i f y u n d e s i r a b l e m i x t u r e s w h i c h w o u l d be s u s c e p t i b l e
to e x c e s s p l a s t i c d e f o r m a t i o n , it w a s d e e m e d d e s i r a b l e to d e v e l o p r e l a -
t i o n s h i p s b e t w e e n Gs and v a r i o u s p h y s i c a l p r o p e r t i e s of t h e m i x t u r e s .
If t h e p r o p e r r e l a t i o n s h i p s c o u l d be d e v e l o p e d , t h e n it w o u l d v e r i f y
t h a t t h e u s e of Gs c r i t e r i a or c r i t e r i o n w a s i n d e e d v a l i d and f u r t h e r
w o u l d at least p a r t i a l l y e l i m i n a t e the d e p e n d e n c y on p a r a m e t e r s s u c h as
air v o i d c o n t e n t and v o i d s in m i n e r a l a g g r e g a t e s , t h e r e b y s i m p l i f y i n g
a s p h a l t m i x t u r e e v a l u a t i o n or design.
A d a t a b a s e w a s d e v e l o p e d by s e l e c t i n g Gs v a l u e s t h a t c o r r e s p o n d e d
to 200 r e v o l u t i o n s of d e n s i f i c a t i o n , as s h o w n in F i g u r e i. This data
w a s p l o t t e d in F i g u r e 2 to i d e n t i f y l i n e a r trends, s u c h as al-a2, be-
t w e e n Gs and a s p h a l t content. Figure 2 illustrates the different typi-
cal p l o t s o b t a i n e d for t h e S-I and S-III m i x t u r e s . T h e s t e e p e r t h e s l o p e
of t h e l i n e a r trend, the m o r e s e n s i t i v e the m i x t u r e w a s to i n c r e m e n t a l
i n c r e a s e s in a s p h a l t content. M i x t u r e s w i t h i0 to 20 p e r c e n t n a t u r a l
sand, a r e p l a c e m e n t of c r u s h e d l i m e s t o n e s c r e e n i n g , p r o d u c e d t h i s ef-
fect. T h e d a t a p o i n t s u s e d in s i m p l e l i n e a r r e g r e s s i o n a n a l y s e s of Gs
vs. A / C are s h o w n in F i g u r e 2 as c u r v e s a, b, and d. D a t a for c u r v e f
w e r e not u s e d b e c a u s e t h i s m i x t u r e c o n t a i n e d a g g r e g a t e s f r o m a d i f f e r e n t
source. D a t a for the h i g h e s t a s p h a l t c o n t e n t for m i x t u r e s t y p i c a l of
c u r v e c w e r e d o c u m e n t e d for s u b s e q u e n t use in m u l t i - l i n e a r r e g r e s s i o n
analyses. In t h i s c a s e it was a s s u m e d t h a t a l i n e a r t r e n d w o u l d e x i s t
if d a t a had b e e n g e n e r a t e d at e v e n h i g h e r a s p h a l t c o n t e n t s . D a t a for
m i x t u r e s s i m i l a r to c u r v e e w e r e not u s e d in the a n a l y s i s .
4--(a) , (b) x (c)
(d) x (e)--A--(0
(f) ~i..~.. S-I. DIFFERENT AGGREGATE SOURCE
,,,,,
OO "",
04 .......~K,,. (e)
(9 ~ LOW VMA (a2) HIGHVMA__~CJ~m ]

i i i i i i i i
PERCENT ASPHALT CONTENT (A/C), (%)

FIG. 2--Interpretation of G T M r e s u l t s
Table 5 gives a l i s t i n g of the r e s u l t s o b t a i n e d f r o m m u l t i - l i n e a r

regression analyses u s i n g d i f f e r e n t c o m b i n a t i o n s of v a r i a b l e s . A
TABLE 5--Comparison of r e g r e s s i o n analyses
Mixture-Composition n (a) R2 a b c d Eqn. m)

DO
S-I, 0% Nat. Sand 13 O. 96 117.18 -12.23 +i. 30 -0.63 1 C
I
S-I, 20% Nat. Sand 12 O. 95 142.00 -41.37 +14.53 -i. 93 2
m
S-III, 0% Nat. Sand 7 O. 89 34.78 -24.36 +12. ii -6.59 3
>
8 0.94 103.93 -8.02 -0.24 +I. 99 4 F
1 0 - 2 0 % Nat. Sand
0
S-I & S-III Combined - 0% S a n d 20 0.80 89.55 -17.22 +5.56 -3.30 5 Z
...............................................................................................................................
6~
S-I, 0% Nat. Sand 13 0.94 138.81 -11.55 -4.89 +0.25 6
DO
S-I, 20% Nat. Sand 12 0.88 152.01 -21.79 -4.14 +1.92 7
9
S-I, Combined 25 0.63 141.03 -12.09 -5.81 +0.33 8 -<
S-III, 0% Nat. Sand 7 0.96 158.51 -12.33 -6.12 +0.88 9 m
S-III, 1 0 - 2 0 % Nat. Sand 8 0.95 172.06 -12.61 -10.43 +0.59 l0 r-

C
5
Z
0"rl
_A:~_PA_~ ................................. ~ ........ ~:~A .... 3-~-~:~A .... z~:~A ..... .:~:~ ...... _:~:~ ........ ~ ......
S-I 0% S a n d 13 0.92 112.98 -10.71 -4.76 +0.28 14
b a t a u s e d t o d e v e l o p r e g r e s s i o n e q u a t i o n s l i m i t e d t o t h a t w h i c h p l o t t e d as a l i n e a r t r e n d of G y r a t o r y S h e a r m
(Gs) at 200 r e v o l u t i o n s v e r s u s p e r c e n t a s p h a l t c o n t e n t . A i r v o i d (AV) a n d v o i d s in m i n e r a l a g g r e g a t e s
(VMA) c o n f o r m t o t h e a s - c o m p a c t e d c o n d i t i o n . Gyratory shear values computed from these equations are m
in psi, m u l t i p l y b y 6895 t o c o n v e r t to Pa.
6Regression equation format: r-
m
E q n s . 1-5 G s = a + b (% A/C) + c (VMA) +d (AV) z
E q n s . 6-11 G s = a + b (% A/C) + c (MF) + d (% Pass. 0.42 mm)
Eqn. 12 G s = a + b (% A/C) + c (MF) + d (% Nat. Sand)
Eqn. 13 G s = a + b (% A/C) + c (MF) + d (% C o a r s e Agg. - % S - I - A a n d S - I - B used)
Eqn. 14 Gs = a + b (% A/C) + c (MF) + d (% Pass. 9.5 mm)
He~lett-Packard model 41CV calculator with statistics module was used

to p e r f o r m all of t h e r e g r e s s i o n a n a l y s e s in T a b l e 5. Regression equa-
t i o n s 1 t h r o u g h 5 are b a s e d u p o n A/C, VMA, and AV v a r i a b l e s . Comparison
of eqn. 1 and 2 s u g g e s t t h a t an i n c r e a s e in a s p h a l t c o n t e n t a n d / o r air
v o i d c o n t e n t p r o d u c e a m u c h g r e a t e r r e d u c t i o n in Gs for S-I m i x t u r e s
c o n t a i n i n g 20 p e r c e n t n a t u r a l s a n d t h a n t h o s e w i t h z e r o p e r c e n t n a t u r a l
sand. Also, an i n c r e a s e in V M A p r o d u c e s a g r e a t e r i n c r e a s e in Gs in
eqn. 2 (20% nat. sand) t h a n for eqn. 1 (0% nat. sand) b e c a u s e t h e 20%
n a t u r a l s a n d m i x t u r e s are m o r e s u s c e p t i b l e to a s p h a l t c o n t e n t c h a n g e s .
C o m p a r i s o n of eqn. 1 and 3 s h o w s t h a t the S - I I I m i x t u r e s w i t h 0%
n a t u r a l s a n d are e x t r e m e l y s e n s i t i v e to c h a n g e s in A/C, VMA, a n d A V and
m o r e s e n s i t i v e t h a n S-I m i x t u r e s w i t h o u t n a t u r a l sand. To i l l u s t r a t e
t h i s e f f e c t t h e V M A was c a l c u l a t e d u s i n g t h e s e e q u a t i o n s w i t h s e l e c t e d
A / C a n d A V v a l u e s b a s e d u p o n a Gs e q u a l t o 372 k P a (see T a b l e 6). Air
v o i d c o n t e n t s of 5 and 7 p e r c e n t w e r e s e l e c t e d to r e p r e s e n t a w e l l c o m -
pacted paving mixture.
TABLE 6--Predicted as-compacted VMA
Predicted V M A - As C o m p a c t e d (a)
% A/C % AV Eqn. 1 0% Nat. Sand E~n. 3, 0% Nat. Sand
6.0 5.0 10.2 16

7.0 11.2 17
6.5 5.0 15.0 17
7.0 16.0 18
7.0 5.0 19.7 18

7.0 20.7 19
It is o b v i o u s t h a t l i t t l e c h a n g e in A/C, VMA, or A V c o n t e n t s are

n e c e s s a r y to p r o d u c e a Gs v a l u e of 372 kPa u s i n g m i x t u r e s s i m i l a r to an
S-III (eqn. 3) w i t h 12.5 m m m a x i m u m s i z e a g g r e g a t e . H o w e v e r , t h e S-I
m i x t u r e s (eqn. i) r e q u i r e c o n s i d e r a b l e c h a n g e in V M A or s i m i l a r l y , t h e
m i x t u r e is not as s e n s i t i v e to c h a n g e s in A / C as t h e S - I I I m i x t u r e s .
A n o t h e r a p p r o a c h for i d e n t i f i c a t i o n of m i x t u r e v a r i a b l e e f f e c t s on
Gs w a s b a s e d on t h e use of A/C, MF, and o n e of t h e v a r i a b l e s d e f i n i n g
g r a d a t i o n , n a t u r a l s a n d content, or a m o u n t of c o a r s e a g g r e g a t e . Compar-
i s o n of Eqns. 6 t h r o u g h ii in T a b l e 5 i n d i c a t e d s o m e s i m i l a r i t i e s in t h e
c o n s t a n t s and c o e f f i c i e n t s for e a c h v a r i a b l e b e t w e e n t h e d i f f e r e n t e q u a -
tions. H o w e v e r , c e r t a i n d i f f e r e n c e s a p p e a r q u i t e logical. For e x a m p l e ,
t h e A / C c o e f f i c i e n t in eqn. 7 (S-I, 20% Nat. Sand) is a l m o s t t w i c e t h a t
of eqns. 6, 9, and I0. The e f f e c t of m i n e r a l f i l l e r (MF) is c o m p a r a b l e
to t h a t in eqn. 6. H o w e v e r , the p e r c e n t p a s s i n g t h e #40 s i e v e has sub-
s t a n t i a l l y m o r e e f f e c t t h a n in the o t h e r e q u a t i o n s w h i c h s u g g e s t s t h a t a
g r e a t e r p e r c e n t p a s s i n g the #40 s i e v e is n e e d e d to i n c r e a s e t h e V M A for
g r e a t e r Gs v a l u e s .
T h e e f f e c t of t h e s e v a r i a b l e s on Gs is i l l u s t r a t e d in F i g u r e 3
u s i n g E q u a t i o n 6. An i n c r e a s e in e i t h e r A/C or MF r e s u l t s in a loss of
shear resistance. A n i n c r e a s e in p e r c e n t p a s s i n g t h e #40 s i e v e p r o d u c e s
a s l i g h t i n c r e a s e in Gs and VMA. It is i m p o r t a n t to r e a l i z e t h a t m i x -
t u r e s w i t h a low p e r c e n t a g e of fine a g g r e g a t e ( g r a d a t i o n b e l o w n = 0.45)
s h o u l d e x h i b i t an i n c r e a s e in Gs and V M A w i t h a r e d u c t i o n in p e r c e n t
p a s s i n g t h e #40 sieve.
F i g u r e 4 is p r e s e n t e d to s h o w the c o r r e l a t i o n b e t w e e n S-I m i x t u r e s
w i t h 20% s a n d u s i n g e q u a t i o n s 2 and 7. In g e n e r a l , t h i s c o m p a r i s o n in-
d i c a t e s t h a t e i t h e r e q u a t i o n is r e a s o n a b l y r e l i a b l e w i t h i n t h e r a n g e of
v a r i a b l e s and m a t e r i a l s u s e d in t h i s study.
MF=3.0% [
500" i MF=5.0% -4l-- MF=7.0%
S-I, 0% NAT. SAND
--~<- MF=3.0%,25%#40 z MF=2.0%
13% PASSING #40
EQUATION 6
400
n-
<
UJ
-r"
n-
0 300
F-
>-
200
6 6[5 7
ASPHALT CONTENT (A/C), PERCENT(%)
FIG. 3--Effect of m i n e r a l filler and A/C on gyratory shear
I 9 PRED.Gs EQ.2 ~ PRED.Gs EQ.7 REGRESSION

4O0 I
c~
cL
L9
C3
LM 300-
0
C3
LU
n-
Q.
~
200
200 360 400
MEASURED Gs, KPa
FIG. 4--Correlation of Gs v a l u e s for S-I m i x t u r e s with 20% sand
The best equation to illustrate the effect of natural sand content

on shear r e s i s t a n c e of the m i x t u r e is eqn. 12 w h i c h was d e v e l o p e d by
c o m b i n i n g S-III m i x t u r e data for 0 and I0 to 20 percent natural sand
content. This equation produces a 14.1 kPa and a 32.4 kPa r e d u c t i o n in
Gs for each 1 percent increase in MF and each i0 percent increase in
natural sand, respectively. Figure 5 shows the effect that natural sand
content has in reducing the shear resistance of the S-I and S-III mix-
tures. The reduction in Gs produced by a 2.3 percent increase in miner-
al filler content appears almost equal to the reduction caused by a I0
percent increase in natural sand content.
9 S-I,NC=7% 9 S-III,NC=7% - - EQ.7

4 5 0 - ~ ~ EQ.12 EQ.12
v 400- 8% PASS. # 4 0
,=_
rr 350
I1~ I--..
"r" ! ..............
>-
n- 300 .... "--.
}-- "'-'-.......
< --~. 25% PASS. #40
>-CE 250" ""--....... MF=3.5%
(.9 """
200 ~" ,
0 5 10 15 20
PERCENT NATURAL SAND, (%)
FIG. 5--Effect of natural sand content
In summary, the results of this i n v e s t i g a t i o n have d e m o n s t r a t e d

that the shear resistance of asphalt mixtures containing different sizes
of crushed limestone from the same source is related directly to the
amount of coarse aggregate (percent of S-I-A and/or S-I-B in aggregate
blend) or the percent passing the 9.5 mm sieve, the amount of fine ag-
gregate (passing #40), the mineral filler content, and the percent as-
phalt content by weight of total mixture. Changes in these parameters
influence the V M A and AV contents of a mix. The d e v e l o p e d r e l a t i o n s h i p s
for p r e d i c t i o n of shear resistance (Gs) using A/C, VMA, and AV content
verify that the G T M is capable of evaluating the composite effect pro-
duced by aggregate blends with different aggregate gradations, surface
texture, and particle shape.
SUMMARY AND CONCLUSIONS
C o n v e n t i o n a l mix design procedures (e.g., Marshall or Hveem) pro-

vide limited reliability in assuring adequate resistance to plastic
d e f o r m a t i o n which may result in excessive t r a n s v e r s e or longitudinal
s h o v i n g (rutting) of a s p h a l t c o n c r e t e p a v e m e n t s . Although most design

m e t h o d s h a v e d e f i c i e n c i e s r e l a t i n g to t h e t e s t i n g d e v i c e a n d p r o c e d u r e s ,
t h e m o s t p r e v a l e n t and i m p o r t a n t d e f i c i e n c y is t h e l a c k of i n f o r m a t i o n
r e g a r d i n g t h e i n f l u e n c e of a g g r e g a t e p r o p e r t i e s on t h e a s p h a l t c o n c r e t e
m i x t u r e t h r o u g h o u t the r a n g e of d e n s i t i e s c o r r e s p o n d i n g to a s - c o m p a c t e d
to t r a f f i c d e n s i f i e d c o n d i t i o n s . T h i s p r o b l e m has b e e n p r o p a g a t e d by
t h e c o n c e p t t h a t m i x d e s i g n is for t h e p u r p o s e of o b t a i n i n g a d e s i g n
asphalt content. The r e c e n t l y c o m p l e t e d A A M A S i n v e s t i g a t i o n has p l a c e d
c o n s i d e r a b l e e m p h a s i s on the n e c e s s i t y for c o n s i d e r i n g o t h e r a s p e c t s
s u c h as d u r a b i l i t y in t h e m i x d e s i g n p r o c e s s [7].
The r e s u l t s of t h i s G T M i n v e s t i g a t i o n u s i n g a s p h a l t m i x t u r e s w i t h
l i m e s t o n e a g g r e g a t e s f r o m the same s o u r c e and a local n a t u r a l s a n d w a s
p e r f o r m e d to i l l u s t r a t e t h a t t h e g y r a t o r y s h e a r v a l u e c o u l d be r e l a t e d
to some of t h e p h y s i c a l p r o p e r t i e s of a g g r e g a t e s and t h e m i x t u r e s . The
c o n c l u s i o n s d e r i v e d f r o m t h i s s t u d y are:
i. T h e G T M c o m p a c t i o n and d e n s i f i c a t i o n t e s t i n g p r o c e d u r e p r o v i d e s
r a p i d a s s e s s m e n t of a m i x t u r e ' s s h e a r r e s i s t a n c e as r e l a t e d to c h a n g e s
in a s p h a l t content, a g g r e g a t e g r a d a t i o n , and d e n s i t y .
2. T e s t s c o n d u c t e d u s i n g t h e air r o l l e r and an i n i t i a l 3 d e g r e e a n g l e
of g y r a t i o n i n d i c a t e d t h e r a t e of d e n s i f i c a t i o n was i n f l u e n c e d s u b s t a n -
t i a l l y by t h e m i x t u r e s ' shear r e s i s t a n c e s . It is o b v i o u s t h a t t h i s con-
d i t i o n w o u l d e x i s t b e c a u s e the a n g l e of g y r a t i o n w i l l a p p r o a c h z e r o de-
g r e e s as t h e shear r e s i s t a n c e (Gs) i n c r e a s e s .
3. The r e s u l t s of r e g r e s s i o n a n a l y s e s d e m o n s t r a t e t h a t t h e g r a d a t i o n
and o t h e r c h a r a c t e r i s t i c s of the a g g r e g a t e b l e n d a f f e c t t h e m i x t u r e ' s
s e n s i t i v i t y to c h a n g e s in a s p h a l t content. P l o t s of G y r a t o r y s h e a r (Gs)
v e r s u s n u m b e r of r e v o l u t i o n s can be r e a d i l y i n t e r p r e t e d to e v a l u a t e t h e
s e n s i t i v i t y to c h a n g e s in a s p h a l t c o n t e n t and to d e t e r m i n e w h e t h e r it is
n e c e s s a r y to m o d i f y t h e a g g r e g a t e b l e n d to r e d u c e s e n s i t i v i t y .
4. R e g r e s s i o n a n a l y s e s i n d i c a t e d t h a t an i n c r e a s e in m i n e r a l f i l l e r
c o n t e n t r e d u c e d the s h e a r s t r e n g t h . H o w e v e r , if t h e air v o i d c o n t e n t
was i n i t i a l l y high, as f o u n d in a p r i o r s t u d y (3), t h e Gs v a l u e w o u l d
i n c r e a s e w i t h a d d i t i o n a l m i n e r a l filler, up to s o m e c r i t i c a l level.
5. An i n c r e a s e in the p e r c e n t p a s s i n g the No. 40 s i e v e p r o d u c e d an
i n c r e a s e in V M A for the m i x t u r e s e v a l u a t e d in t h i s study. This produced
an i n c r e a s e in t h e Gs value. In c e r t a i n o t h e r g r a d a t i o n s t h i s w o u l d
r e s u l t in a r e d u c t i o n in the V M A and Gs values.
6. T h e a m o u n t and size of c r u s h e d c o a r s e a g g r e g a t e or p e r c e n t p a s s i n g
the 9.5 m m s i e v e i n f l u e n c e t h e s h e a r r e s i s t a n c e and s e n s i t i v i t y of the
m i x t u r e to c h a n g e s in a s p h a l t content. Consequently, current aggregate
g r a d a t i o n s p e c i f i c a t i o n s for a s p h a l t c o n c r e t e m i x t u r e s s h o u l d e l i m i n a t e
t h e u s e of the n o m i n a l a g g r e g a t e s i z e r e q u i r e m e n t of, for e x a m p l e , 88 to
I00 p e r c e n t . T h i s s h o u l d be c h a n g e d to a s s u r e t h a t at least 5 or m o r e
p e r c e n t is r e t a i n e d on t h i s sieve (e.g., 85 to 94 p e r c e n t p a s s i n g ) .
7. T h e a d d i t i o n of n a t u r a l s a n d ( r o u n d e d not a n g u l a r ) to r e p l a c e
c r u s h e d s t o n e fines ( s c r e e n i n g s ) r e d u c e s t h e s h e a r s t r e n g t h and m a k e s
the a s p h a l t m i x t u r e m o r e s u s c e p t i b l e to c h a n g e s in a s p h a l t c o n t e n t . The
u s e of n a t u r a l sand and the a m o u n t s e l e c t e d for the a g g r e g a t e b l e n d
s h o u l d be c a r e f u l l y e v a l u a t e d to p r e v e n t a p r e m a t u r e r e d u c t i o n in the
m i x t u r e ' s s h e a r s t r e n g t h w h i c h c o u l d r e s u l t in e x c e s s p l a s t i c d e f o r m a -
t i o n of t h e a s p h a l t c o n c r e t e p a v e m e n t . N a t u r a l s a n d s are not r e c o m -
m e n d e d e x c e p t to improve, w h e r e n e c e s s a r y , t h e w o r k a b i l i t y of t h e m i x -
ture.
8. The r e l a t i o n s h i p s d e v e l o p e d u s i n g a s p h a l t c o n t e n t , v o i d s in m i n -
e r a l a g g r e g a t e s , and air v o i d c o n t e n t to d e f i n e t h e i r e f f e c t on t h e Gs
v a l u e are not i n t e n d e d for any p u r p o s e o t h e r t h a n to e l i m i n a t e t h e n e e d
for V M A and A V c o n t e n t d e t e r m i n a t i o n s in a s p h a l t m i x t u r e d e s i g n and
quality control/quality assurance. S i n c e the V M A is d e p e n d e n t u p o n t h e
a g g r e g a t e b l e n d p r o p e r t i e s , c u r r e n t V M A r e q u i r e m e n t s c a n o n l y be con-.
s i d e r e d as g e n e r a l g u i d e l i n e s . Whereas the GTM mixture evaluation pro-
c e d u r e r e q u i r e s o n l y t h e Gs v a l u e to d e t e r m i n e the a d e q u a c y of a g i v e n
mixture. F u r t h e r m o r e , m o d i f i c a t i o n of t h e a g g r e g a t e blend, e i t h e r by
u s i n g d i f f e r e n t a g g r e g a t e s or a l t e r i n g t h e b l e n d p r o p o r t i o n s , to i m p r o v e
the m i x t u r e can readily be determined using this procedure and the Gs

parameter. Only air void content or air p e r m e a b i l i t y c o r r e s p o n d i n g to
the a s - c o m p a c t e d mix should be of concern for m i n i m i z i n g the potential
for age hardening of the asphalt binder.
ACKNOWLEDGEMENTS
The financial support and technical assistance provided by the Flor-

ida Department of Transportation Materials Office is sincerely appreci-
ated and acknowledged.
REFERENCES
[!] Ruth, B. E., and Schaub, J. H., "Gyratory Testing M a c h i n e Simula-

tion of Field Compaction of Asphalt Concrete," Proceedings, Asso-
ciation of Asphalt Paving Technologists, Vol~ 35, (1966), pp. 451-
480.
[2] Ruth, B. E., "Field Compaction of Soils: A Technique for Control

