Professional Documents
Culture Documents
Characterization of Subgrade Soil Mixed With Recycled Asphalt Pavement
Characterization of Subgrade Soil Mixed With Recycled Asphalt Pavement
ScienceDirect
highlights
Effect of recycled asphalt pavement (RAP) in the subgrade has been investigated.
The resilient modulus of RAP mixed soil increases with the applied stress.
The resilient modulus of RAP mixed soil is less sensitive to confining pressure.
The resilient modulus value reaches a maximum at the optimum moisture content.
Article history: Due to awareness to greener environment, use of recycled asphalt material has become
Received 19 August 2016 popular in asphalt pavement industry. The use of recycled asphalt pavement (RAP) ma-
Received in revised form terials in subgrade soil can be an additional scope of utilizing RAP. This study investigates
14 March 2017 the effect of RAP on the resilient modulus (MR) of subgrade soils mixed with RAP materials.
Accepted 15 March 2017 Note that MR is the principal material input parameter for designing asphalt pavement
Available online 26 April 2018 using the recent mechanistic-empirical pavement design software. As a first step of the
current study, different percentages of RAP and moisture were thoroughly mixed with
Keywords: subgrade soils. Then, the MR of these RAP mixed soils were determined using the AASHTO
Subgrade T 307 (1999) at different stress levels in the laboratory. Results show that the MR of RAP
Recycled asphalt pavement mixed soil increases with the applied deviatoric and bulk stresses, however, it is less
Moisture sensitive to applied confining pressure. Use of RAP materials has made the soils stiff
Resilient modulus enough not to respond to the confining pressure. As expected, the MR value reaches a
Stress maximum at the optimum moisture content and increases linearly with RAP content. The
MR values and characteristics of the RAP mixed subgrade soils, as determined by the
current study, can be used for subgrade design and stabilization using RAP for better
pavement design.
© 2018 Periodical Offices of Chang'an University. Publishing services by Elsevier B.V. on
behalf of Owner. This is an open access article under the CC BY-NC-ND license (http://
creativecommons.org/licenses/by-nc-nd/4.0/).
* Corresponding author. Tel.: þ1 719 549 2612; fax: þ1 719 549 2519.
E-mail addresses: mehedihasan@unm.edu (M.M. Hasan), md.islam@csupueblo.edu (M.R. Islam), tarefder@unm.edu (R.A. Tarefder).
Peer review under responsibility of Periodical Offices of Chang'an University.
https://doi.org/10.1016/j.jtte.2017.03.007
2095-7564/© 2018 Periodical Offices of Chang'an University. Publishing services by Elsevier B.V. on behalf of Owner. This is an open
access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
208 J. Traffic Transp. Eng. (Engl. Ed.) 2018; 5 (3): 207e214
Use of recycled asphalt pavement (RAP) materials obtained The primary objective of this study is to determine the effects
from the removal of existing asphalt pavement material of RAP and moisture on the MR of subgrade soils. In particular,
during resurfacing, rehabilitation, or reconstruction opera- the effects of confining pressure, deviatoric stress and bulk
tions in pavement construction is increasing day by day stress are investigated. The following steps are followed to
(Alam et al., 2010; Apeagyei and Diefenderfer, 2013; ARRA, fulfill the objectives.
2014; Bemanian et al., 2006; Bhavsar, 2015; Cross and Jaka-
timath, 2007; Diefenderfer and Apeagyei, 2014; Kim and Lee, The RAP was mixed with the collected subgrade at five
2006; Kim et al., 2007, 2009; 2010, 2011; Ma et al., 2015; Mar- different RAP contents (0, 25%, 30%, 50% and 75% by weight
tinez-Arguelles et al., 2015; McGarrah, 2007; Kim et al., 2010). of the mixture).
In New Mexico, a maximum of 50% RAP can be mixed with Then, cylindrical samples of 100 mm diameter and 200 mm
virgin aggregate for preparing a base course. A maximum of height were prepared using the AASHTO T 307 test
35% RAP can be mixed with aggregate for creating hot mix protocol.
asphalt concrete. Although there has not been any specifi- Four different moisture contents (5.7%, 6.3%, 7.1% and
cation or limits for using RAP in subgrade materials, yet RAP 7.6%) were used to prepare samples.
materials have not been used in subgrade soils because The MR of the prepared samples were analyzed at different
design parameters (which is MR) and characteristics of RAP deviatoric stress, bulk stress and confining pressure.
mixed subgrade are not known. As a result, a vast amount of
RAP materials end up at contractors' pits or landfills.
Therefore, development of new scope of RAP material is an
immense need. This study evaluates the design data and 3. Materials
characteristics of RAP in subgrade soils.
