REPORT BE96-3843/2001:12-5

Non-Destructive Assessment of
Mechanical Properties of Concrete at
Very Early Age by US Techniques
– Method, Results and Modelling

Matias Krauß1 Karim Hariri2 Ferdinand S. Rostásy3

1,2,3
Institut für Baustoffe, Massivbau und Brandschutz,
Technische Universität Braunschweig

PUBLISHED BY
DEPARTMENT OF CIVIL & MINING ENGINEERING
DIVISION OF STRUCTURAL ENGINEERING
ISBN 91 - 89580 – 12 – 5 • 2001:12-5 • SE
 - ii -

Report No:
S Non-Destructive Assessment of 2001:12-5
PAC Mechanical Properties of Concrete
I
at Very Early Age by US Techniques
Improved Production of
Advanced Concrete Structures – Method, Results and Modelling

Author Ferdinand S. Rostásy1 f.rostasy@tu-bs.de

Author Karim Hariri2 k.hariri@tu-bs.de
Author Matias Krauß3 m.krauss@tu-bs.de
Address1,2,3 IBMB, TU Braunschweig
Beethovenstraße 52
38106 Braunschweig, Germany

Task/Subtask T2.1/T2.2
o
Brite EuRam Contract No. BRPR-CT97-0437
Project no: Brite EuRam Proposal No. BE96-3843

Project title: IPACS - IMPROVED PRODUCTION OF ADVANCED

CONCRETE STRUCTURES

Project co- Betongindustri AB, Dr Mats Emborg

ordinator:

Partners: Betongindustri AB
Cementa AB
Selmer ASA
Technical University of Delft
ENEL
Technical University of Luleå
NCC AB
Skanska Teknik AB
Technical University of Braunschweig
Ismes
Norwegian Public Roads Directorate
Elkem AS
Norcem AS
Technical University of Trondheim

Date of issue of this report: February 2001

Revised date:
Project funded by the European Community under the Industrial & Materials Technologies
Programme (Brite-EuRam III)
 - iii -

IMPROVED PRODUCTION OF ADVANCED CONCRETE STRUCTURES - IPACS

Background
Research and practical experience show that the quality and lifetime of concrete structures largely
depend on the curing conditions in the concrete’s early life, as inadequate curing leads to malfunction
and cracking. A major source of deleterious cracking already in the construction stage is the
occurrence of stresses in the hardening concrete due to restrained volume change related to hydration
temperatures and shrinkage phenomena. It is thus of utmost importance, especially regarding new
high performance concrete, that the proper execution conditions are maintained throughout the
construction period by avoiding the premature cracking.

Objective of project
Main goal of IPACS is to evaluate, integrate and extend the existing knowledge about early age
concrete crack prediction in engineering practice yielding the following benefits:
Contractors and designers will have new and more reliable engineering instruments enabling them to
predict and to optimise the technical effect and cost of alternative designs and execution procedures -
all in the process of fulfilling the quality requirements set up by the owners or the community (codes).
Reduced costs because of the present tendency to specify costly but unnecessarily rigorous crack
criteria will be avoided.
Owners will have access to improved means of specifying and controlling desired quality
requirements regarding serviceability and service life of their structures.
Reduced maintenance costs and increase of service lifetime.

Main tasks and investigations in IPACS and output from the project:
Hydration and volume changes – To acquire data for the modelling of properties of a number of
currently used concrete types.
Mechanical properties - Testing and modelling of mechanical properties.
Behaviour of structures - Computer modelling of structural behaviour.
Field tests - To check and improve the models of the previous tasks in full-scale tests.
Expert System.

The Expert System synthesises the results from the project into a robust engineering tool for planning
and control of the production of concrete structures. It contains modules of varying simplicity, which
can be used in all the phases of a construction project from pre-design to maintenance

Project Partners:
See earlier page

Project Co-ordinator:
Dr Mats Emborg – Betongindustri AB (Heidelberger Zement North Europe) (SE)
Dr Hans-Erik Gram/Mr Mats Öberg – Cementa AB (Heidelberger Zement North Europe) (SE)

Disclaimer
The author/authors and producer of this report have used their best effort in preparing this report. Theses efforts
include the development, research and testing of the theories and programs to determine their effectiveness. The
author/authors and producer make no warranty of any kind, expressed or implied, with regard to these programs
or documentation contained in this report. The author/authors and publisher shall not be liable in any event for
incidental or consequential damage in connection with, or arising out of, the furnishing, performance, or use of
these programs.