Using Strength and Density Tests," Proceedings, III Interamerican
Conference on Materials Technology, (1972), pp. 144-152.
[3] Ruth, B. E., and Schaub, J. H., "A Design Procedure for Asphalt
Concrete Mixtures," Proceedings, Association of Asphalt paving
Technologists, Vol. 36, (1968), pp. 200-225.
[4] Ruth, B. E., Tia, M., and Sigurjonsson, S., "Evaluation of Experi-
mental Asphalt-Rubber, Dense Graded, Friction Course Mixtures:
Materials and Construction of Test Pavements on N.E. 23rd Avenue,"
Technical Report, U.F. Project No. 4910450426912, Department of
Civil Engineering, University of Florida, 1989, pp. 1-56.
[5] Sigurjonsson, S., and Ruth, B. E., "Evaluation of Asphalt Paving

Mixture's Shear Resistance Using the GTM," T r a n s p o r t a t i o n Research
Record 1259, 1990, Transportation Research Board, pp. 63-78.
[6] Ruth, B. E., Tia, M., and Sigurjonsson, S., "Gyratory Testing for
Mix Evaluation," Proceedings, Fourth International RILEM Symposium
on Mechanical Tests for Bituminous Mixtures, No. 8, 1990, pp. 575-
589.
[~] Von Quintus, H. L., Scherocman, J. A., Hughes, C. S., and Kennedy,
T. W., "Asphalt-Aggregate Mixture Analysis System, AAMAS," Na-
tional Cooperative Highway Research Program Report 338, Transpor-
tation Research Board, 1991, pp. 122-124.
Shuler, T. S., and Huber l, G. A.
EFFECT OF AGGREGATE SIZE AND OTHER FACTORS ON REFUSAL DENSITY OF

ASPHALT CONCRETE BY VIBRATORY COMPACTION
REFERENCE: Shuler, T. S., and Huber, G. A., "Effect of Aggregate Size and Other
Factors on Refusal Density of Asphalt Concrete by Vibratory Compaction", Effects of
Aggregates and Mineral Fillers on Asphalt Mixture Performance, ASTM STP 1147,
Richard C. Meininger, editor, American Society for Testing and Materials,
Philadelphia, 1992.
ABSTRACT: The physical behavior of asphalt concrete is related to the volumetric

characteristics of the mixture. A significant challenge facing asphalt pavement
engineers is predicting the future density and air voids of asphalt paving mixtures after
construction and during service. How rapidly changes in density and voids occur during
service can be related to the ability of the pavement to withstand forces imposed from
vehicles and the environment. Densification of asphalt concrete after construction and
during trafficking varies between mixtures and is related to the method of laboratory
compaction, design criteria, and construction technique. This study describes a procedure
for quantifying compactibility of mixtures and suggests a procedure for considering
densification rate as a criteria for describing optimum asphalt mixtures.
The so-called refusal density of six types of asphalt concrete mixtures was evaluated in
this study as function of vibratory hammer compaction. Three aggregate gradings ranging
from 1-1/2 inch to 3/8 inch (37.5 to 9.5 mm) maximum aggregate size were compacted in
the laboratory using three models of portable vibratory compactors administered by two
operators. The resulting factorial experiment was analyzed to determine effect on
volumetric properties due to hammer, aggregate grading, asphalt content and operator.
Vibratory compaction was compared with Marshall, kneading, and gyratory compaction
procedures to determine differences between each method and for each grading with
respect to density and voids characteristics. A new procedure is outlined for using the
vibratory compaction technique for development of a new asphalt concrete design method
and for adapting the technique for determining the sensitivity of asphalt concrete
mixtures to further densification under traffic.
Results of the study indicate that all three vibratory hammers evaluated compacted
asphalt concrete to higher levels of density than achievable by conventional Marshall,
Hveem or gyratory methods and therefore, should be capable of producing mixtures with
higher density than could be obtained after trafficking. The implication of this
efficiency during laboratory compaction is a design tool that could provide information
regarding the potential for mixtures to reach a plastic condition in service as related to
void content.
KEY WORDS: Refusal density, compaction, vibratory compaction, aggregate gradation,

asphalt mixture design.
Scott Shuler is Director of Research and Gerald Huber is Associate Director of

Research at the Asphalt Institute, Lexington, KY 40512-4052
265
Copyright9 1992 by ASTM International www.astm.org
BACKGROUND
The mixture design process for asphalt concrete includes preparation of candidate
materials in the laboratory as a means of providing test specimens which simulate
proposed pavement materials. The type and amount of compaction provided in the
laboratory should be simulative of pavement characteristics at some period after
construction during which the pavement is further densified by traffic. Various
techniques and apparatus have been designed for this purpose. It is not the intent of this
paper to describe the various devices and designs available for compaction of asphalt
concrete. This information is readily available in the literature (1, 2, 3). Instead, the
purpose of this paper is to describe an evaluation conducted using a new technique,
introduced by Cooper, et al (4), for preparation of asphalt concrete test specimens in the
laboratory by vibratory compaction.
There are at least three apparent advantages which justify further evaluation of the
new compaction technique:
9 Relative Efficiency
9 Unrestricted specimen size and shape
9 Portable
9 Low cost
The mixture design method proposed by Cooper and co-workers (4) uses a vibratory
hammer to determine the relative "refusal" density of asphalt concrete compared with
field compaction. Although this refusal density is a relative quantity dependent on the
compactor, the vibratory hammer chosen for use in the method is very efficient, and
provides significantly higher density than other conventional laboratory techniques.
The research described in this paper was conducted to determine the attributes of
laboratory vibratory compacted asphalt concrete compared with laboratory compaction
by other devices. A standard laboratory protocol for vibratory compaction was also
established.
EXPERIMENT HYPOTHESIS
The objective of this experiment was to determine if a vibratory compaction procedure

could be developed which could be used as the basis for a new asphalt concrete mixture
design. Ideally, the procedure would provide maximum densification in the laboratory
simulating the most compaction that could occur without degradation of the aggregate.
Not intended to necessarily represent the level of compaction possible during construction
or in service the so-called refusal density of the asphalt mixture would indicate whether
the mixture could ever achieve a level of density that might result in a loss of shear
strength. By designing mixtures to a level of refusal density where air voids could never
be reduced to some minimum level, plastic flow or loss in shear strength due to a high
percentage of voids filled, should not occur. Mixtures with too low voids at refusal
density could be regraded during design until voids at refusal density are above the
minimum established for eliminating plastic flow.
SHULER AND HUBER ON REFUSAL DENSITY OF ASPHALT CONCRETE 267
RESEARCH APPROACH
Identifying Significant Variables
An experiment was designed to determine the most efficient means of compacting asphalt
concrete into a cylindrical mold. The first phase of the experiment compacted asphalt
concrete in a four inch diameter Marshall mold and collar using a 22 kip (98 kN)
electrohydraulic testing machine. Loads were applied to the loose asphalt mixtures over
the four inch diameter (10.2 cm) at 90, 180, and 360 psi (0.62, 1.24, and 2.48 MPa) on one
face of the test specimen at vibrating frequencies of 0, 10, 20, 30, 50, and 70 Hz for up to two
minutes loading time. These levels of load and vibration were chosen to bracket those
applied by rollers during construction.
The mixture used during this first phase of the experiment consisted of the Asphalt
Institute laboratory standard crushed limestone asphalt concrete paving mixture with
the gradation shown in Figure 1. Optimum asphalt content of 4.7 percent by total mix
weight was chosen at 4 percent air voids based on 75 blow, mechanical, non-rotating base
Marshall compaction.
100
90
80
,/
70
60
50 /
;l/l
it')
40
D.
30
20
10
0
oo ~o ~r
FIGURE 1 - Laboratory Standard Crushed Limestone Gradation
Results of the first vibratory compaction experiment shown in Figure 2 seem to indicate 50
Hz as a frequency where an apparent minimum level of air voids is achieved. This
experiment also seems to indicate that as applied load is increased, changes in frequency
have less effect on density.
Comparison of Hand Held Vibratory Compactor and Other Methods
The experiment in Phase I indicated that pursuit of a compaction technique which

utilized a 50 Hz vibration frequency might provide the desired results. Therefore,
further research investigated use of a portable demolition hammer for compacting
asphalt mixtures which operated at approximately 50 Hz.
A first test of the vibrating hammer effectiveness as a laboratory compactor involved
comparison of mixture density and voids properties. The laboratory standard 3/4 inch
(19.0 mm) mixture was compacted using the electromechanical testing machine operated
at 50 Hz, the Marshall hammer at additional efforts of 35 and 50 blows, the Hveem
kneading compactor using both 4 inch and 6 inch molds (102 and 152 ram), a Texas gyratory
compactor modified for 1~ angle and 6 revolutions per minute and the vibrating hammer
with 4 inch and 6 inch molds (102 and 152 mm). Differences in air voids produced by each
compaction technique are shown in Figure 3.
16
14
~ 90psi
12
/ a ~ ~ ~ ~ 180psi
O
m 10
~ / " Res~ "360psi
Frequency?
6 , I = I ,,,I ,
0 20 40 60 80
Frequency, Hz
FIGURE 2 - Effect of Load and Frequency on Air Voids
The encouraging results shown in Figure 3 indicate that higher densities can be obtained
with the vibrating hammer ('P' vibratory, on figure) than any conventional methods of
laboratory compaction. Therefore, further work to develop a standard protocol for
preparing specimens with the vibrating hammer was conducted.
The objective of developing a standard compaction procedure was to produce mixtures

with the highest density regardless of maximum aggregate size, aggregate type, or
gradation.
FIGURE 3 -- Comparison of Vibratory and Conventional Compactors
DEVELOPMENT OF STANDARD VIBRATORY COMPACTION PROTOCOL
The first experiment was conducted to determine the effect on compaction due to maximum
aggregate size and time of vibratory load application. A wide variation in mixture
characteristics was desired for this phase of the research, therefore, gradations with 3/8
inch (9.5 ram)and 1-1/2 inch (37.5 ram) nominal maximum size crushed limestone
aggregates, respectively as shown in Figure 4 were used. Each mixture was compacted at
the asphalt content which produced 4% air voids after 75 blow Marshall compaction, as
before. The object of this experiment was to determine how aggregate size affected
maximum compaction and to begin to establish a protocol for the compaction method.
Since differences in density were observed in previous work due to mold size, compaction
of both mixtures was in 6 inch (152 mm) cylindrical molds using a 6 inch compactor foot.
The compaction process consisted of applying a seating load to the loose mixture for 10
seconds, applying the compaction load for varying periods from 10 seconds to 80 seconds
for the 3/8 inch (9.5 mm) mixtures and up to 160 seconds for the 1-1/2 inch (37.5 mm)
mixtures and finishing the process with a 10 second levelling load. This process was
repeated for each side of the specimen.
/ Original3/4" Mix
100 /
.i,i
IIIII I
90 3/8 inchnominal.!
80 / / ~ ~ 1 - 1 / 2 in!h nominal
70
60
~3
r
50
40 /
D..
30 ,/
20
10
0
o ~ ,,-
@4 CO ~ CO
FIGURE 4 - Gradations for Aggregate Size Experiment
The effect of this compaction procedure on air voids for the 3/8 inch (9.5 ram) and 1-1/2
inch (37.5 mm) mixtures is shown in Fi~;ure 5.
8.0
7.0 1-1/2"Mix
t
6.0
5.0
4.0 75 blowvoids
3/8" Mix
3.0 -- -- 6" foot
< 2.0
1.0
0.0 I I I I I I I I
20 40 60 80 100 120 140 160

Compaction Time, see/side
FIGLVRE5 - Effect of Compaction Time on Air Voids

Using 6 inch (152 mm) Tamper Foot
Although the 3/8 inch (9.5 mm) mixture could be easily compacted by this procedure, the
1-1/2 inch (37.5 mm) mixture resisted compaction, with air voids greater than that
produced by Marshall 75 blow compaction.
An attempt was made to reduce voids further by using a smaller compactor foot on the
vibratory hammer. A 4 inch (102 mm) diameter foot was fabricated and the compaction
protocol changed to provide compaction in eight segments within the 6 inch (152 ram)
diameter mold. Vibration was applied for 2 seconds within each segment as shown in
Figure 6 for total periods up to 160 seconds.
First Position
6" ID mold~ L a s t Position
33~d 7 6
~
ia foot
FIGURE 6 - Compactor Foot Positions During Compaction
The previous experiment was repeated for the 3 / 8 inch (9,5 mm) and 1-1/2 inch (37.5 mm)
mixtures using this new protocol and the effect on air voids is shown in Figure 7.
~
8.0
7.0 t
6.0
~:~ 5.0
3 ~ 1 ~ ~,,~ 75 blow voids
~ 4.0
3.0
L,.
:~ 2.0
1.0 3/8"; 4" foot
0.0 I ] I I I I 1 |
20 40 60 80 100 120 140 160
Compaction Time, sec/side
FIGURE 7-- Effect of Compaction Time and Imprint Area on Air Voids
An apparent asymptote for air voids occurs after approximately 40 seconds for the 3/8
inch (9.5 mm) mixture and 80 to 120 seconds for the 1-1/2 inch (37.5 mm) mixture with 4
inch foot. However, aggregate degradation was visible on the surface of the compacted
specimen after 120 seconds compaction. Results of sieve analysis of the extracted mixtures
after various compaction times are shown in Figure 8. These results indicate that
degradation is reduced if compaction ceases at 80 seconds.
100 vs
90
80
70
i
i I
60 120 s e c ~ ~ '
r
50
40
30
20
10
0
#1 ri~s7
r CO ~ CO
FIGURE 8 - Effect of Compaction Time on Aggregate Degradation
Comparison of Different Hammers and O p e r a t o r s
The next step in the research was designed to evaluate differences between commercially
available hammers. Also, since the hammers are hand held devices, variability due to
operators was also considered likely. An experiment was conducted using the compaction
protocol described previously to compare density and voids properties of mixtures
prepared with different vibratory hammers, operators, mixture gradation and asphalt
contents.
The experiment was designed as a full factorial with replication to measure the effect on
compacted mixture air voids of the following independent variables:
9 Compaction Hammer
B - Bosch Model 11305
K- Kango Model 638S
P - PowerTron Model 25P
9 Operator
Q
R
9 Asphalt Content
-1% optimum
optimum
+1% optimum
The resulting 3 x 2 x 3 x 2 replicate factorial was conducted for three mixture gradations
with nominal maximum aggregate sizes of:
1-1/2 inch (37.5 ram)

3/4 inch (19.0 mm)
3/8 inch (9.5 mm)
with gradings as previously shown in Figure 4.
ANALYSIS
The effect of each of the controlled variables on air voids for the three types of mixtures
were analyzed by multiple analysis of variance techniques. Only main effects were
judged significant, and therefore, are the only results presented in Tables 1 through 3.
T A B L E 1 - - S i g n i f i c a n t F a c t o r s f o r 1-1/2 i n c h (37.5 m m ) M i x t u r e s
Source of Variation ]Probabiiiiy > F Significant Significant
@ a = .05 ? @o~= .10?
Operator 1 0.066 no yes
Hammer 1 0.0002 yes
AC % 0.0000 yes
TABLE 2 - - S i g n i f i c a n t F a c t o r s f o r 3/4 i n c h (19.0 m m ) M i x t u r e s

Source of Variation Probability > F Significant Significant
@ cc = .05 ? @ a = .10 ?
Operator 0.0001 yes
Hammer 0.0013 yes
AC % 0.0000 yes
TAB!~E..3 -~.Z.S.!gnificantF.~ctors for 3/8 inch (9.5 m m ) M i x t u r e s

Source of Variation Probability > F Significant Significant
@ a = .05 ? @c~= .10?
O
Operator 0.027 .. yes
Hammer 0.112 no 110
AC% 0.000 yes
The practical significance of variation between controlled variables is often better

understood in terms of the dependent variable measured. Tables 4 through 6 indicate the
average differences in air voids observed for each of the control variables with the other
variables held constant.
T A B L E 4 - - V a r i a t i o n i n A i r V o i d s f o r 1-1/2 i n c h (37.5 m m ) M i x t u r e s
Source of Variation 1-1/2 inch (37.5 mm), Maximum i Significant
% Variation, % @ c~ = .05 ?
Operator Q 3.54
R 3.18 0.36 rio
Hammer K 3.98
P 3.33
B 2.77 1.21 y e~
AC% L 5.44
M 3.22
H 1.43 4.01 yes
TABLE 5 - - Variation in Air Voids for 3/4 inch (19.0 ram) Mixtures
Source of Variation 3/4 inch (19.0 mm), Maximum Significant
% Variation, % @ cc = .05 ?
Operator Q 4.13
R 2.75 1.38 yes
Hammer K 4.14
P 3.48
B 2.71 1.43 yes ..
AC% L 5.65
M 3.04
H 1.63 4.02 yes
TABLE 6 - - Variation in Air Voids for 3/8 inch (9.5 r a m ) M i x t u r e s