In the past, the behavior of RAP blended asphalt mixture The subgrade soils and the RAP were collected from the
and base course were widely explored in the literature interstate 40 (I-40) construction site at the mile post of 141
(Alam et al., 2010; Islam et al., 2014; Li et al., 2011a; b; Solanki near Albuquerque, New Mexico in cooperation with the New
et al., 2012; Tabakovic et al., 2010; Tomlinson, 2012; Visintine Mexico Department of Transportation (NMDOT). The sub-
et al., 2013; Wu, 2011). Li et al. (2011a,b) studied the effect of grade soil was collected from the milled section of the pave-
asphalt treatment on the resilient modulus (MR) of Alaskan ment and the RAP material was collected from the stockpile.
base course. Alam et al. (2010) investigated the effect of RAP The RAP was fractioned and screened by the contractor. These
content in base materials on the resilient modulus. They materials are described below.
found that MR increases with RAP content. However, they
did not study the effect of RAP in subgrade. Wu (2011) 3.1. Subgrade
accounted the effect of moisture content on the MR of base
material with RAP. In addition, models were proposed to Sieve analysis and Atterberg limits test were conducted to
account for the effect of temperature on the MR of base determine the soil property. Fig. 1 presents the grain-size
material with RAP. Permanent deformation was compared distribution curve of the subgrade soil. From the gradation
for specimens containing different percent-ages of RAP to and the Atterberg limits test, it is found that the soil was A-
evaluate the rutting potential. It was found adding RAP to 2-6 type. The percentages of materials passing through #4
virgin aggregate not only increased the MR, but also and #200 were found to be 96% and 4.5% respectively.
increased rutting potential under certain conditions.
Constant head permeability tests were conducted for
samples containing different percentages of RAP, and the
results suggested that coefficient of permeability
decreased with the increase of RAP. Bennert and Maher
(2005), MacGregor et al. (1999) and Papp et al. (1998) also
found that inclusion of RAP causes an increase in the MR
of base material.
From the above discussions, it can be concluded that RAP
can cause an increase in the MR of base material. However, the
effect of RAP in natural soil is almost unknown although this
might be another scope of RAP utilization. Evaluation of en-
gineering performance of subgrade material mixed with RAP
is important for pavement design and new constructions. To
this end, this study evaluates the modulus of subgrade
considering the effect of RAP and moisture content at
different stress levels in the laboratory. Fig. 1 e Grain-size distribution of the subgrade soil.
J. Traffic Transp. Eng. (Engl. Ed.) 2018; 5 (3): 207e214 209
3.2. Recycled asphalt pavement where V, N, W, and h are the volume of the specimen, number
of layers, weight of drop hammer, and drop height respec-
Fig. 2 shows the gradation curve of RAP material used in this tively. CE refers to compactive effort of 56,000 ft-lbf/ft3. A total
study. It shows that the NMAS was found to be 19.5 mm. of 27 impact blows in each of the 8 layers were applied. The
The percentages of materials passing through #4 and #200 sample diameter was 100 mm. The heights of the samples
sieves were found to be 52% and 0.3% respectively. From the were taken at least twice of the diameter. Fig. 4 shows the
ignition oven test, the binder content was found to be 2.98%. sample preparation procedure for resilient modulus testing.
Fig. 3 shows the gradation curve of mixed material which Fig. 4(a) shows the collected soil and RAP materials. Fig. 4(b)
consists of 70% subgrade soil and 30% RAP material. From shows the mixing of the soil and RAP. Fig. 4(c) shows the
the sieve analysis, the NMAS was found to be 19.5 mm. The compaction of sample. Fig. 4(d) shows the prepared sample.
percentages of materials passing through #4 and #200 sieves
were found to be 55.2% and 4.5% respectively.
5. Laboratory testing
4. Sample preparation
The MR test was conducted according to the AASHTO T 307
(1999) (16) test sequence for base/subbase materials. This
Subgrade and RAP materials were collected from the con-
method measures the elastic modulus of base and subbase
struction sites. To prepare the sample, first a fixed amount of
materials. A repeated axial cyclic stress of fixed magnitude
material is obtained and water is added to it and mixed
was applied to a cylindrical sample. During testing, the
thoroughly to achieve optimum moisture content. Then the
sample was subjected to a dynamic cyclic stress and a static
loose materials were compacted in eight layers using the dy-
confining pressure by means of a pressure chamber. Note
namic impact compaction by modified proctor effort of
that cell pressure and confining pressure both are the same
56,000 ft-lbf/ft3 into a split mold. The required number of
and interchangeably used in the current study. The total
blows per lift were calculated using Eq. (1) recommended in
recoverable axial deformation of the sample was measured
the NCHRP 1-28A (2004):
using linear variable displacement transducer (LVDT) and
CE$V used to calculate the MR value as shown in Eq. (2).
n¼ (1)
N$W$h
Scyclic
MR ¼ (2)
er
where Scyclic is the applied cyclic axial stress and can be rep-
resented by Eq. (3) and is the resilient axial strain as presented
in Eq. (4).
Pcyclic
Scyclic ¼ (3)
A
where Pcyclic is the applied cyclic load and A is the cross-
sectional area of the sample.
er
er ¼ (4)
L
where er is the resilient axial deformation due to Scyclic and L is
the original specimen length.