Source of Variation 3/8 inch (9.5 ram), Maximum Significant
% Variation, % @ a = .05 ?
Operator Q 2.79
R 2.57 0.22 yes
Hammer K 2.77
P 2.65
B 2.62 0.15 no
AC% L 4.50
M 2.27
H 1.41 3.09 yes
Notice that for each of the three mixtures Operator Q always produces higher air voids
than Operator R. Asphalt content was ordered as expected with AC% L (optimum -1%)
always producing higher air voids than AC% M (optimum) or AC% H (optimum + 1%),
respectively. The hammers were similarly ordered with H a m m e r K producing higher
voids than P and B, respectively. H a m m e r B produced the lowest voids for the 1-1/2 inch
(37.5 mm) and 3/4 inch (19.0 mm) mixtures and was numerically lower than Hammer P for
the 3 / 8 inch (9.5 mm) mixture, although not statistically.
ADJUSTING COMPACTIBILITY DURING DESIGN OF MIXTURES
H o w an asphalt concrete mixture handles in the laboratory during design provides clues
about how the mixture will behave during construction and after trafficking. Technicians
experienced with kneading compaction can tell the difference between a potential
"tender" mixture and one that will perform satisfactorily by the way the mixture is
displaced under the action of the kneading compactor foot. Technicians experienced with
the Texas gyratory compactor can relate similar notions depending on the rate at which
the mixture reacts to applied loads.
The vibratory compactor described herein can provide similar information about the
"compactibility" of asphalt concrete mixtures during design. Since the maximum density
achievable by the vibratory method is likely to be greater than that by traffic, a design
based on vibratory compaction would require use of some intermediate level of compaction
which more closely relates to in-service compaction. The rate at which the design
mixture reaches each of these two levels of density in the laboratory should be useful to
judge the adequacy of the mixture.
For example, mixtures A and B in Figure 9 both reach maximum density at approximately
80 seconds per side of the specimen but at different rates.
18.0 III. .___
16.0
14.0
o~ 12.0
9o
,m
10.0
o
> 8.0
.,-
,,r 6.0
acceptable ~ 4.0
r
unacceptable ~ 2.0
o.o I T I I
0 20 40 60 80
Compaction Time, sec/side
FIGURE 9 -- Compaction Rate for Two Mixtures During Vibratory Compaction
Mixture A reaches maximum density at a faster rate than Mixture B and is more sensitive
to compaction, achieving less than 3 percent air voids after only 10 seconds of vibratory
compaction. It is possible this mixture could become plastic in service if the low level of
voids shown after only 20 seconds compaction were obtained under traffic. Adjustment to
this rapid compaction could be accomplished by reducing asphalt content or changing
aggregate gradation so the minimum air void content could only be reached after higher
levels of vibratory compaction in the laboratory.
Ideally, the relationship between laboratory compaction, construction compaction, and

traffic compaction is known. The process of adjusting behavior of the asphalt concrete
during the design process is a relatively simple task of providing some limiting level of
voids to correspond with compaction after construction and later after trafficking. These
two levels are shown, hypothetically, on Figure 9 correlated with vibratory compaction
time.
The concept of monitoring the compaction rate of mixtures in the laboratory during design
has been suggested by others (5) and is repeated here because of its applicability to a
vibratory compaction device. Obviously, more work is necessary before the vibratory
compactors described herein could be utilized as the basis of a new asphalt concrete
mixture design procedure since the needed correlation between the vibratory compaction
protocol and construction/traffic densification has not been done.
CONCLUSIONS
A standard protocol for vibratory compaction of asphalt concrete in the laboratory

was established which produced test specimens with lower air voids than mixtures
produced by Marshall, gyratory or kneading compaction. Although no comparison
with rotating base Marshall or rolling wheel compaction was made, the relative
similarity of these methods to kneading compaction and gyratory with respect to
density indicates the vibratory method should produce higher relative density than
these devices, as well.
2. The refusal density of asphalt concrete paving mixtures is probably not a

fundamental value for a given mixture, but is related to the method of laboratory or
field compaction. The level of density achieved for the three mixtures evaluated
was different depending on the compactor used. Significant differences were
measured, not only between conventional and vibratory, but between the vibratory
compactors themselves.
3. Crushing of aggregates during vibratory compaction was dependent on compaction

time. An optimum vibratory compaction time was determined which minimized
crushing while achieving maximum density for both 3/8 inch (9.5 mm) and 1-1/2 inch
(37.5 ram) nominal mixtures.
4. Air voids in mixtures produced using the vibratory compaction protocol established in
this study were statistically significant for operator, hammer, and asphalt content in
each of 1-1/2 inch (37.5 ram), 3/4 inch (19.0 ram), and 3/8 inch (9.5 ram) mixtures.
5. Variation in voids due to operator effect was one of the lowest sources of variation in
the factorial experiment. It is believed this source of variation can be reduced further
by developing a mechanical method of keeping the vibrating hammer in contact with
the specimen.
Variation in voids due to hammer effect was more significant than variation due to
operator. Therefore, any standard compaction procedure using a vibratory compactor
should be based on specifications for weight, frequency and amplitude and not simply
by commercial brand name or model since changes in design could effect compaction
results.
7. Since the vibrating hammer compaction method leads to higher densities than can be
achieved with more conventional methods, a volumetric mixture design procedure
based on ultimate density with such hammers would require a reduction in the design
air voids criteria. Clearly, designs based on the current 3 to 5 percent air voids would
lead to insufficient asphalt content and would be difficult to compact in the field.
8. The vibratory compactors evaluated could be used to judge compactibility of design

asphalt concrete mixtures. This rate of compactibility could be used as the basis for
judging a mixture suitable for use.
REFERENCES
1. Marshall, B. G., "Marshall Stability Method for Design and Control of Asphalt
Paving Mixtures", published by the author, 1946, Asphalt Institute Library.
2. Hveem, F. N., "Kneading Compaction Method for Bituminous Mixtures to be Tested in

Hveem Stabilometer', Bituminous Paving Mixtures: Fundamentals for Design,
Highway Research Board Bulletin 105, 1955.
3. Ortolani, L. and Sandberg, H. A., "The Gyratory Shear Method of Molding Asphaltic
Concrete Test Specimens, ITs Development and Correlation With Field Compaction
Methods", Association of Asphalt Paving Technologists Vol. 21, 1952.
4. Cooper, K. E., Brown, S. F., Preston, J. N., Akeroyd, F. M. L., "Development of a

Practical Method for the Design of Hit Mix Asphalt", Paper No. 910546,
Transportation Research Board 70th Annual Meeting, January, 1991.
5. Huschek, S. and Pigois, M. L., "Orni6rage, 6tude comparative de diff6rentes

dispositifs d'essai', Institut f/ir Strassen-, Eisenbahn- und Felsbau (ISETH) an der
Eidgenossischen Technischen Hochschule Z/irich, Laboratoire des voies de circulation
(LAVOC) de l'Ecole polytechnique f6derale de Lausanne, November, 1984.
Fatigue, Modeling, and Theoretical
Dr. Osama Abdulshafi, P.E.
"Effect of Aggregate on Asphalt Mixture Cracking Using

Time-Dependent Fracture Mechanics Approach"
REFERENCE: Abdulshafi, 0., "Effect of Aggregate on

Asphalt Mixture Cracking Using Time-Dependent Fracture
Mechanics Approach", Effects of Aggregate and Mineral
Fillers on Asphalt Mixture Performance, ASTM STPII47,
Richard C. Meininger, Ed., American Society for
Testing and Materials; Philadelphia 1992.
ABSTRACT: Fatigue failure can be attributed to cyclic

(dynamic) loading energy which induces stresses and/or
strains below the allowable strength and/or fracture
strain level of a material. Since fatigue failures
evolve over time, damage theories have been postulated
to describe the mechanism by which this failure occurs.
When a crack exists or forms early in the life of a
pavement, fracture mechanics becomes the most relevant
theory to explain the damage process. If conditions
prevail to promote brittle fracture, the stress
intensity factor and its critical value, K (or Go),
becomes the most relevant parameter controlling the
cracking and fatigue damage process. If conditions
prevail to promote ductile fracture, the ductile
fracture
0
parameter,
0
J c , ~becomes
~
the oparameter for
conslderatlon. If condltlons prevall to promote creep
fracture, the C* - Line Integral offers the most
relevant fracture criteria. Since the definition of J
or C* in fracture mechanics imposes the condition that c
either parameter will degenerate into K (or Go) when
brittle fracture conditions prevail, the utilization of
J or C* provides the user with a powerful tool to
characterize the fatigue/fracture process in a more
general form. Since asphaltic mixtures can exhibit
brittle and/or creep fracture, it is apparent that the
C* - Line Integral can more closely capture the
fatigue/fracture damage process than K or J 9 In this
paper, a simple laboratory procedure to determxne the
C* - Integral, which is interpreted as the energy
release rate, is presented. In addition, a mixture
efficiency factor (~*) is introduced. The mixture
efficiency factor is used to examine and compare the
efficiency of asphaltic concrete mixture in resisting
fracture cracking using different aggregates and
asphalt cement binders.
Dr. Osama Abdulshafi, P.E., is the Director of R & D at

CTL Engineering, Inc., Columbus, Ohio.
281
Copyright 9by1 9ASTM

Copyright 92
Int'l (all Irights
by ASTM
reserved); Thu Dec
nternational
22 07:29:10 EST 2011
www.astm.org
KEYWORDS: Cracking of asphalt mixtures, time dependent

fracture mechanics, effects of aggregates and binders
upon asphalt mixture cracking, testing for c r a c k i n g
potential of asphalt mixes.
BACKGROUND: The field parameter approach for the

Fracture Mechanics Theory is used to describe the
strength of the energy field around the crack tip. In
this approach, a variety of field parameters are
available, such as the linear elastic stress intensity
factor K c (or the linear elastic energy release
parameter G ) [10,6], the contour integral, J, [ii],
the crack-t~p opening displacement, 4, [12, 13, 14],
and the energy rate line integral C*, [i, 2, 19, 22].
The success of any of these parameters in d e s c r i b i n g
the crack-growth rate depends upon several factors,
including: the geometry of the specimen, the time-
dependency of the material, the rate of d e f o r m a t i o n or
stressing, the temperature of the environment, as well
as whether the material obeys a linear or non-linear
law. If it is assumed that a crack exists in a body,
as shown in Figure (i), three (3) primary zones will be
generated as a result of this crack.
t Zone I: Non-Prolx~rgonal
Zone 1I: Fracture Dorr~nant
.Zone m: Remote Body
0~",.,.. X
"''"",,,,
"', ,, m x
u - Str~n
U * - Power~ E ~ e r g y Rate
Fracture A - Displacement
Condition Brittle Ductile 11me-dependent
,~ - Displacement Pate
Parameter Kc or Gc Jc C"
Energy __Su ~5 u 1]a ~_[hgu*
Formulation 5c
Figure 1
Fracture Mechanics Criteria
ABDULSHAFI ON TIME-DEPENDENT FRACTURE MECHANICS APPROACH 283
The first zone, directly around the crack, is d e n o t e d

as Zone I, the non-proportional zone; the second zone
as Zone II, the f r a c t u r e - d o m i n a n t zone; and the t h i r d
zone as Zone III, the remote zone. The most i m p o r t a n t
zone is Zone II, which actually c o n t r o l s the crack-
propagation. If the material in Zone II e x h i b i t s
brittle fracture, the field p a r a m e t e r w h i c h can most
readily describe the c r a c k - g r o w t h is G . If the
m a t e r i a l in Zone II exhibits ductile fracture, the
field parameter which is most d e s c r i p t i v e of the c r a c k
growth is J c . If the material in Zone II e x h i b i t s
creep f r a c t u r e (e.g., as in a steady-state, time-
dependent material), the field p a r a m e t e r w h i c h is most
d e s c r i p t i v e of the crack- growth is C*.
Since it is well known that asphaltic c o n c r e t e is a

t i m e - d e p e n d e n t material, it is more a p p r o p r i a t e to
utilize the time-dependent fracture m e c h a n i c s a p p r o a c h
to describe the crack growth in such material. There
are three primary approaches to t i m e - d e p e n d e n t f r a c t u r e
mechanics; namely, extended l i n e a r - e l a s t i c f r a c t u r e
m e c h a n i c s concepts, crack- growth in discrete, time-
d e p e n d e n t materials, and the field p a r a m e t e r s approach.
The constitutive used in the field p a r a m e t e r s a p p r o a c h
can be expressed as:
en = 1 + v/E olj - v/E akk 6ij + (3/2)B a "-le Sij
where:
e z~ = Strain Tensor Rate

o x3. = Stress Tensor Rate
Sij = Deviatoric Stress = o i. - (I/3)a~k61j
o = Effective Stress { (~/2)SijSij }
6" = K r o n e c k e r Delta
E ~J = Young' s Modulus
v = Poisson's Ratio
B = Creep Coefficient
n = Stress Exponent j
The superimposed dot denotes the t i m e - d e r i v a t i v e
G o l d m a n and Hutchinson [15], using e q u i l i b r i u m and

c o m p a t i b i l i t y equations in c o m b i n a t i o n w i t h the above
equations, concluded that the stress and strain fields
for such materials are of the HRR {named after
Hutchinson, Rice and R o s e n g r e n [16, 17]} type.
These fields can be expressed as:

0.. = C ( t ).i/.+I
~ bIr olj (e)
E..
13
-- C ( t )../.+I
Blnr Eli (8)
where:
r,8 = Polar coordinates centered at the crack tip

I n = Dimension factor, given g r a p h i c a l l y by Shih
[18]
o , e = Dimensionless angular functions, given

13 ~3
graphxcally by Shlh [18]
C(t) is the loading parameter which determines the

strength of the crack-tip singular fields and depends
upon the load, specimen geometry, and time. For
sustained loading, Riedel [19], Ohji, Ogura, and Kubo
[20] showed that directly after load application, C(t)
is a function of the stress intensity factor, Young's
modulus, stress exponent and time, whereas for a longer
time period, C(t) = C*. The C* integral is the steady-
state, time-dependent analog of the J integral. Thus,
the equation is:
C(t) = KI 2 (l-v 2) directly after load

(n+l) E t application
and
C(t) = C* = S W* dy - T i (aul/ax) dS
F for steady creep state
where:
W* = the strain energy rate density
----- T Gij d ~ij

0
= Line contour taken from the lower c r a c k

surface in a counterclockwise d i r e c t i o n to
the upper crack surface.
Ti = The traction vector defined by the outward

normal nj along (T i = olj nj).
S = The arc length along the path of the

integral.
EXPERIMENTAL METHOD:
Test Procedure: In order to use the C* - Integral as

a parameter to correlate creep c r a c k - g r o w t h rates, it
is assumed that the material is following a steady-
state creep law. This assumption ignores the transient
effects, which should be small in this case. This
path-independent integral (i.e. C* - Integral) was
introduced by Landes and Begley [21].
Marshall-size specimens were used in this

investigation. As shown in Figure No. 2, a right-
angled wedge was cut into the specimens to accommodate
the loading device, and cracks were i n t r o d u c e d either
by cyclic loading or by sawing. A four-piece loading
device was used to transmit the vertical force applied
by the loading machine.
WedQe
Initiation Cut
Crack. 05 inch
~plaorL
Figure 2
C* Integral Experimental Setup
The experimental d e t e r m i n a t i o n of the C* - Integral

involves the c o l l e c t i o n of load and c r a c k - l e n g t h data,
versus time for a constant d i s p l a c e m e n t rate. These
data are then used to determine load as a f u n c t i o n of
d i s p l a c e m e n t rate for various c r a c k - l e n g t h s , and c r a c k -
growth rate versus c r a c k length. The p o w e r or e n e r g y
rate input, U*, is m e a s u r e d as the area u n d e r the load-
d i s p l a c e m e n t rate curve. The energy rate, U* v e r s u s
c r a c k - l e n g t h for d i f f e r e n t d i s p l a c e m e n t rates is
plotted, and the slopes of these c u r v e s c o n s t i t u t e the
C* - Integral. The C* - Integral is t h e n p l o t t e d as a
function of the d i s p l a c e m e n t rate. Finally, the c r a c k -
growth rate is p l o t t e d as a function of the C* -
Integral.
The steps involved in obtaining the C* - Integral,

versus crack speed are shown in Figure No. 3.
Time (T) Z~ a
Step 1 Step 2 Step 3
z~ C'
Step 4 Step 5
Figure 3
Determining C* Parameter Schematically
Two types of aggregates and five types of asphalt

cements were used for this study. Typical material
test data for the aggregates and the asphalt cement are
shown in Tables (i), (2), and (3).
TABLE (1)
AGGREGATE TEST DATA
Aggregate Properties Aggregate Types
Gravel Granite
Los Angeles Abrasion 24.0 30.0

Water Absorption 0.9 1.03
Bulk Specific Gravity (Sp. Gr.) 2.568 2.742
Saturate Surface Dry Sp. Gr. 2.593 N/A
Apparent Sp. Gr. 2.635 2.656
Flakiness Index 24 10
Sand Equivalent 86 N/A
TABLE (2)
ASPHALT CEMENT TEST DATA
AC-30 AC-10 AR-4000 AR-2000 AC-10
Viscosity @
140F Poises 3,260 1,000 1,860 1,060 920
Penetration
(i00, 55) 70 102 53 76 154
TABLE (3 )
MIXTURE AGGREGATE GRADATION
Sieve
size 1" 3/4" 1/2" 3/8" #4 #8 #16 #3O !#50 #100 1200
Percent
P a s s i n g I00 95 81 69 49 35 24 17 12 8 5.5
Results and Analysis: Figures (4) through (8) show the

test results of C* - Integral, versus c r a c k - s p e e d for
the different mixes.
Since the asphaltic mixtures are crack-notch

insensitive, an additional parameter is introduced to
help in the characterization of these materials. This
parameter is the mixture efficiency factor (~*) w h i c h
is defined as:
--~i-
where:
~2 = Crack-speed at C* = i00
(ibs/in. min) [292.7 Kg/(cm 2. min)]
&1 = Crack-speed at C* = 50
(ibs/in. min) [146.4 Kg/(cm 2. min)]
As ~* increases, the efficiency of the m i x t u r e to

resist cracking is reduced. If ~* is equal to, or
greater than, 2, the mixture can be c h a r a c t e r i z e d as
not efficient in resisting cracking.
For the mixtures in Figure (4), the gravel mix has an

~* = 2.1 and the granite mix an ~* = 1.8, which, in
turn, means that when using these p a r t i c u l a r a s p h a l t
cement and aggregate combinations, the g r a n i t e m i x t u r e
is efficient in resisting cracking, w h i l e the gravel
mixture is not.
CRACK SPEED (IN/MIN)

1
/
0.8
0.6
0.4
0.2
0
20 40 60 80 100 120 140
C* INTEGRAL (IN LBS/IN IN MIN)
-- GRAVEL ~ GRANITE
Fig(4) C* INTEGRAL RESULTS FOR

T W O AGGREGATES
AC-IO Viscosity ~140F- 1000 Poises
1(IDs/in
2rain)= 2.927(kglcm
2n~n)

turn, means that when using these p a r t i c u l a r asphalt
cement and aggregate combinations, neither m i x t u r e is
efficient in resisting cracking.
i E 1 J J
20 40 60 80 1OO 120
-- GRAVEL ~ GRANITE

TWO AGGREGATES
AC-10 Viscosity e 140F-900 Poises
l(Ibs/in~ rain) = 2,927(kg/cm~rain)

cement and aggregate combinations, both m i x t u r e s were

1.4
1.2 /
J
018 .
0.4
0.2 ~
i i 1 i i i
0 60 100 160 200 260 300 350
-- GRAVEL ~ GRANITE
Fig(6) C* I N T E G R A L R E S U L T S FOR
TWO AGGREGATES
AR-4000 Viscosity e140F- 1860 Poises
l(Ibs/in2 m~) = 2.927(kg/cm2min)

cement and aggregate combinations, both mixtures were

1.2=
0.6 ~
0.4 ~
0.2
0 ~ 1 I i i
50 100 160 200 260 300
C" INTEGRAL (IN L B S / I N I N M I N )
-- GRAVEL ~ GRANITE

T W O AGGREGATES
AR-2000 Viscosity e 140F= 1060 Poises
t(Ibs/in 2 m h ) = 2 . 9 2 7 ( k g / c m 2 rain)

cement and aggregate combinations, both mixtures were

1.2
1 J
0.8
0.6
0.4
0.2
0 i i T i i
0 50 100 150 200 250 300
C ~ INTEGRAL (IN LBS/IN IN MIN)
-- GRAVEL ~ GRANITE

T W O AGGREGATES
AC-30 Viscosity e 140F= 3260 poises
l(Ibs/~n 2 rain) = 2 . 9 2 7 ( k g / c m 2 rain)
CONCLUSIONS
There are two competing mechanisms involved in time-

dependent fracture. One mechanism is the creep
deformation which is characterized by crack-tip
blunting in the material ahead of the crack tip. The
other mechanism results in an accumulation of creep
damage in the form of microcracks and voids that
enhance crack growth as they coalesce. Whichever
mechanism dominates determines the required energy for
crack growth.
The steady-state crack growth occurs when an

equilibrium between these two mechanisms is attained.
If the energy rate required to propagate the crack is
higher than the steady-state, the material is
considered more efficient in crack resistance than
material requiring lower energy rate than the steady-
state condition. The mixture efficiency factor (~*) is
equal to 2 at the steady-state condition. Therefore,
mixtures with ~* smaller than 2.0 is considered
efficient in reducing crack growth.
Based upon the above results, it can be concluded that

the C* - Line Integral and ~* could be used as a
meaningful tool to characterize the cracking potential
of asphaltic concrete mixtures.
ACKNOWLEDGMENTS: The author acknowledges, with

appreciation, all personnel who worked in the SHRP
AIIR-12 project, including Dr. Shafi Abdulshafi, Mr.
Kamil Kaloush, Mr. Ali Karaki, and Ms. Bozena
Kedzierski. Special thanks are due to Mr. Leon Talbert
for his review and editing of the manuscript of this
paper. The author also would thank Mr. Tate Antrim and
Ms. Angela Hickey for the graphic presentation and
typing of the manuscript.
REFERENCES:
i. Silverns, M. J. and Price, A. T. International

Journal of Fracture; Vol. 9, No. 2, 1973.
2. Majidzadeh, K. et. al.; "Application of Fracture

Mechanics for Improved Design of Bituminous
Concrete", Final Report, Federal Highway
Administration, Washington, D.C., FHWA-RD-76,
1976.
3. Sadanada, K. and Shahinian, P. Metallurgical

Translations; Vol. 8A, March, 1972.
4. Haigh, J. R. Material Science Enqineerinq; Vol.

20, 1975.
5. Harrison, C. B. and Sandor, G. N. Engineerinq

Fracture Mechanics; Vol. 3, No. 4, 1971.
6. McEvily, A. J. and Wells, C. H. "On Applicability

of Fracture Mechanics to Elevated Temperatures."
International Conference on Creep and Fatigue in
Elevated Temperature Appl~cations, American
Society of Mechanical Enqineerinq, ASME, 1973.
7. Abdulshafi, O. "Rational Materials

Characterization of Asphalt Concrete Pavements."
Ph.D. thesis submitted to The Ohio State
University, 1983.
8. Strategic Highway Research Program SHRP AIIR-12

"Innovative Techniques to Distinguish Performance
of Asphalt-Aggregate Interaction and Mixtures",
Final Report, 1990.
9. Riedel, H. "Crack-Tip Stress Fields and Crack-

Growth Under Creep-Fatigue Conditions." American
Society for Testinq and Materials, PhiladelDhia~
ASTM Symposium on Elastic-Plastic Fracture
Mechanics, 1981.
i0. Riedel, H. "Creep Deformation at Crack Tips in

Elastic-Viscoplastic Solids." Journal of
Mechanical Physical Solids; Vol. 29, 1981.
ii. Goldman, N. L. and Hutchinson, J. W. "Fully

Plastic Crack Problems: The Center-Cracked Strip
Under Plane Strain." International Journal of
Solids and Structures; 'Vol. II, 1975.
12. Hutchinson, J. W. Journal of the Mechanics and

Physics of Solids; Vol. 16, No. i, 1968.
13. Rice, J. R. and Rosengren, G. F. Journal of the
Mechanics and Physics of Solids; Vol. 16, No. l,
1968.
14. Shih, C. F. "Fracture Analysis." American

Society for Testinq and Materials, Philadelphia,
ASTM, STP 560, 1974.
15. Kubo, S., Ogura, K., and Ohji, K. "Engineering

Aspects of Creep", Proceedinqs, International
Conference, Sheffield, The Institution of
Mechanical Enqineers; Vol. 2, 1980.
16. Landes, J. D. and Begley, J . A . "Mechanics of

Crack-Growth." American Society for Testinq and
Materials, Philadelphia, ASTM, STP 590, 1976.
Dario Perdomo, Joe W. Button, and Robert L. Lytton
A NEW APPROACH FOR PREDICTION OF PERMANENTDEFORMATION
REFERENCE: Perdomo, D., Button, J. W., and Lytton, R. L., "A

New Approach for Prediction of Permanent Deformation," Effects
~f Aqqreqates and Mineral Fillers on Asphalt Mixture
Performance, ASTMSTP 1147, Richard C. Meininger, Eds., American
Society for Testing and Materials, Philadelphia, 1992.
ABSTRACT: Alaboratory investigation was performed in which the

influence on resistance to plastic deformation as demonstrated
when rounded, smooth, sand-sized aggregate particles were
replaced by rough, angular, porous particles while other
aggregates and the total gradation remain the unchanged. Five
asphalt concrete mixtures with varying amounts of these
aggregates were tested. Resulting data were used to (I)
evaluate the a b i l i t y of certain test procedures to differentiate
rutting potential of the mixtures, (2) develop a new rutting
model for predicting pavement performance which incorporates
aggregate characteristics, (3) examine fractal dimension
analysis as a method of quantifying aggregate angularity and
surface texture, and (4) demonstrate that octahedral shear
stress theory can properly differentiate the rutting potential
of asphalt concrete mixes.
KEY WORDS: asphalt concrete pavements, rutting, permanent

deformation, hyperbolic model, fractal dimension, octahedral
shear stress.
INTRODUCTION
A field investigation in Texas [1] revealed that excessive sand-size

aggregate particles, and the round shape and smooth texture of the
particles were major contributors to the rutting problem in asphalt
concrete pavements. A laboratory program was initiated to develop
methods to quantify the influence of aggregate on permanent deformation
in asphalt mixtures and j u s t i f y a new aggregate specification.
D. Perdomo is a researcher with the California DOT, Sacramento,

California. J. W. Button and R. L. Lytton are researchers at the Texas
Transportation Institute, Texas A&M University, College Station, Texas.
295
Copyright9 1992byby ASTM International
Downloaded/printed
www.astm.org
Five asphalt concrete mixtures containing the same aggregate

gradation and the same asphalt cement were developed for laboratory
testing. The material retained on the No. 10 sieve was crushed limestone
and was the same for all five mixtures. That passing the No. 10 for the
different mixtures varied from 100 percent crushed to 100 percent
uncrushed (natural, rounded) sand. The mixtures contained zero, 5, 10,
20, and 40 percent, rounded sand by weight of total mix. The five
mixtures were designed and tested to demonstrate the gain in resistance
to permanent deformation when angular particles replace round particles
and to provide a medium for the development of tests capable of
identifying rut-susceptible asphalt mixtures.
Space limitations do not allow details of derivations and

developments to be presented herein. For further details see references
I, 2 and 3.
DEVELOPMENT OF A NEW RUTTING MODEL
A novel theoretical approach, founded on the creep and recovery

behavior of asphalt concrete mixtures, was developed into a rutting
model. The resulting hyperbolic equations accurately model the
deformation performance of a paving mixture.
I t has been found that both creep and recovery compliance behavior
of an asphalt mixture follow curves similar to the ones shown in Figures
I and 2. Using a philosophical approach developed by Badillo [4], an
equation for predicting the creep compliance behavior of an asphalt
concrete mixture can be derived:
D(t) - D~ +D at ~ (i)
I +at m
where
DO i n i t i a l creep compliance,
Dm = maximum creep compliance,
a = regression constant,
t = time, and
m = slope factor.
The same philosophical approach can be applied to the recovery

compliance, even though i t shows a different slope factor and a partial
strain recovery. The equation obtained for t h i s recovery compliance is
represented as follows:
R(t) - R~ + Rmbt mp (2)

I + bt ~P
where
RO i n i t i a l (elastic) recovery compliance,
Rm maximum recovery compliance,
b regression constant, and
P slope modifier factor.
Indian Institute of Technology Kharagpur pursuant to License Agreement. No further reproductions aut
PERDOMO ET AL. ON PREDICTION OF PERMANENT DEFORMATION 297
Compliance, 1E-8 Pascals-1

2
1.5
0.5 . . . . . . . . . . . . . . .
1
6.5 % Air Voids Legend
Opfimurn Asphali Content Experimental Data
Theoretical Data
O I _ _ I ..... I
0 500 1000 1500 2000

Time, seconds
Figure I -- Linear representation of typical creep and recovery

behavior of a zero percent natural sand mix at high
air void contents.
Compliance, 1E-8 Pascals-1

4.5
3 . . . . . . . . . . . . . . . . . . .
1.5
Optimum Asphalt Conlent Legend

- e - Experlmental Date
/ - ~ - The~ calData
0 500 1000 1500 2000

Time, seconds
Figure 2 -- Linear representation of typical creep and recovery

behavior of a 40 percent natural sand mix at high
air void contents.
In order to obtain all the unknown parameters in Equations I (Do, Dm,

a, m) and 2 (Ro, Rm, b, p), an optimization technique known as pattern
search was used. Pattern search is a technique based on an optimization
method developed by Hooke and Jeeves [5]. The method consists,
basically, of finding the unknownparameters that minimize the sum of the
squared differences between observed data and predicted values.
Test Results
Figures i and 2 represent typical results from the laboratory test
program for mixtures with low and high s u s c e p t i b i l i t y to r u t t i n g ,
respectively. The mix considered highly susceptible to r u t t i n g is the
40 percent natural sand mix, while the mix considered to be rut-resistant
is the zero percent natural sand mix. Both figures include experimental
results, as well as theoretical predictions obtained by using the
hyperbolic models described above.
The results obtained from all the cases analyzed reveal that:
I. The hyperbolic modeling of the creep and recovery behavior

produced near-perfect predictions.
2. The value of "p" defined for Equation 2 increases as the
percentage of natural sand in the asphalt mixture increases
(Table I ) .
Based on these results, one can infer that the "p" value could account
for the role of the natural sand in the permanent deformation behavior
of asphalt concrete mixtures. This constitutes a novel way of
characterizing the influence of the aggregate's physical nature in a
permanent deformation model. More research is needed to analyze the
s e n s i t i v i t y of the "p" value to variations in asphalt binder
characteristics and to changes in the type and character of the coarse
aggregate used in the mixture as well as changes in test parameters such
as temperature and loading rate.
Table I -- Mean "p" values at high air void contents.
Mix type Mean "p" value

for 3 tests
0 % natural sand 0.8

Permanent Deformation Model
A permanent deformation model was derived from Equations I and 2.

Suppose that a load pulse of duration At is applied repeatedly, N times.
Then, compliance equations for both creep and recovery may be used to
estimate the total deformation and recovery that occurs:
D(N) - D~ + O rN ~ (3)
1 § Thu
Copyright by ASTM Int'l (all rights reserved); ~ Dec 22 07:29:10 EST 2011
R(N) - R~ + RmrpN~ (4)

I + rpN
where
N = numberof cycles,
r = a (At) ~ #,
rp = b (At) n~ pp,
t = N (At), and
#,#p = load pulse f a c t o r s .
When the load is not a square wave, the term (At) m must be
m u l t i p l i e d by p, which is a function of m and the loading wave shape, and
varies between 0 and i . For a square wave, p i s equal to I . The f a c t o r ,
pp, f o r the recovery curve has the same meaning as p, and i t i s a
function of #, mp, and the wave shape. For a square wave, p, is also
equal to 1.
The t o t a l accumulated s t r a i n a f t e r N r e p e t i t i o n s f o r a constant

stress pulse, Oo, is defined as:
~a(N) = E ( N ) -~ (N) =Oo[D(N ) - R ( N ) ] (5)

The rate of change of the permanent s t r a i n with load r e p e t i t i o n s is:
Replacing Equations (3) and (4) in Equation (6), and d i v i d i n g by the

r e s i l i e n t s t r a i n , E,, gives:
I aEa(N)
- ErDmmN~-I q(1 - qN ~) [(Dm - Do) - ( R - Ro)PS] (7)
Er aN
where
r
q - - - (8)
1 +rN ~
s - qp(1 - qpN~) (9)

q(1 - qN m)
r N re(p-l)
qp - P (10)
I + rpN mp
Er = r e s i l i e n t modulus
Now, r e c a l l i n g the permanent s t r a i n response model used in the VESYS

approach (6):
aE
p - Er.#.N -a (11)
ON
where
= parameters determined from Equation ( l ) ,
N = number of cycles,
Er = elastic or resilient strain, and
aEp
= rate of change of permanent strain with load repetitions.
aN
Thus, i f
/~N-m = Er L aN J
then, i t is apparent from Equation (7) that
e=1-m (12)
# = ErDmmq (1 -qN~) [(Dm -Do) - ( R - Ro)PS] (13)
where
_ m)
s = qp(1 qpN (14)
q(I - qNm)
The # and e parameters defined in Equations 12 and 13 has been
incorporated into the Texas Flexible Pavement System (TFPS) program [7]
developed at the Texas Transportation Institute in order to predict
rutting. In this approach, the loading-unloading response (a-e curve)
of the pavement material is modeled as in Figure 3. The strain response
is decomposed into (e-elastic (resilient) strain, and c,-permanent
strain. The total strain, (t, is the sum of (o + (,. The elastic
(resilient) strain remains f a i r l y constant during the l i f e of the
pavement except at low number of load repetitions and near failure. The
change in permanent strain per load application (AEp) decreases with the
number of load applications until the sample reaches failure, where Acp
starts to increase dramatically [ ] ] . In general, the permanent strain
is represented by Equation (5).
From Figure 3 i t is observed that

o
EIo(N) - = modulusduring loading (15)
E,(N)
o
Eun - = modulusduring unloading (16)
Ee
Now, rewriting Equation 15 by using Equation 11 gives:
E,o(N) - a _ a _ a _ E ,
AEp(N) + e 8 ( +(, E(I +/~V-~ 1 +#N -~ (17)
aN
This equation provides a relation between E~o(N) and Eu, (constant),
as a function of e, #, and N. The permanent deformation under load is
then calculated
Copyright by(all
by ASTM Int'l subtracting the surface
rights reserved); rebound,
Thu Dec 22 07:29:10 calculated
EST 2011 while all
PERDOMO ET AL, ON PREDICTION OF PERMANENT DEFORMATION 301
Applied
Figure 3 -- Typical loading - unlodaing response for an asphalt

pavement material.
the pavement layers are assigned the unloading moduli (Eun), from the
surface deflection, calculated while all pavement layers are assigned the
loading moduli (Eto).
The influence of rounded sand in the laboratory mix is i l l u s t r a t e d

by this r u t t i n g model [ i ] in that the Ioadingmoduli, (E,o), are 1.8 times
greater for the zero percent natural sand mix than for the 40 percent
natural sand mix. This indicates a much lower r u t t i n g potential for the
zero percent sand mix at a given level of t r a f f i c and temperature.
The hyperbolic model provides a near perfect predictor of actual

creep and recovery performance of asphalt mixtures. The "p" value
defined within the hyperbolic model shows a strong correlation with the
percentage of natural sand in a mix, hence, i t accounts for the influence
that the character of the aggregate particles has on deformation
behavior. The hyperbolic models for creep and recovery can be
mathematically transformed, as shown in previous pages, into a permanent
deformation prediction model. Additional work is needed to perform a
complete calibration of the new model.
FRACTAL DIMENSION ANALYSIS
Fractals are a family of mathematical functions that describe natural

texture and shapes. As a result, the theory has begun to receive
considerable attention as a method used for texture analysis and, in
general, for describing natural surface shapes [8 through 11].
The defining characteristic of fractals is that they possess a

fractal dimension. Measurement of the two-dimensional image fractal
dimension allows an estimation of the three-dimensional fractal
dimension, and t h i s three-dimensional fractal dimension relates well to
the common perception of roughness [12]. In order to evaluate the
fractal dimension, a black-and-white digitized image of the surface in
question is required. This digitized image is obtained from either a
photograph or a video frame, and then analyzed by a computer program that
includes the fractal mathematical functions.
The digitized image must be black and white because the analysis of
fractal dimensions is based on a gray scale. The gray levels of an image
are usually plotted on a histogram. This gray-level histogram is a
function that gives the frequency of occurrence of each gray level in the
image [ ] 3 ] . In other words, i f the gray levels are measured from 0 to
n, the value of the histogram at a particular gray level p, denoted h(p),
is the number of pixels (any of the small discrete elements that together
constitute an image) in the image with that gray level. Higher fractal
dimensions are produced by abrupt, large-scale changes, while lower
fractal dimensions are produced by gradual small-scale changes.
Experimental Analysis
The hypothesis was that images of crushed limestone aggregate had a
higher fractal dimension than images of uncrushed r i v e r gravel aggregate,
and that the difference between t h e i r fractal dimensionwas s t a t i s t i c a l l y
significant. Four digitized images (400 x 400 pixel regions) from black-
and-white video frames (Figures 4 and 5) of 13 mm aggregate were
analyzed. Mean fractal dimension, D, for the coarse r i v e r gravel was
2.34 and for the crushed limestone was 2.46. This difference seems
i n s i g n i f i c a n t until one realizes the maximum possible range of values is
2.0 to 3.0. I t was found that the crushed limestone images had a higher
fractal dimension than the r i v e r gravel images and the differences were
s t a t i s t i c a l l y significant.
The second procedure consisted of analyzing five non-overlapping 128
x 128 pixel subregions (within the whole image) for the purpose of
obtaining more s t a t i s t i c a l information. Fromthe results of this second
procedure the mean fractal dimension of the ten 128 x 128 pixel
subregions for a single aggregate type and size is very similar to the
mean fractal dimension of the two 400 x 400 pixel regions for the same
aggregate.
Although the fractal dimensions of the fine aggregate images showed
a significant s t a t i s t i c a l separation between crushed limestone and river
gravel, the standard deviations for the fine aggregate was twice that of
the coarse aggregate, which makes characterization of aggregate based on
fine material unreliable. Furthermore, visual analysis indicated that,
for the fine aggregate images, the fractal analysis was extracting
features from the mass ( d i s t r i b u t i o n of particles, contrasting colors,
etc.) rather than features from the individual particles. In other
words, the fine particles are too small to be analyzed using fractal
analysis on standard photographs or video frames. However, based on
results from the coarse aggregate particles, i t may be inferred that
photomicrographs of a dispersed layer of fine aggregate particles can be
analyzed
Copyright usingInt'l
by ASTM fractals.
(all rights Therefore, this
reserved); Thu Dec process offers
22 07:29:10 potential for
EST 2011
Indian Institute of Technology Kharagpur pursuant to License Agreement. No further reproductions au
PERDOMO ET AL. ON PREDICTION OF P~RMANENT DEFORMATION 303
Figure 4 -- River gravel images representing sets of (a) coarse

material and (b) f i n e m a t e r i a l . (Shown 40 percent
l arights
Copyright by ASTM Int'l (all r g e r reserved);
than actual size)
Thu Dec 22 07:29:10 EST 2011
Figure 5 -- Crushed limestone images representing sets of (a)

coarse material and (b) fine material. (Shown 40
percent larger than actual size)
quantifying the angularity of sand-sized or even f i l l e r - s i z e d particles.

I t should be pointed out that there are currently no test methods capable
of objectively evaluating aggregate shape and surface texture.
Additional analysis procedures consisted of applying a technique

known as histogram-flattening to the coarse aggregate images, in an
attempt to improve the statistical performance of the data. Histogram-
flattening, better known as histogram equalization, is a very useful
image-transform technique. The technique consists of defining a
"mapping" of gray levels, p, into gray levels, q, such that the
distribution of gray levels, q, is uniform {13]. This mapping compresses
contrast (reduces the range of gray levels) in areas with gray levels
near histogram minima and stretches contrast (expands the range of gray
levels) for gray levels near histogram maxima. Since contrast is
expanded for the majority of the image pixels, the transform generally
improves the resolution and recognition of many features in the image.
The results from the histogram flattened images (Table 2) consistently
supported the same conclusions drawn earlier, however, the differences
between means were notably higher than before.
Fractal analysis is a very promising technique for studying the

texture of aggregates. The technique unambiguously identifies and ranks
texture in coarse aggregates according to their fractal dimension.
Caution must be used when analyzing aggregates that present different
color tones within the individual particle structure or between different
particles, because fractal analysis is solely based on a gray scale and,
therefore, contrasting colors will yield unrealistically high values of
fractal dimension. However, this potential problem is not without
possible solutions. Two suggested solutions that deserve further study
include:
1. Coat aggregate particles with a very thin film of white or light
gray paint or dye.
2. Apply a correction factor to the fractal dimension parameter to
account for the contrasting colors.
Table 2 -- Fractal Dimension (Di) for five histogram-flattened

128 x 128 pixel subregions.
AGGREGATE IMAGE DI D2 D3 D4 D5 MEAN STD
Crushed Limestone 9 2.676 2.617 2.682 2.612 2.624 2.642 0.034

(coarse) 10 2.598 2.577 2.612 2.640 2.564 2.598 0.030
2.620 0.038
River Gravel 11 2.414 2.337 2.424 2.414 2.424 2.403 0.037
(coarse) 12 2.401 2.414 2.403 2.444 2.447 2.422 0.022
2.412 0.031
The fractal analysis technique is simple and of very practical

value. In this particular study, fractal analysis showed good
correlation with the character of the aggregate particles used and,
therefore, i t can be i n t r i n s i c a l l y related to the permanent deformation
performance of the mixture. I t should be possible to develop an
Copyright by ASTM classification
aggregate Int'l (all rights reserved); Thu Dec 22 07:29:10
system EST 2011
to objectively measure aggregate shape
and surface texture using fractal dimension analysis and provide

acceptance c r i t e r i a . Development of correlations between pavement
performance (or surrogate tests) and fractal dimension analysis should
enable pavement materials specialists to identify aggregate that w i l l
produce rut-susceptible mixtures.
OCTAHEDRAL SHEARSTRESS ANALYSIS
Freeman and Carpenter [14] used octahedral shear stress theory to

analyze premature deformation in asphalt concrete overlays over Portland
cement concrete pavements. They found that the octahedral shear stress
in a pavement can indicate how close to f a i l u r e a mixture w i l l be when
loaded. This indication of incipient f a i l u r e was given by the ratio of
actual octahedral shear stress in the pavement to the f a i l u r e octahedral
shear stress (strength) predicted by theory. Ameri-Gaznonand L i t t l e [2]
further developed the octahedral shear stress theory for a series of
typical pavement sections found across Texas. Ameri-Gaznon, et al [3]
estimated that rutting is imminent i f the octahedral shear stress ratio
(OSSR) exceeds 0.65.
In general, the octahedral shear stress is a scalar parameter that

defines the influence of the nine normal and shear stresses at a specific
point within a material. The theory is delineated in References I, 2,
3, and 14 and w i l l not be repeated here.
Using an appropriate computer program such as modified ILLIPAVE

[15], where a three-dimensional pavement is modeled by using a two-
dimensional halfspace of a f i n i t e solid of revolution, one can obtain
reasonably good information by which to evaluate the stress state within
the pavement structure under any loading and pavement boundary
conditions. Octahedral shear stress theory was used in t h i s manner to
analyze the potential for rutting of the two most widely different mixes
studied herein: the zero percent natural sand mix and the 40 percent
natural sand mix. Two different hypothetical pavement sections were used
in the investigation to give a total of four computer analyses. Contours
of stresses and OSSR's portrayed the d i s t r i b u t i o n of the failure
potential within the pavement structures. An example is shown in Figure
6. The relative potential for rutting was assessed by evaluating the
OSSR.
A series of contour charts were developed showing the variation of

the OSSRfor the different mixes and cases analyzed under both single and
dual t i r e s during four thermal seasons. Data from these contours were
used to produce plots l i k e the one in Figure 7, where the maximumOSSR's
are plotted against r e s i l i e n t modulus at pavement depths of 13 and 3.8
mm. Four data points for the seasons analyzed are plotted on the curves.
For the hottest season ($I), i t is observed that the potential for
f a i l u r e within the f i r s t inch of the surface layer is 1.6 to 1.8 times
higher for a 40 percent natural sand mix than i t is for a zero percent
natural sand mix.
There seems to be a c r i t i c a l modulus value above which variations

in mix characteristics have no major influence on the maximumoctahedral
shear stress ratio values for the top 38 mm of the pavement structure.
Copyright
Thisby ASTM
fact Int'l
may (all rights reserved); Thu
indicate thatDec 22 07:29:10 EST 2011
octahedral shear stress theory should be
OSS RATIO - CASE I - SEAS I/GRA

~0 I 2 3 4 5 6 7 8
-4
Figure 6 -- Octahedral shear stress ratio contours for 40

percent natural sand mix surface layer in
traditional pavement structure under single t i r e
loading, and for hottest season (season I ) .
Maximum Octahedral Shear Stress Ratio