Haversine shaped load form of 1cosðqÞ is applied in a triaxial
Fig. 2 e Grain-size distribution of the RAP material. 2
pressure chamber. Each test sequence (1 s) has 0.1 s of load
pulse and 0.9 s of rest period. Confining and deviatoric stress
were applied as specified which are listed in Table 1. The test
setup is shown in Fig. 5. The sample was confined in the
pressure chamber and external LVDT was used. The vertical
deformation ratio of the two LVDTs was monitored for
accurate measurement.
To determine the effect of moisture content and RAP par-
entages on the MR value of the subgrade soil, sixteen samples
were prepared and tested in the laboratory. Two replicate
samples were prepared for each of the four different moisture
contents and RAP percentages on partially saturated subgrade
materials. Table 2 presents the test matrix used in this study.
It is worth mentioning that the sample preparation and
testing take huge amount of time. Therefore, a limited
Fig. 3 e Grain-size distribution of the mixed material (30% number of samples was tested; however it suffices the
RAP). purpose of this study.
210 J. Traffic Transp. Eng. (Engl. Ed.) 2018; 5 (3): 207e214
Fig. 4 e Sample preparation. (a) Soil and RAP. (b) Mixing. (c) Compacting. (d) Prepared sample.
the MR value increases with bulk stress. This is expected for 6.2. Effect of RAP content
subgrade soil as well as base material; the bulk stress
includes the effect of confining pressure which causes an Fig. 9 shows the effect of deviatoric stress on the MR value of
increased MR value. the subgrade soil as well as the mixed material having 75%
Fig. 6 e Resilient modulus with deviatoric stress (30% RAP). (a) Moisture content of 5.7%. (b) Moisture content of 7.6%.
Fig. 7 e Resilient modulus with bulk stress (30% RAP). (a) Moisture content of 5.7%. (b) Moisture content of 7.6%.
212 J. Traffic Transp. Eng. (Engl. Ed.) 2018; 5 (3): 207e214
Fig. 9 e Resilient modulus with deviatoric stress (7.1% moisture). (a) Soil 100%, RAP 0%. (b) Soil 25%, RAP 75%.
Fig. 10 e Resilient modulus with bulk stress (7.1% moisture). (a) Soil 100%, RAP 0%. (b) Soil 25%, RAP 75%.
J. Traffic Transp. Eng. (Engl. Ed.) 2018; 5 (3): 207e214 213
Solanki, P., Hossain, Z., Zaman, M., et al., 2012. Volumetric and Dr. Md Rashadul Islam is an assistant pro-
mechanistic characteristics of asphalt mixes containing fessor and the program coordinator of civil
recycled asphalt pavement. In: GeoCongress 2012. Oakland, engineering technology at the Colorado State
2012. University-Pueblo. He is also a licensed pro-
Tabakovic, A., Gibney, A., McNally, C., et al., 2010. Influence of fessional engineer (P.E.). Dr. Islam received
recycled asphalt pavement on fatigue performance of PhD in civil engineering with the distinction
asphalt concrete base courses. Journal of Materials in Civil categorized research, a Masters degree in
Engineering 22 (6), 643e650. civil engineering, and a B.S. degree in civil
Tomlinson, C.R., 2012. The Effect of High RAP and High Asphalt engineering. Dr. Islam has intensive
Binder Content on the Dynamic Modulus and Fatigue research experience in transportation geo-
Resistance of Asphalt Concrete (Master thesis). Virginia techniques. As of 2016, he has published 1
Polytechnic Institute and State University, Blacksburg. full-text book, 38 journal papers, 9 book
Visintine, B., Khosla, N., Tayebali, A., 2013. Effects of higher chapters, and 36 conference articles. Dr.
percentage of recycled asphalt pavement on pavement Islam is also the principal investigators of
performance. Road Materials and Pavement Design 14 (2), several funded research projects.
1e6.
Wu, M., 2011. Evaluation of High Percentage of Recycled Asphalt
Pavement as Base Course Materials (Master thesis). Dr. Rafiqul A. Tarefder is a professor and
Washington State University, Pullman. Regent's Lecturer of civil engineering at the
University of New Mexico (UNM), associate
director of two federally funded trans-
Mr. Md Mehedi Hasan is a PhD student with portation centers: SPTC, and SOLARIS. He is
a major concentration on civil engineering also a registered professional engineer (P.E.)
at University of New Mexico since August in the state of New Mexico and Idaho. Dr.
2014. His PhD research focuses on charac- Tarefder has been working on pavement
terization and modeling of unbound mate- materials and design for more than seven-
rials such as base, sub-base and subgrade teen years and published about 250 journals
materials in pavement industries. He and conference articles, 21 books, special
completed his B.S. degree in civil engineer- issues (edited), and book chapters, and 30
ing in 2012. His research findings have been technical reports. Dr. Tarefder has received
published in different journals and several over $11 million in research grants and su-
conferences. pervised/supervising 42 PhD/MS students.