1
-B- 40 ~l, 8and. 13mm Deep I
~. ................................................ - 0 - 40 '~ 8and, 38mm Deep I ......
0.8
......................." ~ .... -~.- 0 , S I n d , 13rnmOeep I
--A- 0 " 8 " n d . 38mmDeep
0.6
0.4
0.2
I I I I I
0 200 400 600 800 1000 1200
Resilient Modulus, pascal x 1000
Figure 7 -- Maximum octahedral shear stress ratio versus

r e s i l i e n t modulus for the upper most surface
layer in traditional pavement structure under
single t i r e loading.
limited to the analysis of shear strength data collected at relatively

high temperatures due to the overwhelming influence (on shear strength)
on the cohesive strength of the asphalt cement at low temperatures.
CONCLUSIONS
i. A hyperbolic theoretical model was developed that provides a near

perfect predictor of actual laboratory creep and recovery
performance of asphalt mixtures. The "p" value defined within the
model showed strong correlation with the amount and character of the
sand-size aggregate particles, thus indicating i t is representative
of the influence that these particles have on permanent deformation.
From the state-of-the-art theoretical approach, a new permanent
deformation prediction model was developed which is believed to
represent actual field performance.
2. Fractal analysis appears to be a very practical method for
objectively quantifying the shape and surface texture of aggregate
particles.
3. Octahedral shear stress theory can be used effectively to evaluate
the permanent deformation potential of asphalt mixtures under
specific stress states.
ACKNOWLEDGEMENTS
This work was sponsored by the Texas Department of Transportation in

cooperation with the Federal Highway Administration.
REFERENCES
I. Perdomo, D. and Button, J.W., "Identifying and Correcting Rut-

Susceptible Asphalt Mixtures," Research Report 1121-2F (Draft),
Texas Transportation Institute, Texas A&M University, College
Station, Texas, 1991.
2. Ameri-Gaznon, M. and L i t t l e , D.N., "Permanent Deformation Potential
in Asphalt Concrete Overlay Over Portland Cement Concrete
Pavements," Research Report No. 452-3F, Texas Transportation
Institute, Texas A&M University, College Station, Texas, 1988.
3. Ameri-Gaznon, M., Button, Joe W., Perdomo, D., L i t t l e , D.N., and
Zollinger, D . G . , "Avoiding Early Failure of Intersection
Pavements," Research Report 1172-1F, Texas Transportation
Institute, Texas A&M University, College Station, Texas, 1989.
4. Juarez-Badillo, E., "General Compressibility Equation for Soils,"
Tenth International Conference on Soil Mechanics and Foundation
Engineering, Stockholm, Sweden, Vol. i, pp. 171-178.
5. Hooke, R. and Jeeves, T.A., "Direct Search Solution of Numerical
and Statistical Problems," J. Assn. Comp. Mach., 8, 2, April, 1961.
6. Kenis, W. J., "Predictive Design Procedures," VESYS User's Manual,
FHWA Report 77-154, 1978.
7. Uzan, J. and Smith, R.E., TFPS Technical Report, Texas
Transportation Institute, Texas A&M University, College Station,
Texas, August, 1988.
8. Mandelbrot, B.B., "Fractals: Form, Chance and Dimension," San
Francisco, California, 1977.
9. Mandelbrot, B.B., "The Fractal Geometry of Nature," San Francisco,

California, 1982.
10. Richardson, L.F., "The Problem of Contiguity: An Appendix of
Statistics of Deadly Quarrels," General Systems Yearbook, Vol. 6,
pp. 139-187, 1961.
11. Fournier, A., Fussel, D., and Carpenter, L., "Computer Rendering of
Stochastic Models," Commun. ACM, Vol. 25, No. 6, 1982, pp. 371-384.
12. Gagalowicz, A., "A New Method for Texture Fields Synthesis: Some
Applications to the Study of Human Vision," IEEE Trans. Pattern
Anal. Machine I n t e l l . , Vol PAMI-3, September, 1981, pp. 520-533.
13. Gonzales, R.C., "Digital Image Processing," Advanced Book Program,
1977.
14. Freeman, T.J. and Carpenter, S.H., "Characterizing Permanent
Deformation in Asphalt Concrete Placed o v e r Portland Cement
Concrete Pavements," Transportation Research Record 1070,
Washington, D.C., 1986.
15. Duncan, J.M., Monismith, C.L., and Wilson, E.L., "Finite Element
Analysis of Pavements," Highway Research Record 28, 1968, pp. 18-
33.
Y. Richard Kim', Nakseok Kim 2, and N. Paul Khosla 3
EFFECTS OF AGGREGATE TYPE AND GRADATION ON FATIGUE

AND PERMANENT DEFORMATION OF ASPHALT CONCRETE
REFERENCE: Kim, Y. R., Kim, N., and Khosla, N. P., "Effects of Aggregate
Type and Gradation on Fatigue and Permanent Deformation of Asphalt
Concrete," Effects of Aggregates and Mineral Fillers on Asphalt Mixture
Performance, ASTM STP 1147, Richard C. Meininger, Editor, American Society
for Testing and Materials, Philadelphia, 1992.
ABSTRACT: This paper presents laboratory test results demonstrating the effects
of aggregate type and gradation on fatigue and permanent deformation of asphalt
concrete. Diametral fatigue tests and uniaxial incremental static creep tests were
performed under varying temperatures and mixture variables, including aggregate
type and gradation.
The effects of aggregate type and gradation on permanent deformation were
evaluated under test combinations with changing asphalt type, asphalt content, air
voids content, temperature, and applied stress level. For the fatigue study, the
effect of aggregate type was evaluated by changing asphalt content, air voids
content, and temperature. The test results were analyzed using statistical analysis
and graphical comparison of data. The analysis of variance (ANOVA) tests were
performed to investigate the main effects and interactions of the test variables
with the aggregate type or gradation.
The analysis revealed that, with the size of experimentation used in this study,
aggregate type has significant effects on fatigue resistance and permanent
deformation of asphalt concrete, indicating better performance from the mixtures
comprised of aggregates with a rough surface texture and an angular shape.
Coarse gradation, meaning a larger proportion of coarse aggregates with the same
nominal maximum aggregate size compared to medium gradation, did not show
significant effects on permanent deformation. Interactions of aggregate type with
gradation, asphalt type, air voids, and temperature were found to be significant for
the permanent deformation of asphalt concrete, whereas no interaction appeared
to be significant for fatigue with the given size of experimentation.
KEYWORDS: aggregate type, gradation, fatigue, permanent deformation
'Assistant Professor, 2Graduate Research Assistant, and 3Professor, respectively,

Department of Civil Engineering, North Carolina State University, Raleigh, NC
27695-7908.
310
Copyright 9 1992 by ASTM International www.aslm.org
KIM ET AL. ON FATIGUE AND PERMANENT DEFORMATION 311
INTRODUCTION
Aggregate is a major structural component of asphalt concrete of which

physical and chemical properties significantly affect the performance of asphalt
concrete pavement in service. In order to design asphalt concrete mixes properly,
one has to understand the basic properties of aggregates, interactions between
aggregates and binders, and the effect of environmental conditions and mixture
variables on the performance of asphalt concrete with different aggregate
characteristics.
The relationship between aggregate properties and pavement performance has

been reported by many researchers [!-9]. Dukatz [i], based on the earlier
research [2-5], concluded that the permanent deformation of asphalt mixtures is
mainly affected by the shape, gradation, durability, and the toughness of the
aggregate. Also, the aggregates that are very hard and tough were found to be
more fatigue-resistant.
State-of-the-art review on permanent deformation of asphalt concrete was

recently conducted by the University of California at Berkeley as a part of the
Strategic Highway Research Program A-003A project [6]. Based on the review of
others work [7-9], the effect of aggregate properties on permanent deformation of
asphalt concrete was summarized in that the rutting resistance increases with a
rougher surface texture, with continuous gradation compared to gap gradation,
with a more angular particle shape, and with larger maximum size aggregates.
As demonstrated in these references, there is a considerable amount of data

verifying the significant effects of aggregate properties on the asphalt concrete
performance. Nevertheless, there is little work done on dealing with the
significance of interactions among aggregate properties and other mixture
variables or environmental factors on the performance of asphalt concrete.
The objectives of this paper are first, using the laboratory test data generated
at the North Carolina State University as a result of the SHRP A-003A
subcontract, to evaluate the effects of aggregate type and gradation on fatigue and
permanent deformation, and secondly to identify the significant interactions among
aggregate properties and other variables.
MATERIALS
Asphalt Cement
Two asphalts used in this research were AAK-1 (AC 30) and AAG-1 (AR
4000) asphalts according to the SHRP Materials Reference Library codes. These
were selected because of their vastly different compositional and temperature
susceptibility characteristics. AAG-1 asphalt was found to be more temperature
TABLE 1--Asphalt contents (%) by weiclht of a.q,qreclate [10].
RB Aggregates RL Aggregates
AAK-1 AAG-1 AAK-1 AAG-1
Low High Low High Low High Low High
5.1 5.7 4.9 5.5 4.3 5.0 4.1 4.8
susceptible than AAK-1 asphalt as defined in the SHRP Asphalt Concrete

Specimen Preparation Protocol [10].
Two levels of asphalt content for each asphalt were determined by the
University of California at Berkeley as shown in Table 1. The higher asphalt
content was about 0.6 to 0.7 percent more (by weight of aggregate) than the lower
asphalt content.
Ag_gre~ates
The aggregates used in this research were RB and RL aggregates (SHRP

Materials Reference Library codes). The main reason for this selection was due
to their surface textures which result in different stripping potentials and bond
strengths between aggregates and asphalt cement. RB aggregates have a rough
surface texture which results in good bonding and low stripping potential, while
RL aggregates have a smooth and polished surface texture resulting in poor
bonding and high stripping potential.
Medium and coarse gradations were selected in accordance with the

specification limits by ASTM Test Method for Standard Specification for Hot-
Mixed, Hot-Laid Bituminous Paving Mixtures (D 3515) as shown in Figure 1. The
gradations shown in Figure 1 are the gradations after washing the aggregates. The
same gradations were used for both types of aggregates. It should be noted that
the difference between two gradations is not in the nominal maximum size of
aggregates, but in the proportions of coarse and fine aggregates. The detailed
sieve analysis procedure can be found elsewhere [10].
SPECIMEN PREPARATION AND COMPACTION
Mixing and Curing
From the Bitumen Test Data Chart the optimum mixing temperatures were
determined to be 284~ (140~ for AAG-1 asphalt and 300~ (149~ for AAK-1
iO0 --
ASTM D 3515 ..--3:4~
Specification .."~/'/
80
(3)
u)
t~ 60 7 J "~
Medium ~,," / 4 ,(/ ^
E Gradation ,," J / , / Coarse
,~ .-//,~i~- Gradation
40
tD
ft. ," /./,,""
m
t~
~$7. 7" f .'"J .J
20
I I ! ! I1 1 v 1 11/2 I
1200 II100 150 130 t 6 18 I14 3/8 314 ]-
U.S. Standard Sieves

FIG. 1--Selectedaggregate gradations.
asphalt [10]. The optimum temperatures correspond to a viscosity of 170___20
centistokes. More detailed information on mixing and curing can be found
elsewhere [10,11].
Compaction
Cured mixtures were compacted at 240~ (116~ using the Gyratory Testing
Machine designed by the United States Army Corps of Engineers (COE). One
degree of gyratory angle was used to produce briquets with 4 in. (10.16 cm)
diameter and 2.6 in. (6.604 cm) height and cylinders with 4 in. (10.16 cm)
diameter and 8 in. (20.32 cm) height. Two levels of compaction effort were
imparted to mixtures in order to produce specimens with two air void contents,
4% and 8%. The method used for measurement of the air voids data presented
in this paper was wet-with parafilm (WWP) method. This method employs elastic-
wax paper, parafilm, as a substitute for paraffin wax. The use of parafilm during
the air voids measurement prevents water from entering the specimen during
submerged weighing which could result in underestimation of air voids [12].
EXPERIMENTAL DESIGN
Fatigue
Five factors were investigated in the fatigue testing: aggregate type, asphalt
type, asphalt content, air voids content, and test temperature. The final testing
combinations for the diametral fatigue tests are presented in Table 2. Due to the
change in the air voids measuring technique during this project, fatigue data on
RB aggregates are available only with AAG-1 asphalt cement. Four to six
specimens were tested for each test combination.
The factors investigated in the course of permanent deformation study include

aggregate type, aggregate gradation, asphalt type, asphalt content, air voids
content, temperature, and applied stress level. The original experimental design
was a half-factorial design resulting in a total of 64 test combinations. Since the
specimens at 140~ (60~ failed at the early stage of testing with the applied
stress level of 20 psi (138 kPa), all the tests at 140~ (6&C) were performed with
the stress level of 10 psi (69 kPa). Final testing combinations are presented in
Table 3. One replicate was used for each test combination.
TABLE 2--Final test combinations for diametral fatiaue testin.q.
abcde" K1~ Kz~
01000 5.401 x 1013 4.405

01101 5.759 x 1011 3.559
01001 1.597 x 10" 3.708
01010 5.221 x 10"11 3.758
01100 7.005 x 10-19 5.970
01111 4.081 x 10~ 2.817
10100 8.421 x 1018 4.719
10010 2.163 x 10" 3.797
10001 1.842 x 1015 4.016
10111 2.406 X 1 0 -13 3.484
11000 3.000 x 10.9 2.491
11110 9.378 x 10lz 3.163
11011 2.000 x 10.9 2.404
11101 7.000 x 10.9 2.222
NOTE:
a a
= Aggregate Type (0 = RB 1 = RL)
b = Asphalt Type (0 = AAK-1 1 = AAG-1)
c = Asphalt Content (0 = Low 1 = High)
d = Air Voids Content (0 = 4_+0.5% 1 = 8___0.5%)
e = Temperature (0 = 32~ (0~ 1 = 68~ (20~
b Fatigue constants from the power law between the number of cycles to
failure and inverse of recoverable horizontal strain.
TABLE 3--Final test combinations for permanent deformation testin.q.
abcdefg a Slope b Interceptb e;
0010001 0.172 1.384 x 10.3 2.00 x 10-4

0011101 0.231 1.698 x 103 3.25 x 10.4
0011000 0.183 6.823 x 10.4 1.25 x 10.4
0010100 0.198 1.119 x 103 1.58 x 10.4
0001001 0.208 6.934 x 10.4 2.20 x 10.4
0000101 0.199 1.086 x lff 3 3.13 x 10.4
0000000 0.243 3.090 x 10.4 1.00 x 10.4
0001100 0.229 4.667 x 10.4 1.48 x 10.4
0010010 0.236 8.072 x 10.4 3.58 x 10.4
0010110 0.162 2.188 X 1 0 .3 3.75 x 10 3
0011010 0.151 1.687 x 103 4.40 x 10.4
0011110 0.360 1.472 x 10.3 N/Aa
0000010 0.146 1.016 x 10 -3 2.40 x 10.4
0000110 0.147 1.641 x 103 3.63 x 10.4
0001010 0.143 1.172 x 103 3.58 x 10.4
0001110 0.217 1.538 x 103 3.51 x 10.4
0111001 0.188 1.346 x 10 -3 2.90 x 10.4
0110101 0.196 1.324 X 1 0 .3 2.50 X 104
0110000 0.201 5.200 X 10.4 1.00 X 10.4
0111100 0.200 1.164 X 10.3 1.25 X 10.4
0100001 0.198 7.278 X 10.4 2.25 X 10.4
0101101 0.210 1.099 X 1 0 .3 2.75 x 10.4
0101000 0.265 2.931 x 10.4 1.75 x 10.4
0100100 0.231 5.445 x 10.4 1.25 x 10.4
0110010 0.207 8.590 x 10.4 7.50 x 10.5
0110110 0.191 1.762 x 10"3 3.68 x 10.4
0111010 0.254 7.447 x 10.4 3.25 x 10.4
0111110 0.372 1.233 x 10-3 N/A
0100010 0.204 4.808 x 10.4 3.25 x 10.4
0100110 0.212 7.261 x 10.4 3.18 x 10.4
0101010 0.212 5.224 x 10.4 3.13 x 10-'
0101110 0.126 2.023 x 103 4.50 x 10.4
1011001 0.287 1.581 x 10 .3 4.00 x 10.4
1010101 0.356 1.442 x 10.3 N/A
1010000 0.235 5.080 x 10.4 1.50 x 10.4
1011100 0.373 7.210 x 10.4 2.50 x 10.4
1000001 0.331 2.720 x 10.4 1.85 x 10.4
1001101 0.288 1.012 x 10.3 3.25 x 10.4
1001000 0.314 2.050 x 10.4 1.25 x 10.4
1000100 0.315 3.920 x 10' 1.48 x 10.4
TABLE 3--Continued.
abcdefg" Slope b Intercept b e, ~
1010010 0.214 1.824 x 103 4.20 x 104

1010110 0.555 1.511 x l(t3 N/A
1011010 0.186 2.799 x 10 .3 N/A
1011110 0.552 1.421 x 10 .3 N/A
1000010 0.209 9.710 x 104 3.25 x 104
1000110 0.358 1.042 x 10~3 N/A
1001010 0.208 9.550 x 104 3.25 x 104
1001110 0.338 1.652 x 10 .3 N/A
1110001 0.204 1.057 x 10 .3 2.15 x 104
1111101 0.359 1.040 x 10 .3 N/A
1111000 0.318 5.380 x 104 2.25 x 104
1110100 0.310 8.300 x 104 1.50 x 104
1101001 0.292 5.120 x 104 2.10 x 104
1100101 0.424 6.620 x 104 N/A
1100000 0.371 1.180 x 104 1.35 x 104
1101100 0.345 3.490 x 104 1.63 x 10 "4
1110010 0.447 1.851 x 10 .3 N/A
1110110 0.538 1.845 x 10 .3 N/A
1111010 0.465 1.846 x 10 .3 N/A
1111110 0.644 1.849 x 10 .3 N/A
1100010 0.257 1.236 x 10 .3 4.20 x 104
1100110 0.306 2.382 x 10 .3 N/A
1111010 0.297 1.384 x 10 .5 N/A
1101110 0.409 1.912 x 10 .5 N/A
1110000 0.228 4.850 x 104 1.11 x 104
1110101 0.302 1.959 x 10 .3 N/A
1100001 0.340 1.760 x 10 "4 2.32 x 104
NOTE:
a a
= A g g r e g a t e T y p e (0 = R B 1 = R L )
b = A g g r e g a t e G r a d a t i o n (0 = M e d i u m 1 = C o a r s e )
c = A s p h a l t T y p e (0 = A A K - 1 1 = A A G - 1 )
d = A s p h a l t C o n t e n t (0 = L o w 1 = High)
e = A i r V o i d s C o n t e n t (0 = 4+_.0.5% 1 = 8___0.5%)
f = T e m p e r a t u r e (0 = 104~ (40~ 1 = 140~ (60~
g = Stress Level (0 = 10 psi (69 k P a ) 1 = 20 psi (138 k P a ) )
b S l o p e a n d i n t e r c e p t at 0.1 s e c o n d f r o m the r e l a t i o n s h i p b e t w e e n t h e
a c c u m u l a t i v e p e r m a n e n t strain a n d t h e i n c r e m e n t a l l o a d i n g t i m e in a
l o g a r i t h m i c scale.
~ R e c o v e r a b l e strain during 200th cycle o f d y n a m i c loading.
d N o t available d u e to the early failure b e f o r e d y n a m i c loading.
TEST PROTOCOL
Fatigue
A servo-hydraulic system with a closed loop control was used in this research.
The testing fixture was originally designed for the determination of resilient
modulus of asphalt concrete in accordance with ASTM Test Method for Indirect
Tension Test for Resilient Modulus of Bituminous Mixtures (D 4123) [13]. The
schematic presentation of this fixture is shown in Figure 2. The f'Lxture was
installed inside an environmental chamber in which temperature could be
maintained within _+I~ for extended periods of time.
The horizontal and vertical displacements were measured by extensometers

with full scale travels of 0.15 in. (0.381 cm) and 0.16 in. (0.4064 cm), respectively.
Ranging the transducers to 10% of full scale calibrates the output to a finer scale
travel and allows higher resolution measurements of small deformations. The
operation and merits of this device are discussed elsewhere [11,13].
The haversine load with 0.1 second load duration and 0.5 second rest period
was repeated until the specimen "failed." The stress amplitude was kept constant
ECIMEN
VERTICAL DI
GAGE
W FRIC~ON
~OTATE FIXTURE
LATERAL DEI
EXTENSOME
A N D SPECIMEN
XTURE
LOWER EXTENSION ROD
FIG. 2--Schematic of diametral fatigue testing apparatus.
throughout testing, and time histories of horizontal and vertical deformations were
recorded at the 200th cycle. Two load levels, low and high, were used which
would induce the failure of the specimen under a given condition at about 100,000
and 10,000 cycles, respectively. The recoverable horizontal strain in the middle
plane (Ae0) was determined [11] from the following:
Ae o = 0.5235Ab u
where ASH is the recoverable horizontal deformation measured by horizontal

extensometers.
Some preliminary testing was conducted to identify the failure criterion for the
given testing conditions. Irrespective of the test temperature and the applied load,
it was found [11] that the horizontal deformation increased dramatically after a
value of 0.1 in. (0.254 cm) of horizontal deformation. Even though the dramatic
increase in horizontal deformation occurred earlier than this value in some cases
(e.g., tests done at 32~ (0~ the difference in the number of cycles between the
point where dramatic increase in horizontal deformation occurred and the point
with the horizontal deformation equal to 0.1 in. (0.254 cm) was relatively small. It
was thus concluded that the failure of the material in the controlled-stress
diametral fatigue testing at 32~ and 68~ could be described as the moment when
the total horizontal deformation reached 0.1 in. (0.254 cm).
The recoverable horizontal strain during unloading of 200th cycle was plotted
against the number of cycles to failure on a log-log scale. A linear behavior
between logarithms of the recoverable horizontal strain and the number of cycles
to failure confirmed the commonly used power form of the fatigue model. The
main reason for using the recoverable horizontal strain in lieu of the maximum
total horizontal strain is that the recoverable horizontal strain results in less
variation from the fatigue line than the maximum horizontal strain does [1_!1]. The
measurements from the 200th cycle were used to allow preconditioning of the
testing system and specimens.
The uniaxial incremental static creep test was performed in accordance with
the VESYS method [14] to determine the permanent deformation characteristic of
the asphalt concrete. Axial deformation was measured by using two
extensometers which were held by upper and lower clamps directly fastened to the
4 in. (10.16 cm) by 8 in. (20.32 cm) cylindrical specimen with a gauge length of 4
in. (10.16 cm). These extensometers were the same ones as used in measuring
horizontal deformations of diametral specimens.
Before the actual test loading, the specimens were preconditioned by applying
one ramp load with the same magnitude of the actual test load and holding the
peak load for one minute followed by a 5-minute unloading period. If the total
axial deformation exceeded 0.0025 in. (0.098 mm) during one-minute loading, the
loading was stopped and followed by a 5-minute unloading period. Then a series
of loading and unloading was applied to the specimen with rest periods, and the
total permanent deformations were measured after each rest period. The
accumulative permanent strain versus the incremental loading time was plotted in
a logarithmic scale. Finally, the best-fit line was constructed through the data
points.
DATA ANALYSIS METHODS
The test results are summarized in Tables 2 and 3 for fatigue and permanent
deformation, respectively. Before these results are discussed, it is important to
describe several sequential steps taken during the data analysis. A typical way of
evaluating the effect of a certain variable is the graphical comparison of the data
from two levels of the variable while the other variables are fixed in one level. In
other words, wanting to evaluate the effect of aggregate type on the fatigue
performance of the mixture with AAG-1 asphalt, low asphalt content, and low air
voids at 68~ (20~ one needs to plot the fatigue test results (strain versus
number of repetitions to failure) of the test combinations 01001 and 11001 in a
logarithmic scale. Then the difference between two fatigue curves is used to
determine the significance of the effect of the aggregate type on the fatigue
performance. However, because of the nature of the half-factorial design used in
both the fatigue and permanent deformation testing, it was infeasible to evaluate
the effect of one variable by directly comparing graphs.
Another way of investigating the effects of the several variables from the half-
factorial design is the analysis of variance (ANOVA) test with the generalized
linear models. The ANOVA test procedure employs the F-value as the test
statistic to test the null hypothesis that all the slopes or intercepts of fatigue or
permanent deformation lines are the same. The level of significance (p-value) for
this test is the probability of having F-value larger than the calculated F-value
from a data set for the factor in question. Smaller value of this probability implies
the heavier weight of the sample evidence for rejecting the null hypothesis. That
is, a p-value of 0.001 on slope from one variable indicates a much more significant
effect on the performance than a p-value of 0.5 on slope from another variable.
The ANOVA tests for equality of slopes and intercepts of performance lines
(logarithms of horizontal strain versus number of repetitions to failure for fatigue
and logarithms of permanent strain versus incremental loading time for permanent
deformation) were conducted to statistically determine which variables have
significant effects on the fatigue or permanent deformation of asphalt concrete.
Because the p-values are relative statistical numbers, a critical p-value for
significance had to be determined using engineering judgement for this study.
Therefore, plots were made for each variable depicting performance lines of two
levels of the variable regardless of levels of the other variables. The difference in
general trends of groups of performance lines between two levels was compared
with the p-value from the ANOVA test. The comparisons made for all the
variables and interactions led the researchers to conclude a p-value of 0.05 as a
critical number. Therefore, for the remainder of this paper, effects of each
variable is considered to be significant if a p-value on either slope or intercept
from the variable is less than 0.05.
Once the variables with significant effects on the performance were identified,
average performance lines at two different levels were plotted using mean values
of slope and intercept for each significant variable. This graph was then used to
determine which level of the variable resulted in better performance meaning
longer fatigue life or smaller permanent deformation. The same approach, as
discussed above, was made for interaction of different variables. Since the scope
of this paper covers the effects of aggregate related parameters on the
performance of asphalt concrete, only those interactions involving aggregate type
and gradation were investigated.
DISCUSSION OF FATIGUE TEST RESULTS
ANOVA test results for diametral fatigue testing are summarized in Table 4.
Factors including aggregate type, asphalt type, and temperature had p-values less
than 0.05 indicating significant effects of these factors on the fatigue performance
of asphalt concrete. Also, the A_NOVA results revealed that, with the given size
of experiment, no interaction with the aggregate type appeared to have significant
TABLE 4--Summary of ANOVA test results from diametral fatiaue testin.cl.
Level of Significance (p-value)
Dependent Variable
Factor Slope Intercept a
Aggregate Type 0.006 0.250

Asphalt Type 0.004 0.189
Asphalt Content 0.513 0.659
Air Voids Content 0.197 0.954
Temperature 0.027 0.011
Agg. Type x AC Content 0.984 0.885
Agg. Type x Air Voids 0.129 0.989
Agg. Type x Temp. 0.480 0.463
NOTE: ~The values are the intercepts at the number of cycles to

failure equal to 1000 cycles.
effects on the fatigue. It should be noted however that a larger size of

experimental design could show greater significance of the factors or interactions.
Average fatigue curves of recoverable horizontal strain versus number of

repetitions to failure are shown in Figure 3 for different aggregate types. The
data indicated that the mixtures with RB aggregates were more resistant to
fatigue. This difference can be explained by a rough surface texture of RB
aggregates. That is, the rough texture of RB aggregates provides better interface
bonding between aggregates and binder, and therefore, better fatigue resistance.
This is well demonstrated in Figure 4 showing the fracture surfaces of diametral
fatigue testing specimens at 32~ (0~ The careful examination of the fracture
surfaces in Figure 4 revealed that the fracture of RL mixtures under the repetitive
loading occurred through the interfaces between aggregates and asphalt, probably
due to poor bonding on the smooth and polished surface of RL aggregates. On
the other hand, the fracture of RB mixtures progressed through the coarse
aggregate particles indicating very strong bond between the aggregates and
asphalt. At 68~ (20~ no aggregate failure was observed, regardless of
aggregate type.
This observation suggests that, at 32~ (0~ where the fatigue is of greater
concern, the strength and toughness of coarse aggregate particles play an
important role on the fatigue failure of the mixtures with rough-textured
aggregates, whereas the interface bond strength controls the fatigue failure of the
mixtures with polished aggregates. Knowing that the strength of aggregate
particles is much higher than the interface bond strength, one can easily
understand the greater fatigue resistance of RB mixtures demonstrated in Figure
3.
Recov. Horiz. Strain
1.0E-03
-.8 /
..... RL
1.0E-04
1.0E-05 I i I B I hi I I I ~ ]IILL I d I I I ] iJ
1.0E*03 1.0E*04 1.0E+05 1.0E*06

Number of Repetitions to Failure
FIG. 3--Effect of aggregate type on fatigue resistance.
FIG. 4--Failure surfaces of specimens from the diametral fatigue tests:

(a) with RL aggregates (b) with RB aggregates.
DISCUSSION OF PERMANENT DEFORMATION TEST RESULTS
Table 5 presents the ANOVA test results from the permanent deformation
test data. Among the seven main factors investigated, only aggregate gradation
did not fall under the critical p-value of 0.05. As stated earlier, the coarse
gradation had the same nominal maximum aggregate size but a greater proportion
of coarse aggregates. Again, the half-factorial design of the permanent
deformation testing might not have been large enough to show the significant
difference due to change in gradation.
Effects of changing aggregate type and gradation are depicted in Figure 5

using mean slopes and intercepts of the permanent strain versus incremental
loading time lines in a logarithmic scale. The permanent strain of the RB
mixtures was significantly smaller than that of the RL mixtures. This can be
explained by the fact that, at high temperatures as used in the permanent
deformation study, the applied compressive load to the specimen is mainly carried
by the interlock among aggregates. Since better interlock is provided by the
aggregates with an angular shape and a rough surface texture, smaller permanent
deformation is expected from the RB mixtures.
Among the interactions presented in Table 5, the following interactions

appeared to have significant effects on permanent deformation: aggregate type
and gradation, aggregate type and asphalt type, aggregate type and air voids, and
aggregate type and temperature. None of the interactions with gradation was
TABLE 5--Summary of ANOVA test results from permanent deformation testing.
Level of Significance (p-value)
Dependent Variable
Factor Slope Intercept"
Aggregate Type 0.0001 0.0001

Aggregate Gradation 0.149 0.549
Asphalt Type 0.039 0.0001
Asphalt Content 0.373 0.014
Air Voids Content 0.001 0.0001
Temperature 0.117 0.0001
Stress Level 0.911 0.0001
Agg. Type x Gradation 0.405 0.011
Agg. Type x AC Type 0.393 0.014
Agg. Type x AC Content 0.947 0.638
Agg. Type x Air Voids 0.002 0.553
Agg. Type x Temp. 0.041 0.0001
Agg. Type x Stress Level 0.475 0.653
Gradation x AC Type 0.921 0.883
Gradation x AC Content 0.371 0.810
Gradation x Air Voids 0.243 0.437
Gradation x Temp. 0.118 0.576
Gradation x Stress Level 0.991 0.631
NOTE: "The values are the intercepts at the incremental loading time
equal to 1 second.
found important from the given size of experimentation. Again, mean slopes and
interactions were used to generate Figures 6 through 9.
Figure 6 indicates that the trends of change in permanent deformation with

gradation changes are opposite for the two aggregate types. The greater
permanent deformation from the coarse gradation of the RL mixtures is probably
due to a larger amount of round, polished coarse particles which result in poorer
interlock, and therefore, larger permanent deformation.
The interaction between aggregate type and asphalt type is depicted in Figure
7. Two important observations can be made from Figure 7" one that there is
larger permanent deformation of the mixtures with AAG-1 asphalt, and the other
that the effect of change in asphalt type is greater in the RL mixtures than in the
RB mixtures.
Permanent Strain
1.0E-01
-- -=-- RB i
-
--Coarse
RL
-- M e d i u m
I '~fJ~
1.0E-02
__ I I 1 1 I [ I II I I I I ~ I ! II I 1 I I 1 I I [
1.0E-03
10 100 1000
Incremental Loading Time (seconds)
FIG. 5--Effect of aggregate type and gradation on permanent strain.
Permanent Strain
1.0E-01
~- RB, Medium
- +-
....
--
RB, C o a r s e
RL, M e d i u m
RL, C o a r s e
I
|
1.0E-02
__ ] l [ J J I ill I I I t I I Ill I I I I [ I f
1.0E-03
10 100 1000
FIG. 6--Interaction between aggregate type and gradation for permanent strain.
Permanent Strain
1.0E-01
-- RB, AAK-1
RB, AAG-1
..... RL, AAK-1
- *- RL, AAG-1
1.0E-02 J
z
j ~ ~ : S . . . Z~ " i j t ! 1 1
J I I L I I II] l I I I I I III I ] I I I I J 1
1.0E-03
10 100 1000
FIG. 7--Interaction between aggregate type and asphalt type for permanent
strain.
Permanent Strain
1.0E-01
=-- RB, L AV
- - RB, H AV ,J
.... RL, L AV
-'- RL, H AV
1.0E-02
1.0E-03
1 10 100 1000
incremental Loading Time (seconds)
FIG. 8--Interaction between aggregate type and air voids for permanent strain.
Permanent Strain
1.0E-01
RB, 104 F
f
J
J
RB, 140 F /
/
..... / f
RL, 104 F /
/
- - - /
RL, 140 F /
/
J
1.0E-02 7"
j // / f .
1.0E-ON
_ _ _ _ _ 1 i i i t I *It I I I i i I III I I I I I I II
10 100 1000
FIG. 9--Interaction between aggregate type and temperature for permanent
strain.
The larger permanent deformation in the AAG-1 asphalt mixtures is due to a

lower viscosity of the AAG-1 asphalt at the test temperatures; the viscosity of
AAG-1 asphalt at 140~ (60~ is approximately half of that of AAK-1 asphalt.
The viscosity of asphalt binder would have a larger influence on permanent
deformation of the mixtures with a round particle shape, because more load has
to be carried out by the binder due to poor interlock among aggregates. This
explains the second observation made above.
Same argument can be used for explaining the interactions between aggregate
type and air voids in Figure 8 and between aggregate type and temperature in
Figure 9. As shown in these figures, the differences in air voids and temperature
affect the permanent deformation of the RL mixtures more significantly than
those of the RB mixtures. The poorer interlock among RL aggregates makes
asphalt-aggregate mixtures more dependent on the strength or hardness of asphalt
binder. Also in Figures 8 and 9, as expected, higher air voids and higher
temperature result in larger permanent strain for both the aggregate types.
CONCLUSIONS
Based on the half-factorial statistical experiment employed in this study, the

following conclusions could be made from the results of the diametral fatigue tests
and uniaxial incremental static creep tests:
. Aggregate type has a significant effect on the fatigue resistance of asphalt

mixtures, demonstrating the importance of better interface bonding between
asphalt and aggregates of a rough surface texture.
. With the given size of experimentation, no variable seemed to interact with

the aggregate type in the fatigue testing.
. The effect of aggregate type was found to be significant on permanent

deformation of asphalt concrete. Better interlock from angular particles
resulted in lower permanent deformation.
. Changing the proportions of fine and coarse aggregates with the same nominal
maximum aggregate size did not affect the permanent deformation
significantly.
. The effects of other variables, such as asphalt type, air voids, and temperature
on permanent deformation were more amplified with the RL mixtures. This is
probably because the load bearing capacity of the mixtures with poorer
interlock among round aggregates depends on the viscosity of a binder much
more than the mixtures with better interlock among angular aggregates.
ACKNOWLEDGEMENTS
The statistical analysis for this paper was conducted under the guidance of
Professor Francis Giesbrecht in the Department of Statistics at the North Carolina
State University. The authors are grateful to the Strategic Highway Research
Program for its support and sponsorship of this research.
REFERENCES
[_1] Dukatz, E. L. Jr., "Aggregate Properties Related to Pavement Performance,"

Proceedings of Association of Asphalt Paving Technologists, Vol. 58, 1989,
pp. 492-5O2.
[2] "Effects of Aggregate Size, Shape and Surface Texture on Properties of
Bituminous Mixtures," Proce.edings of 47th Annual Meeting of Highway
Research Board, HRB SP 109, 1970.
[3] Monismith, C. L., Epps, J. A., and Finn, F. N., "Improved Asphalt Mix
Design," Proceedings of Association of Asphalt Paving Technologists, Vol.
54, 1985, pp. 347-406.
[4] Emery, J. J. and Johnston, L., "Rutting in the Urban Context and Its
Mitigation," proceedings of Canadian Technical Asphalt Association, Vol.
32, 1987, pp. 254-283.
[5_] Kalcheff, I. V. and Tunnicliff, D. G., "Effects of Crushed Stone Aggregate
Size and Shape on Properties of Asphalt Concrete," Proceedings of
Association of Asphalt Paving Technologists, Vol. 51, 1982, pp. 453-483.

[6] Sousa, J. B., Craus, J., and Monismith, C. L., "Summary Report on
Permanent Deformation in Asphalt Concrete," Report TM-UCB-A-003A-
89-5, Institute of Transportation Studies, University of California, Berkeley,
CA, July 1990.
[7] Brown, S. F. and Pell, P. S., "Repeated Loading of Bituminous Materials,"
Report CAPSA 74, National Institute for Road Research, Republic of South
Africa, Durban, 1974.
[8] Uge, P. and van de Loo, P. J., "Permanent Deformation of Asphalt Mixes,"
Koninklijke/Shell-Laboratorium, Amsterdam, November 1974.
[9] Davis, R., "Large Stone Mixes: An Historical Insight," Report IS-103,
National Asphalt Paving Association, 1988.
[10] Harvey, J., "Asphalt Concrete Specimen Preparation Protocol," SHRP
Asphalt Project A-003A, Version 2.0., University of California at Berkeley,
May 1990.
[11] Kim, Y.R., Khosla, N.P., and Kim, N., "Effect of Temperature and Mixture
Variables on Fatigue Life Predicted by Diametral Fatigue Testing," To be
published in Transportation Research Record, Transportation Research
Board, 1991.
[12] Harvey, J., Sousa, J.B., Deacon, J.A., and Monismith, C.L., "Effect of Sample
Preparation and Air-void Measurement on Asphalt Concrete Properties,"
To be published in Transportation Research Record, Transportation
Research Board, 1991.
[13] Fairhurst, C.E., Kim, Y.R., and Khosla, N.P., "Resilient Modulus Testing of
Asphalt Specimens in Accordance with ASTM D 4123-82," Proceedings of
RILEM Fourth International Symposium on Mechanical Tests for the
Characterization, Design and Quality Control of Bituminous Mixes, Edited
by H.W. Firtz and E. Eustacchio, Budapest, Hungary, October 1990, pp. 402-
418.
[14] "Predictive Design Procedures, VESYS Users Manual - An Interim Design
Method for Flexible Pavements Using the VESYS Structural Subsystem,"
Report PB-285 488, U. S. Federal Highway Administration, Washington, D.
C., January 1978.
James M. Matthews and Carl L. Monismith
THE EFFECT OF A G G R E G A T E GRADATION ON THE CREEP RESPONSE OF ASPHALT

M I X T U R E S A N D P A V E M E N T R U T T I N G ESTIMATE'S
REFERENCE: Matthews, J. M. and Monismith, C. L., "The Effect of

Aggregate Gradation on the Creep Response of Asphalt Mixtures and
Pavement Rutting Estimates," Effects of A~Eregate and Mineral
Fillers on Asphalt Mixture Performance, ASTM STP 1147, Richard C.
Meininger, Ed., American Society for Testing and Materials,
Philadelphia, 1992.
ABSTRACT: The objective of this paper is to determine the influence of

aggregate grading on asphalt mixture performance in rutting on heavy duty
highways. For this purpose, the Hveem stabilometer method and simple
creep tests (unconfined) were utilized. In both methods, Watsonville
granite aggregate and AR 4000 and AR 8000 asphalts were used to cast
specimens by kneading compaction. Two types of aggregate gradations were
used: (I) California standard 0.5 inch (13 nun) maximum size, medium
gradation and (2) California standard 0.5 inch (13 mm) maximum size,
coarse gradation. The rut depths were predicted for a total of 15 cases,
consisting of 4 temperature conditions, 2 aggregate gradations and 2
asphalts. The conclusions drawn, from the results of both methods of
measuring performance (stabilometer and creep tests), were that (i) the
medium graded mixtures provided significantly better performance in
rutting than the coarse graded mixtures, and (2) temperature has more
influence o n rutting than the aggregate grading.
KEY WORDS: asphalt pavements, asphalt mixtures, permanent deformation,

creep, stiffness, rutting, aggregates
INTRODUCTION
With increasing frequency, agencies in the U.S. are reporting

problems with premature distress in the performance of asphalt concrete,
which is used for new construction or for the rehabilitation of existing
pavement, involving both flexible and rigid types. Problems related to
fracture, both traffic and temperature related, permanent deformation, and
surface wear are being reported by user agencies at all levels; cities,
counties, state and federal [!] - One of the most common forms of distress
of asphalt concrete pavements is rutting (permanent deformation).
IAssistant Professor of Civil Engineering at Temple University,

Philadelphia, PA 19122.
2Robert Horonjeff Professor of Civil Engineering and Research

Engineer at the University of California, Berkeley, CA 94720.
329
Copyright by9 ASTM

Copyright 1992 Int'l
by (all
ASTMrightsInternational
reserved); Thu Dec 22 07:29:10 EST 2011
www.astm.org
PURPOSE AND SCOPE
This p a p e r attempts to determine the influence of aggregate grading on

asphalt mixture p e r f o r m a n c e in rutting on h e a v y duty highways based on
l a b o r a t o r y study results. The scope covers a design system appropriate
for h e a v y duty asphalt concrete pavements (AC) subject to traffic and
temperature conditions as those in the State of California. This work
also concentrates on the development of solutions to mitigate rutting.
This p a p e r does not cover the design p r o c e d u r e s for other forms of

distress (for example, fatigue, thermal cracking and ravelling). This
study is limited to h o t - m i x e d asphalt concrete, excluding, for example,
o p e n - g r a d e d friction courses and drainage layers. The m a t e r i a l s studied
in this investigation are commonly used in the State of California.
The m e t h o d o l o g y followed in this p a p e r is b a s e d on the procedure

developed by the Shell researchers and developed by Monismith. According
to the Shell researchers good correlations exist between the laboratory
and field related estimates [2] . Hence the correlation b e t w e e n laboratory
and field is not c o n s i d e r e d in this paper.
MATERIALS AND THEIR COMBINATION
The aggregates and asphalt used, their requisite p r o p e r t i e s along with

the test methods and criteria u t i l i z e d to define these properties, are
given in [3] . The properties of the desired mixture, with some of the
factors that influence these properties, are also given in [~] . From past
experience at the U n i v e r s i t y of California, it was seen that A R 4000 and
A R 8000 asphalt cements, with the p r o p e r t i e s (Table i), and Watsonville
granite aggregate with 0.5 inch (13 mm) m a x i m u m size m e d i u m gradation
(California Standard) pass these standards and requirements for the
preparation of asphalt concrete mixes for c o n s t r u c t i n g h e a v y duty
pavements (Fig. I) [4].
STABILOMETER TESTS
The H v e e m S t a b i l o m e t e r m e t h o d [5] was used to determine the design

asphalt content. A n average design asphalt content of 5.2 % (by weight of
aggregate) was selected for the m e d i u m gradation and 5.1% was u s e d for the
coarse gradation aggregate. These asphalt contents were derived
r e c o g n i z i n g the tolerance of q u a l i t y control of the field equipment used
during construction of the highway. The results for the m e d i u m and coarse
gradation aggregate are given in Figs. 2 and 3.
Most asphalt cement specifications call for • to • asphalt

cement. So these mixes have a p p r o x i m a t e l y the same asphalt cement
content.
SAMPLE PREPARATION, TESTING EQUIPMENT AND TESTING TECHNIQUES
Sample Preparation
Two types of mixes were p r e p a r e d using Watsonville granite aggregate

and A R 4000 and A R 8000 asphalt cements for the gradations selected. All
the samples were compacted using the Triaxial Institute Kneading
Compactor. For creep tests, cylindrical specimens, 4 inches (102 nun) in
d i a m e t e r a n d 9 inches (229 mm) in height, were initially compacted. Then
the material at the ends was trimmed off with a d i a m o n d saw to produce the
final specimens a p p r o x i m a t e l y 8 inches (203 ran) in height. The top and
b o t t o m surfaces were then capped with a thin layer of hydrostone to obtain
smooth surfaces for load application.
MAI-rHEWS AND MONISMITH ON CREEP RESPONSE 331
TABLE 1 PROPERTIES OF A S P H A L T S
PROPERTIES OF A S P H A L T S USED IN T H I S STUDY
AE-8000 AE-4000
CR88R-5009 CR88K-5020
Penetration at 77F, dmm 46 72

Viscosity at 140F, Poise 3786 ZL25
Viscosity at 275F, cSt. 425.8 350.8
RTFC Residue
Penetration at 77F, dmm Z9 37
Viscosity at 140F, Poise 8764 4882
Viscosity at 275F, cSt. 620.8 501.4
(Tests are performed by Chevron Research Corporation)
I00
90 -
I
I
// ?
r/ /
//
80
L~
:Z 70
///
///
~n
~n
<
60
///
50
V//
/,,'/
40
,// /.--co~s~ (EPPS)
<
5~
/.//
30 .... MEDIUM (EPPS &
O
5~ /7 / PRESENT STUDY)
20
/~ ~/ / COARSE (PRESENT
.~ -'- STUDY)
I0
"
0 'I(X) 270 200
III I II I I I I I ! I I It l II I I
I I 1
I00 50 30 16 O 4
SEIVE SIZES
FIG. 1 AGGREGATE GRADING CURVES
MATTHEWS AND MONISMITH ON CREEP RESPONSE 333
Test Procedure
Specific gravity and air void content were d e t e r m i n e d and cross

sectional m e a s u r e m e n t s were taken for all specimens before testing.
Creep Tests: Creep tests were p e r f o r m e d in axial compression in the

u n c o n f i n e d condition. The d e t a i l e d procedure followed in conducting the
creep test is given in [3] . Tests were p e r f o r m e d at temperatures of 77~
(25~ 100~ (38~ and 120~ (49oC) . An IBM PC was used to control the
load, record both the load and the deformation, and determine the
compliance as a function of time. Creep moduli, as a function of time,
were calculated as the quotient of a p p l i e d stress d i v i d e d b y total strain
at a p a r t i c u l a r time (Figs. 4,5). A total of 24 creep tests were
conducted. Three specimens were tested for each test condition and the
average value was computed.
DEVELOPMENT OF D E S I G N M E T H O D WITH RUTTING AS T H E M A J O R CONCERN
A general framework was developed by M o n i s m i t h et al. (Fig. i) [6] .

A procedure, d e v e l o p e d b y the Shell researchers [2] and m o d i f i e d b y
M o n i s m i t h et al. [!], is followed in this study for rut depth predictions,
m a k i n g use of the results of simple, u n c o n f i n e d creep tests. When
suitably transformed, observations of the development of rut depth exhibit
the same form as test results obtained from the laboratory creep test in
uniaxial compression [2] . The procedure followed in this research
investigation is the one that was suggested b y Monismith, et al (Fig. 6)
[~,Z] 9 Steps i, 2 and 4 (Fig. 6) have been discussed in p r e v i o u s
sections. Step 3 is beyond the scope of this paper. Step 5 is e x p l a i n e d
in the following section.
CATEGORIZATION OF T E M P E R A T U R E CONDITIONS OF T H E S T A T E OF C A L I F O R N I A
The temperature conditions of the State of C a l i f o r n i a were divided

into five categories (Table 2). D e t a i l e d mean m o n t h l y air temperatures
for all these categories were obtained [8] .
DEVELOPMENT OF P A V E M E N T TEMPERATURE PROFILES
For each temperature condition, b a s e d upon the research conducted by

the U.S. Environmental Data Service [8], the hottest M e a n M o n t h l y A i r
Temperature (MMAT) was selected. U s i n g Barber's equation [9] and solar
radiation data [i0], Mean M o n t h l y Pavement T e m p e r a t u r e s (MMPTs) were
obtained for each two-inch interval, from the surface to a depth of
twelve inches, and pavement temperature profiles were plotted, up to 30
inches (762 mm) depth. As an example, the v a r i a t i o n of temperature with
respect to depth for temperature condition 2 is shown (Fig. 7).
Temparature condition 5 was not a n a l y z e d because rutting is not a concern
at this cold temperature [3] .
DEVELOPMENT OF P A V E M E N T STIFFNESS PROFILES
For each temperature profile (Fig. 7), a c o r r e s p o n d i n g stiffness

profile was obtained, which describes the change in stiffness with respect
to depth of the A C layer, for each temperature condition (Fig. 8).
PREDICTION OF R U T D E P T H
Results of the Shell procedure [2] and m o d i f i e d b y M o n i s m i t h [~,Z] can

be stated to include only the permanent deformation in the a s p h a l t - b o u n d
CO
CO
5o?_ I I
I I I
I I I >
I I I
"0
I I I -I-
L >
. . . . . . . . . I_ L r-
45 I
i I
I
~i~ ' ~ "~ I
C
-n
I
_ _ |. . . . . . . . . . . m
40 I -o
.< I m
! ,
I--
bq ! .
I 0
I
'
I--
Z I , e >
hi
I_ I. . . . . . . . . . I. . . . . . . Z
o0 2 3 5 I 0
m
I
n
I
I "~ ~i- - - A V, BllOll
I I
I. . . . . . . . . . . I. . . . . . . . . . . . . . . "IF- - AI{ 40l)(I
30 I I 1
I I I
| I I
l I |
I I I
25 "1' I'--I I I "'1--'1--1--1 - I--I ""1 " i--i--r--l--! i i

4 5 6
PERCENt ASPHAI.T CONTEHr (B'( WEIGHT OF AGGREGATE)
FIG. 2 STAI31HTYVs. ASPHALT CONTENT

Aggregote: Medium
44 I
I I I
I I I I
I I I I
_ ~ ~ _ _ I i I
I t I
_ _ -L . . . . . . . . . I~ ~I
40
Y - - ~ - - v ' '
I ~ r * I
I I ~ I I
36.: _ L . . . . . . ~ ...... ~ - ~ - - -'
I I ~ 7 I
"r
I I 1~ I 11"I
, , ! \ i
I I I ~ I I>
. . . . . . . . . . i . . . . t. . . . . . . . . . _,L.-- -----~ . . . . i z
~ 32 I I ! ~ I
U I I I ~ I E
Q
' ' ' \A--- AR 8 0 0 0 Z
I I I
I I i E
AR 4 0 0 0
_1 . . . . . . . . . . . . L . . . . . . . . ..L . -- -I
28; I I i i "n-
I I I I O
I I I I i
I I I I C)
I I I l :D
m
24 -i--i--iTl--I--i-'l--r-I-i - , "1--~--I-,--,--, -'''-T''fI-l'~r ~ ' 1 - ~ ' ~ . . . . . . 4 m
"o
4 5 6
PERCENT ASPHALT CONTENT (BY WEIGHT OF AGGREGATE) m
fj~
"D
Q
z
FIG. 3 STABILITY Vs. ASPHALT CONTENT m
Aggregate,= Coarse
O)
Oa
.0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
l I.~]_IZI-IIIIIE_ ]--IZr-IIITtEZ-IZI_IqII~E Z]__izr_~iiiiE Z _iZiZl_~iiill
- ~q -I-I~ TITIF -- 9 - I - I ~ TIfF -- 9 - I - - 1 9 T I f F -- 3 - I - 1 7 T I T I F -- 3 -I-19 TITII
-1 ---- ~ . ~ I Z I Z I Z I Z I E -" ..J -'lZl_~.J J_II.IE Z ] Z I Z I - i ~-I~JE Z ] Zl-I-I ]IZIE Z ] -I-I-I J_l]li
~.i-~_ J _~lJllIJllllL _ J _I_I-ILI.LIL _ J --I--IJllllL _ J -I-IJllllL _ J -I-IJ LIJ-II O)
i~' I --I ~IlIll , I , ' ''"1 , I ' I 'Ill, , I , , IIlll , I I ' IIlII "0
"-r
T~- ] - - I - I i - ~ T I F - ] - I - I - I TI]-Ii- - q - i - ( - i TITIF - ] -I-I-I TITIF - 9 -J-13 TITli >
- ~ --~'~--~l'-J~l l l ~ - - -1 - I - I - I J - I l l I -- J - - I - - I J L I Z I L -- J - I - I J .LIII I- - J _ I - I J .L I l l l -q
I I~l~l I't-Illl ~l I I Illlll I I I I Illll I I I I Illll I I I I Illll E
I I I~II~... ~ I IIIIII I I I IIIIII I I IIIIIII I I IIIIIII X
-d
1 0 . l ~ -- -1 ~ - ! - - J J - _'H2J.~_- .~l-__~.l_~-~J.~lJJL _ J _ l _ l J ] - I J A L _ J -I-iJ IlilL _ J _ l _ l J .I-I~11 C
---6~ ~ _ - 1 - - 1 3 I I I I # _ - I - - _ I ~ L ~ -- ] --I_--I3 I I ] 3 [ -- ] - I [ t - I I l I I C [ ] [I]l]IlIII
"~ ] " ~ - - I - - I ~ T I T I C -- 3~"~ - - ~ Th~ __~I--17 TI~ F -- 3 - I - 1 3 T I T I F -- 3 - - I - - I ~ T I T I I Ill
] ] , , . ~ ] _-'l-Zl-I Z I Z I L __ ] Z l - - I ~ ' ~ z t ~ C --'~~ - - - - - - ~ - ' - I ~ ] I Z I E Z ] ZiZl_-i illlL _- ] - - I - I _ - I i i l l i
Ill
J ~1-~...~_ ~ J.-LLIL _ J _ l _ l J J.I II1-.~ " ~ . l ~ l . . ~ t . ~ " ~ L . . . _ J _ l _ l _ l J. I I I L _ J _ l _ l J J. IJ.II "ll
/ ~41 I~l"t--LJJl"l" ~ I I I I I I I I I I I ~ ~ 1 1 1 1 1 1 1 I I I IIIIII -rl
U) o
ill -'/. ~ " ' 1 T I ~ I I - - ~ " ~ r ~ - - I . - ~ . ~ L J ~ " " ~ " T ~ t ~ - - _ , ~ _ "~ --I--I--I T I T l i
z 8 "-_ ~_ , . = ~ ~ ~ _T~~-~'t~,.L--._.
_~__ ,._, .,_u.,, E
hi' o I I I IIIII I / ~1 f=rl~ ~_ 1 ~ - 1 4 i -- -- ~ 1 i iiiii z
i i iiiiiii i~ i iiiiii i ~ i - ' ] ~] i1]1 ~ ~ ~ 1 ~ 1 ~ i i i iiii 0
b~ l O i. _ _1 _ l _ l J J_ILIL _ J _l_lJ II~JL _ / _l_lJ LIIIL _ / _~_.~ ll/l~_~'j _ l J 11111 m
-- -I ZI--I3 IIIIE Z ] ZI_-I-I IlIIC Z ] ~1-'13 III]E Z ] -I-13 IlIIE _- -l -I-I-I IlIII
- -I --I--I-I TITIF -- "1 --I--I--I T I T l l " -- "1 --I--I--I / l ' r I F -- -I - - l - I ' l % l ' l T l F -- -I --I--I-1TITII
- ] Zl_-i-I ZI~_,C - ] -i_-i-I i,]-IC - "-i _-I_-I-I ZiZiC - -i -i-i'-I ILT.i i" - ] "-i-t'-I J. iili
_ J _l_lJ iI.LIL _ J _I_I/.LI.LIL _ J _1_1_1 i l / I L . _ J _1_1_1 .LI~I._ _ J _l_iJ /I.LII
_ _1 _ 1 _ 1 _ 1 L i / i L _ J _ l _ i _ i L i L i L _ i _l_l...i I_1/I L _ j _1_1_1/iLl L _ i _1_1_1 _.l I_li!
I I I I IIIII I I I I IIIII I I I I IIIII I I I I IIIII I I I I IIIII
_ J _l_l_J • _ J _l_lJ LIZIL _ J _l_IJ • _ J _ l i I . J J.I.LID _ J _1_1_1 .LI•
I I I I IIIII I I I I IIIII I I I I IIIII I I I I t1111 I I I I 11111
I I I III'!I I I I IIIIII I I ! IIIIII I I I IIIIII I I I IIIIII
10' I I IIH-ttl I I llItttl I I llllttt H-lllh'tl I t-H-t-Nil
10-' 1 I0 101 10 ~ -I0
TIME (Seconds)
FIG. 4 INFLUENCE OF AGGR. GRADATION AND TEMPERATURE

O N CREEP SFIFFNESS A spahall. Type: AR8000
Solid Lines : Med. Aggr. ( 1 ) 7 7 F ( 2 ) 1 0 0 F (3)12_0F (4)140F
Dashed Lines: Coarse Aggr. (5)77F (6)100F (7) 120F (8)140F
10
- 5 -i-i_5 7_EiE - 5 -i-iS u E - 5 -i-iS ~i-Z E: - 5 --i-iS ~EiF_ - 5 --;--iS T:i~i i
-- "1 ZI-l-I 1-11"If- - "1 -I-'I] TI_TI I- - ] -IZI-I 1-I_11 I- - -I -" I-I_-I TIT_I[ - 3 _-I-I-1 "TIT•
_ _i _i_L_i •177 _ J _i_i._i,Li.U L _ _i _ L i . J • _ -i _ I _ i ] • _ _i _ I _ i ] •
-- J - I - I - I . L I • _ J --I--I-I.LI ill_ _ i -I-IJ ,LI I l L _ J - I - I - I lille _ J - I ~ - I . - 1 •177
__1_1 i_ll_li_l L__l_i_l_l_ili_l L__i_l_l_ll_l/I L_.!_I_I_I • L_/ I I l/l!•
I I I IIIIII I i i IIIlil I I I IIitll I I I lillll I I I ililll
L ' " t -" .l .- .I .-~I -II l - l / I L ' - / - I - I - ' l ' L l l l L - / - I - l / • I li• ..1 I I-lll/ll
IIIII I I I IIIIII I I I I IIIII I-I-I I IIIII - I-I-I I IIIII
)--"I-.I'1"-~IIIII I I I I IIIII I I I I IIIII I I I I IIIII I I I I IIIII
10 _ J_ _ - ~ I ~ _ i _i_,J • _ d _l_li • L _ i _i_Ii LillE _ d -l_ii •
2 - _n -!-!-~ ~zlq_'z-.q.~_-.,-,-~ z , ~ 1- - -1 - - I - - 1 3 I I T i I - -- -1 : l - - l - - I I l I l r " _-- -I --l_--l--I I I I l l
n *-'~ [iEl_.11"lZi_P~]~ TI3"11 '~ -- "1 --I--I-1 1-1rip -- -1 --i--I-1TITIF -- G -I-- -1 T 1"11
9" . . . . ~ ~ - 2-,_.-; ~ z ] _ - , z , ] z , z , c _- ] - , - , - , z,z, c___- ] _-,:,_-1 z,z,i
"J~iliL--l--I--I.~-J'~--~;--iJ/i.l..IL_/ I_,Jii• j ' 'Jl,lli "F
FIG
IS) ~ ~ ' _ - _ , . _ ~ , Z , , 1, ,,,,1 , , , , ,111,
i,i
Z
~ ~ TJ.,~-i-,-, T,7,i- - i - , - , - , T,T,i
I.L ~ ' L - _ . - ~ ; L _ 9 _ ,_,- . , . , , >
t~ o ~ ~ _--_ - - - - - - ~ ,,,,,,, Z
I I I IIIIII -I- I J IIIIII I I - - ~""~ - --t'~'~l~-"-'~"-""'~'~l I IIIII
O~
lO ~ ~ ~ _ - ' _ - , . 9 ,~,,
0
-- -1 Z l - I - i I l I I [ - -I E I Z l ' - I I i r i 1- _~ "1"~1-_13 I i ' r i c _- -1 - 1 _ - 1 " 7 I l I I C -" "I ~ " I'-I I l l l Z
-- "1 - - I - - I . 1 T I T I F -- "1 - - I - - 1 " 1 1-1311- -- -1 - - I h l q T I T I 1 - -- "1 - - I - - I - 1 T I T I I - - -- -1 ~'1-- i'-i 1 - i ' r i l
z ] -_,:,-, Z,.T_,~- _- ] - , - , ] T,T,C - ] - ,- ,]'~r C - i _-i-;-; .t;iii- - ..i -1-15 ~.;Z"
-I
_ J _l_lJllllL _ i _l_lJ.illl/ _ d _I_IJilIIL _ J _I_IJ• _ / _l_lJ/lil "r
_ / _l_l_lii/i L _ j _ll_l I_l i__lL _ j _ l _ l _ l / l • L _ / _1_1_1 i_l/i L _ j _1_1_1 _ l l / i
0
I I I IIIIII I I I I IIIII I i I I IIIII I I I I II111 I I t 1 IIIII Z
_ ..I _l_lJ • _ d _1_1_1 • L _ i _l_lJ l.lllL _ J _ I_lJ .LI.I.IL _ / _l_lJ 11111
' i ' lilii' ' i ' 'Jill, ' ' i 'Jill' ' i i iilii, ' i i ' 'ill, 11
m
I I I IIIIII I I I IIIIII I I I I IIIII I I I I IIIII I I I I IIIII m
lllll ~ I I IIIIII I "13
I0 "-I I l llllll I I I Iiliill
'' , ,,,,,,i , I IIIIIII I1 I IIH-ttl ~D
1 0 "' 1 10 10 I 10 ~ I0 ' m
(#3
TIME (Seconds) "D
0
Z
FIG. S I N F L U E N C E O F AGGR. GARDATION A N D TEMPERATURE m
ON C R E E P S F I F F N E S S A.sDahalt T y p e : AR . 4 0 0 0 ..
Solid Lines : M e d . A g g r . ( 1 ) 7 7 F (2)1.00F (3)I.20F (4J140F
D a s h e d L i n e s : C o a r s e Aggr. ( 5 ) 7 7 F (6)100F (7) 120F (8)140F O0
TABLE 2 CLASSIFICATION OF OVERALL TEMPERATURE CONDITIONS
AVERAGE
YEARLY
AVERAGE AIR TLMP.
YEARLY FOR THE RANGE OF AIR T--vM.P.
AIR TEMP. GROUP JAN. JULY
CITIES BY GROUPS DEGREE F. REPRES . MIN. MAX.
VERY HOT
Death Valley 75.3
Needles FAA Airpor~c 73.2 73.2 52.1 97.2
(This is taken as Representative city for Group I)
Barstow 69.0
II HOT
Bakersfield 63.3
Modesto 61.4
Fresno 63.9 63.9 43.0 86.0
(This is taken as Representative city for Group If)
III M O D E R A T E L Y HOT
Sacramento Airport 59.9 59.9 42.4 77.0
IV COASTAL
San Diego 64.5
(Extreme South)
Eureka 52.4
(Ex~:reme North)
San Francisco 58.6 58.6 50.0 64.1
(This is taken as Representative city for Group IV)
COLD
Eureka 52.4 52.4 47.9 58.6
MAT'I-HEWS AND MONISMITH ON CREEP RESPONSE 339
L Select ospho/t grode end

oggregate choro cter/st/cs
1
T I
2. Select ospholt conlent at I

y mix anticipated end '" -4 ~
I
mix/S wqter
sens/t/ve
3. Check mix for woter sensit/v/t
y l
I IO. Zncorporote
additive
I 4. Perform creep tests at ]
7 7 ~ - a n d I00~-
I
rut depth/s
5. Perform ru~/ing onalys/s using excessive
modified SHELL procedure for I
r when Fair > 7;'~- 7
15Q Fede'Ign mi'; I
rut depth is consider use
o/Modifier
~ tolerqb/e
6. Check for fatigue

I" r~rtdepth/s
/olerob/e
Stop
FIGURE 6 DESIGN SYSTEM FOR ASPHALT-AGGREGATE MIXTURE

W I T H R U T T I N G A S D O M I N A N T C O N C E R N (4)
0 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 0
I
l
>
I 00
"U
l -r
l >
l -q
l
30 I ~- 7 -I l X
--4
C
II _I Il Il ll ~3
m
O9 I I I I "U
I I I I m
I I I I 33
I I I I 0
33
I - __ - J L . . . . . . l l
>
I ~ I I I z
I ~ I I I C)
m
I ~ I I I
C3 I ~ I I I
I ~ I I I
I ~ I I I
I ~ I l l
10 84 . . . . . . I . . . . .. . . . . . ~. . . . r . . . . . . . . . . . . . . . . . . . . . . . . I
I ~ - I I
I I ~ I I
I I ~ I
I I
I I I I
I I I ~ I
0
0
70 80 go I00 I 0 120 150 140
PAVEMEN[ II:MI'II,~AIIII,:I (I)
FIGURE 7 DISTRIBUTION OF TEMPERATURE IN THE PAVEMENT FOR
GROUP 2 TEMPERATURE CONDITIONS
16.00 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
4 3 2 1
12.00 - -- -- F -- -- F - - - - - I- -
I
I
If) I
w I
"1- I
r
Z I
I
- -- -- L L . . . . _ - --L _ _ L _ _ , __L _ L _ m
"~ 8.00
I 6o
T
I--"
I
13_ I z
LI_I
I
I
0
I z
I
4.00 F - - -I- - - F - - --[-I-- -H
I
I 0
1 z
m
m
0.00
-/-ij /
m
10 10 2 6o
STIFFNESS (,lO00 PSl) 0
z
6o
FIGURE 8 DEPTH Vs. STIFFNESS PROFILES FOR AC GROUP 2 m
TEMPERATURE CONDITIONS
O~
AR 4000; Medium; AR 4 0 0 0 ; Coarse 4~
..&
layer:
O" a w
Aht = C M " h~ ..... , (1)
S~
where
h I = thickness of the a s p h a l t - b o u n d layer,
~ = average stress in the a s p h a l t - b o u n d layer,
Smi x = m i x stiffness determined in [~],
CM correction factor for the s o - c a l l e d "dynamic effect"

which takes account of d i f f e r e n c e s between static
(creep) and dynamic (rutting) behavior. This factor is
dependent on the type of m i x a n d has b e e n found
e m p i r i c a l l y to be in the range 1 to 2.
If the value of C M was taken as equal to unity, and if it was d e s i r e d

to subdivide a thick a s p h a l t - b o u n d layer into a n u m b e r of sublayers, the
e q u a t i o n could be stated as:
n (ffa~) i ]
Ah i = ~ hi
...... (2)
i=l (S~) i
The E L S Y M computer p r o g r a m [ i i ] was u s e d to analyze the asphalt

concrete layer. A dual wheel configuration, each wheel of 4,500 ibs.
(2041 kgs), with i00 psi (689 N / m s ) tire pressure, radius of contact of
3.79 inches (96 mm), and a free space of 3.79 inches (96 mm) b e t w e e n the
two wheels, was u s e d in this study. A n axle has 2 sets of such
configuration.
The asphalt concrete layers were d i v i d e d so that the average stiffness

could be d e t e r m i n e d in the sublayers to reflect the temperature gradient.
For accurate p r e d i c t i o n of rut depths, the first four inches (102 mm)
of asphalt concrete (where the m a j o r p o r t i o n of rutting occurs) were
d i v i d e d into four I inch (25 mm) thick sublayers, a n d the next four inches
were d i v i d e d into two 2 inch (51 mm) thick sublayers. From the seventh
sublayer onwards, the thickness was d e t e r m i n e d in such a w a y that the
t e m p e r a t u r e change w i t h i n the a s p h a l t i c sublayer is •176 (0.5~ 9 A total
of 15 case studies were analyzed for the four types of material
combinations, u n d e r the four temperature conditions. Since A R 4000
asphalt w i t h c o u r s e - a g g r e g a t e g r a d i n g is not c o n s i d e r e d a suitable m i x for
v e r y hot temperature conditions, it was not u s e d in the case where MMPTs
rise more than 130~ (54~ This is b e c a u s e of excessive creep
d e f o r m a t i o n at one hour loading time.
To determine the average stresses in a sublayer, it is n e c e s s a r y to

estimate mix stiffness at a loading time interval of 0.i sec
(corresponding to r e l a t i v e l y s l o w - m o v i n g traffic). Stiffness values at
this rate of loading and Poisson's ratios (over the temperature range in
pavement) are shown (Table 3) for temperature c o n d i t i o n 2. The stiffness
values were o b t a i n e d from the creep test results (Figs. 4 a n d 5).
To estimate the amount of r u t t i n g from Eq. I, it is n e c e s s a r y to have

the stiffness of the mix, as d e t e r m i n e d from creep tests as a function of
time of loading (Figs. 4 and 5). From 1 hour creep tests, m i x stiffnesses
as a function of depth were obtained (Fig. 8). The resulting computations

for rut depth detez~inations were made for temperature conditions 1
through 4 (Table 3).
Provided everything remains the same, for heavy-duty asphalt pavements

(in this study, all AC layers were more than 14 inches thick), final rut
depth is not affected by an increase in the thickness of the AC layer or
by a change in the subgrade modulus [12].
SUMMARY OF DESIGNS WITH RUTTING AS THE MAJOR CONCERN
Predicted rut depths for all 15 conditions are given (Table 4). For
all these pavements, total ruts on the asphalt concrete are in the range
of 0.04 (1 mm) to 0.06 inch (1.5 nun), which is well within the allowable
limit (considering that 0.4 inch (10 mm) is the allowable limit to
mitigate hydroplaning).
Therefore, all of the asphalt concrete heavy-duty highway pavements

studied in this study did pass the rutting criteria. A check for pavement
fatigue is beyond the scope of this paper and therefore has not b e e n
included. However, detailed calculations for pavement fatigue design are
shown in [3].
Both methods (stabilometer and creep tests) validated better

performance in rutting for medium graded mixes than for coarse graded
mixes.
A review (Table 4) indicates that environment is more important than

the gradation for rutting prediction. For example, environment 1
increases rutting by 86% in comparison to environment 4 when same asphalt
(AR 8000) and aggregate gradation (medium) are used. There is only 15%
increase in rut depth due to the change of aggregate gradation (coarse to
medium) for the same environment (environment 1) and asphalt (AR 8000).
Similar results were observed for the remaining cases (Table 4).
ACKNOWLEDGMENTS
Thanks are due to Sonali P. Khukde and A. Satyanarayana for

organizing and typing the manuscript on time.
REFERENCES
[1] Epps, J.A., Civil Engineering, April 1986, pp. 56.
[2] Rao, C.S., "Development of asphalt aggregate mixture analysis

system." NTIS Report 9006594, Doctor of Engineering Dissertation,
University of California, Berkeley, 1989.
[3] State of California, Department of Transportation, "Standard

Specifications", Sacramento, CA, July 1984.
[4] State of California, Department of Transportation, Test Methods No.

366, "Materials Manual", Vol. 2, Sacramento,CA, 1978.
[5] Monismith, C.L., Epps, J.A., and Finn, F.N., Proceedings, Association
of Asphalt Paving Technologists, 1985, pp. 347-406.
[6] Shell International, "Shell Pavement Design Manual", London, 1977.
[7] Finn, F.N., Monismith, C.L., and Markevich, N.F., Proceedings,

Association of Asphalt Paving Technologists, 1983, pp. 121-150.
[8] U.S. Environmental Data Service, Department of ConEnerce, "Monthly

Normals of Temperature, Precipitation and Heating and Cooling Degree
9
0
I I I I I I I I I I I I - I I
I I I I I I I I I I I I I I I I
it i I t I I t tt o t I I I
T rl-~- i--T - T - -I- ~-j-

l
~-'3-
i I ,r-- i!~~
'
t
G
0
I I I I I I I I I I I I I
off
(a~] .-
I I I I I I I I 1 I I I I I
I 1 I I I I I I I I I I I I Q
L L - - I - - I - - L - l - - .L _ -- L. I--I--
I -C _ 0
I I I I I
I I I I I ,
I I I I I I .00
~ 0 0
I I I t I > 0 0
I I rq- 7' - -
I I I I I ' rY r ~ r,1
I I I I I I D < <
I I I I I
4.
, w w
I I I I I I
I I I I I I
I I I I I I I
r~ ~- .~ ._~
I I I I I I
I --C-----'-o e-
Fi 'C~- i" g
I
I
I
I
I
I
I
I
I
I
l
I
"0~
I I I I I r :o < O o
-'I" - - I- - - - =Q
I I I I I I I , ::?
"n
I I I I I I I I
I I I I I I I <
t 0
Is I - - I. . . . . -0
l ---I--~-'L-I
~ lI I - .l__
__llI I t d
I I I I I I I
I I I I H c~ m~ II I I I I
I I I I I II I I I I 0 d
LI.
T rl-r I--I- ~-- r - - - n I"I-i-i- i- -I 3"
I I I I I I I I I i i i I I I I
I I I I I I I I I I I I I I I I I
I I I I I I I I i i i i i I I I I 0
III I I I I I I!II I I I I 1 o
0 0
Os ) SS N US
TABLE 3 RUT DEPTH PREDICTION FOR TEMPERATURE CONDITION 2
AR 4000; MEDIUM AGGREGATE GRADATION
VERTICAL
STRESS RUT DEPTH
LAYER DEPTH STIFFNESS A T THE STIFFNESS COL.(Z)X
O F CENTER POISSONS AT 0.i C E N T E R OF AT COL.(6) =
Thick- O F LAYER TEMP. RATIO SECOND O F LAYER 1 HOUR COL.(7)
No. ness
in. in. de E . F *i000 psi psi *i000 psi inches i

m
*0.001
Go
>
1 1.0 0.5 129 0.46 32 99.85 9.2 10.85 z
2 1.0 1.5 120 0.44 43 96.95 i0.6 9.15 0

z
3 1.0 2.5 113 0.43 54 87.52 11.6 7.54
.-I
4 1.0 I
3.5 107 0.41 64 75.93 12.5 5.9 Q
z
5 19.0 11.5 87 0.41 103 16.16 14.3 21.47 0
~D
11"i
6 8.0 N~A -- 0.55 20 (untreated base layer) l'rl
'1-i
7 Infinite N/A -- 0.35 i0 (subgrade layer) m
Q
TOTAL RUT DEPTH~ 0.0549 inches (1.37 mm.) z
cD
m
TABLE 4 SUMMARY OF RUT DEPTH PREDICTIONS
Environment with Mix

Identification RUt Depth
Environ-
SI mental T7pe of Aggregate
No. Condition Asphalt Gradation Inches mm
1 1 AKS000 Medium 0.053 1.3
2 1 AE8000 Coarse 0.060 1.5
3 1 A/~4000 Medium 0.067 1.7
4 2 AK8000 Medium 0,042 1.0
5 2 AE8000 Coarse 0.049 1.2
6 2 AR4000 Medium 0.055 1.4
7 2 AK4000 Coarse 0.066 1.7
8 3 AE8000 Medium 0.040 1.0
9 3 AES000 Coarse 0.047 1.2
i0 3 ARS000 Medium 0.051 1.3
Ii 3 ~4000 Coarse 0.067 1.7
12 4 AE8000 Medium 0.030 0.7
13 4 ARS000 Coarse 0. 035 0.9
14 4 KK4000 Medium 0.036 0.9
15 4 AE4000 Coarse 0.053 1.3
Days: 1951-80", 1982.
[9] Barber, E.S., Bulletin No. 168, Highway Research Board, Washington
D.C., 1967, pp. 1-8.
[1o] State of California, Department of Water Resources, Division of

Resources Development, "Solar Radiation Measurements in California,"
Sacramento, CA, January 1974.
[11] Ahlborn, G., "ELSYM5, Computer Program for Determining Stresses and
Deformation in a Five Layer System". University of California,
Berkeley, 1991.
[12] McLean, D.B., "Permanent Deformation Characteristics of Asphalt

Concrete", Ph.D. Dissertation, University of California, Berkeley,
1974.
STP1147-EB/Oct. 1992
Author Index
A K
Abdulshafi, O., 281 Khosla, N. P., 310

A1-Fozan, F., 107 Kim, N., 310
Ali, N. A., 3 Kim, Y. R., 310
A1-Mudaiheem, J., 107
A1-Suhaibani, A., 107 L
Anderson, D. A., 131, 154
l.Agar6, P.-P., 177
Lytton, R. L., 295
B
M
Bahia, H. U., 131
Bayomy, F. M., 19 Matthews, J. M., 329
Bergan, A. T., 3 Monismith, C. L., 329
Bouchard, G. P., 35
Brown, E. R., 45, 68 P
Button, J. W., 295
Papagiannakis, T., 3
Parker, F., 68
C Perdomo, D., 295
Chan, J. S. S., 3 R
Chapuis, R. P., 177
Chrxstensen, D. W., III, 154 Ruth, B. E., 252
Cross, S. A., 45
S
D Saloukeh, B. G., 187

Sanders, C. A., 90
Dongre, R., 131, 154 Shahrour, M. A., 187
Dukatz, E. L., 90, 154 Shen, X., 252
Shuler, T. S., 225, 265
F T
Fwa, T. F., 211 Tan, S. A., 211
Theriauh, E. G., 3
H W
Huber, G. A., 225,265 Wang, L.-H., 252
349
Copyright 9 1992 by ASTM International www.aslm.org
STP1147-EB/Oct. 1992
Subject Index
A Bituminous concrete, 35
Bonding surface volume, 35
Absorption, water, 35
Aging, oxidative, 131 C
Aggregates, 90, 154, 187
angularity, 295, 310 Calcite, 131
blends, 252 Cement, 187
cement coated, 19 coating, 19
compaction, 265 portland, 107, 177
cracking, effect on, 281 CEMPHALT, 19
crushpercentage, 35 Coating, cement, 19
face, fractured, 45, 68, 90 Compaction, 265, 329
flow test, 45 Concrete
gradation, 295 asphalt, 3, 19, 68, 107, 187,
creep response, effect on, 329 265, 295
curves, 177 bituminous concrete, 35
refusal density, effect on, 265 Cracking
rutting, effect on, 45, 68, 310 aggregate effect on, 281
void space and, 225 cold temperature, 3
mineral, 131, 154, 225, 252 fatigue, 131
size, 45 potential, 281
steel slag, 3 Creep, 3, 310, 329
surface area, 177 Crushed gravel, 68, 90, 177
surface texture, 187, 295, 310 Crushers, aggregate, 35
texture, 68
Air void content, 225, 252, 310 D
Angularity, aggregate, 295, 310
ANOVA, 310 Deformation, permanent, 3, 19, 252
Arterial mixes, 35 aggregate e,:ffectson, 45, 68,
Asphalt cement, 45, 225 310, 329
Asphalt concrete filler effect on, 107
aggregate effect on, 68 potential, 90, 211
cement coated aggregate in, 19 predicting, 295, 329
cracking, 281 Deformation, plastic, 252
filler effect on, 107, 187 Density, refusal, 265
predicting performance, 295, 310
refusal density, 265
steel slag in, 3
Asphalt mastics, 131, 154 Fatigue, 3, 19, 154
ASTM standards cracking, 131, 281
C 204:177 resistance, 310
Filler-bitumen system, I87
Fillers, 177, 187
B asphalt mastics, 131, 154
mineral, 154, 252
Binders, 281 Flow test, aggregate, 45
Bitumen cement, 187 Fractal dimension analysis, 295
351
Fracture mechanics, time P

dependent, 281
Fracture percentage, gravel, 90 Plastic deformation, 252
Fracture toughness test, 155 Portland cement, 107
C 204:177
G film, 19
Gradation, aggregate, 295 Q

creep response, effect on, 329
curves, 177 Quartz, 131
refusal density, effect on, 265
rutting, effect on, 45, 68, 310
void space and, 225 R
Gravel fracture percentage, 90
Gyratory testing machine, 45, Resilient modulus, 3, 107
252, 265 Rheometer, bending beam, 131
Rutting, 252
aggregate effect on, 45, 68,
H 310, 329
filler effect on, 107
Hveem stabilometer method, model, 295
107, 265, 329 potential, 90, 211
Hydration time, 19 pre.dicting, 295,329
Hyperbolic model, 295 resistance, 3, 19
K S
Kneading procedures, 265,329 Sand, 177

angularity, 90
L angular natural, 68
Shear, gyratory, 252
Lime, hydrated, 107, 187 Shear stress, octahedral, 295
Limestone dust, 107 Shear, under load, 35
Load energy, 281 Sieve size, 225
Load intensity, increased, 68 minus No. 200, 154
Load speeds, 211 passing No. 200, 187
Single edge notched beam test,
154
M Size, 45
sieve, 154, 187, 225
Marshall properties, 90, 107, 265 Slag, steel, 3
stability, 35 Sodium silicate filler, 187
Mastics, asphalt, 131, 154 Stability, 107
Mineral filler, 131, 154, 225,252 Stabilometer, 107, 265, 329
Moisture damage, 3, 19 Steel slag, 3
Stiffness, 329
Stone, crushed, 68
N Storage modulus, 154
Strategic Highway Research
National Aggregate Association Program, 131
flow test, 45 Surface area, filler, 177
INDEX 353
Surface texture, 310 Voids

filler, 187 air, content, 225, 252, 310
quantifying, 295 characteristics, 265
characteristics, fillers, 154, 187
T content, 45
mineral aggregate, 35, 225, 252
Tensile strength, indirect, 3, 107
Tension test, direct, 131
Time dependent fracture
mechanics, 281 W
V
Water absorption, 35
Variance, analysis of, 310 Water cement ratio, 19
Viscoelasticity, 154 Wheel tracking test, 107, 211
ISBN 0-8031-1468-0

Effect of Aggregates - BM - STP1147

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Effect of Aggregates - BM - STP1147

Uploaded by

Copyright:

Available Formats

Copyright by ASTM Int'l (all rights reserved); Thu Dec 22 07:29:10 EST 2011

Effects of Aggregates and

Richard C. Meininger, editor

ASTM Publication Code Number (PCN)

Effects of aggregates and mineral fillers on asphalt

Peer Review Policy

Printed in Ann Arbor, MI

T h e U s e o f Steel Slag in A s p h a l t i c C o n e r e t e - - N . A. ALl, J. S. S. CHAN,

D e v e l o p m e n t and Analysis of C e m e n t - C o a t e d A g g r e g a t e s for A s p h a l t M i x t u r e s - -

Effects of A g g r e g a t e Absorption and C r u s h P e r c e n t a g e on Bituminous C o n c r e t e - -

Selection of A g g r e g a t e P r o p e r t i e s to M i n i m i z e Rutting of Heavy Duty

Evaluation of P e r c e n t F r a c t u r e of Hot Mix Asphalt Gravels in lndiana--c. A.

Effect of Filler Type and Content on P r o p e r t i e s o f A s p h a l t C o n c r e t e M i x e s - - A . AL-

R h e o l o g i c a l P r o p e r t i e s of M i n e r a l F i l l e r - A s p h a l t Mastics and Its Importance to

A Simple M e t h o d for D e t e r m i n i n g t h e Surface A r e a of F i n e A g g r e g a t e s a n d Fillers

Laboratory Evaluation of Rutting Potential of Asphalt MixturesuT. F. FWA AND

Providing Sufficient Void Space for Asphalt Cement: Relationship of Mineral

Gyratory Evaluation of Aggregate Blends to Determine Their Effect On Shear

Effect of Aggregate Size and Other Factors on Refusal Density of Asphalt

FATIGUE, MODELING, AND THEORETICAL

Effect of Aggregate on Asphalt Mixture Cracking Using Time-Dependent Fracture

A New Approach for Prediction of Permanent DeformationuD. PERDOMO,

Effects of Aggregate Type and Gradation on Fatigue and Permanent Deformation

The Effect of Aggregate Gradation on the Creep Response of Asphalt Mixtures

Author Index 349

Subject Index 351

SHRP and Other Research

Importance of Mixture Performance

THE USE OF STEEL SLAG IN ASPHALTIC CONCRETE

ABSTRACT: Advanced testing and evaluation were performed on laboratory prepared

KEYWORDS: steel slag, mechanical properties, pavement performance

1Assistant Professor, Department of Civil Engineering, Technical University of Nova

The objectives of this study are:

A P P L I C A T I O N O F STEEL SLAG IN ASPHALTIC CONCRETE

The main disadvantages of steel slag asphalt mixes am:

ADVANCED TESTING AND RESULTS

An IBM XT compatible micro-computer was used to provide signals to the MTS

TABLE 1 -- Aggregate Properties

COARSE AGGREGATES I'--

0.8 2.6 4.94 13

Absorption 2.750 2.350 0.590 1.870 0.820 0.940

Soundness (MgSO4,%) 10.7 18.7 12.6

Specific Gravity (ASTM C128):

Absorption 5.120 4.920 1.377 1.380 1.860 1.130 1.690 1.570

28000 (1.06 in.) 100.0 .--4

An environmental chamber, capable of providing air temperatures between -40~ and

Resilient Modulus Test

TABLE 3 -- Diametrical Resilient Modulus (MPa)

A 22000 19500 4200 3250 1050 930

Creep and Permanent Deformation Test

Low Temperature Characteristics

The low-temperature tensile fracture strengths of briquette specimens were determined

Moisture Damage Test

10 -~ i I IIIIII I I I IIIIII I I L ttl~l= I I I IIIIIl l

0.1 1 10 100 1000

FIGURE 1 -- Permanent Strains at 0 Degree Celsius

FIGURE 2 -- Permanent Strains at 20 Degrees Celsius

10 -s I I I I Jill ] I I I i lit I I I I I I III I I ~ I I I Ill 1

0.1 I 10 100 1000

FIGURE 3 -- P e r m a n e n t Strains at 44 Degrees Celsius

10 I I I I Illl[ I l I I IIll I I I I I Illl I I I I I Elll I

0.1 1 10 100 1000

FIGURE 4 -- Creep M o d u l u s at 0 Degree Celsius