Editorial/production supervision: Prof. Lennart Elfgren

Cover design:
Prepress material:
Printed and published by
Hans Hedlund
By report authors
Luleå University of Technology,
Department of Civil and Mining Engineering,
Division of Structural Engineering
SE-971 87 Luleå, Sweden
 - iv -

CONTENTS

1 INTRODUCTION ....................................................................................................................... 1

2 MODELING OF MECHANICAL PROPERTIES OF CONCRETE AT VERY EARLY

AGES .................................................................................................................................................... 2

2.1 PROBLEM AND SCOPE OF INTEREST ....................................................................................... 2

2.2 THE USE OF NDT TECHNIQUE................................................................................................ 4

3 MATERIAL PROPERTIES OF YOUNG CONCRETE......................................................... 4

4 DETERMINATION OF INITIAL DEGREE OF HYDRATION WITH THE

COMPRESSIONAL WAVE METHOD (CWM)............................................................................. 6

4.1 INTRODUCTION ...................................................................................................................... 6

4.2 COMPRESSIONAL WAVE METHOD .......................................................................................... 6
4.2.1 Application of the sound penetration technique and testing......................................... 7
4.3 TEST RESULTS OF CWM...................................................................................................... 10
4.3.1 Velocities of compressional waves .............................................................................. 10
4.4 DEVELOPMENT OF DYN E ACC. TO CWM............................................................................ 11
4.4.1 Influence of cement type .............................................................................................. 12
4.4.2 Influence of water - cement ratio ................................................................................ 13
4.5 DETERMINATION OF INITIAL DEGREE OF HYDRATION ......................................................... 14

5 DETERMINATION OF INITIAL DEGREE OF HYDRATION WITH COMBINED

SHEAR AND COMPRESSIONAL WAVE METHOD (SCWM) ................................................ 15

5.1 INTRODUCTION .................................................................................................................... 15

5.2 TEST SET-UP ......................................................................................................................... 16
5.3 ADIABATIC CALORIMETER ................................................................................................... 16
5.4 ULTRASONIC MEASUREMENT .............................................................................................. 16
5.5 TESTING OF AXIAL TENSILE STRENGTH AND STATIC MODULUS OF ELASTICITY .................. 18
5.6 CONCRETE COMPOSITIONS .................................................................................................. 18
5.7 MODELS AND TEST RESULTS ............................................................................................... 19
5.7.1 Test results................................................................................................................... 19
5.7.2 HPC ............................................................................................................................. 19
5.7.3 Material properties at early ages of HPC................................................................... 20
5.7.4 Application of NDT on HPC........................................................................................ 22
5.7.5 Improvement of iBMB – model.................................................................................... 23
 -v-

5.7.6 Improvement of Scandinavian model .......................................................................... 24

5.8 PARAMETER STUDIES REGARDING END OF DORMANT PHASE .............................................. 24

6 CONCLUSIONS........................................................................................................................ 25

7 REFERENCES .......................................................................................................................... 26
 - vi -

Abbreviations and Units

3
max QCo maximum heat release of concrete [kJ/m ]

max QCe maximum heat release of cement [kJ/kg]

cCo heat capacity of concrete [kJ/(kg⋅K)]

cCe heat capacity of cement [kJ/(kg⋅K)]

ρCc
3
density of concrete [kg/m ]

c1, tk parameter of modified JONASSON model [-], [h]

a, b parameter of Danish model [h], [-]

d, td0 parameter of Shrinkage Core Model [1/h], [h]

amount of cement, aggregate, fly ash, silika fume and

C, G, FA, SI, W [kg/m3]
water in concrete

α degree of hydration [-]

meas α measured degree of hydration [-]

cal α model of degree of hydration [-]

C3A, C2S, C3S, C4AF clinker phases -

meas ∆T measured adiabatic temperature rise [K]

maximal adiabatic temperature rise calculated by

max ∆Tad [K]
Bogue - Method

te equivalent age [h]

 -1-

ABSTRACT: For the prediction of the mechanical properties of early age concrete, as well
as for the prediction of thermal stresses it is essential to know the initial degree of hydration,
from which on the development of stiffness and strength begins and which denotes the so-
called dormant phase of concrete´s hydration. This paper deals with the determination of the
end of dormant phase by means of US-technique. With compression wave and shear wave
transducers the hardening of concrete is observed under adiabatic and isothermal curing
conditions. From the development of dynamic Young's modulus and Poisson's ratio a model
of the initial degree of hydration is derived to improve existing models of the development of
tensile strength and modulus of elasticity for very early age concrete. Typical results are
presented for different concrete compositions, especially for high strength concrete.

Keywords: ultrasonic testing, young concrete, setting and hardening, on-set time, degree of hydration,
Young’s modulus, Poisson’s ratio

1 INTRODUCTION

Ultrasonic testing for controlling the quality of metallic and composite materials is one of the
most successful and frequently applied testing methods. The development of electronics as
well as of the microprocessor techniques during the last 20 years led to considerable
improvements of automatic testing systems and their integration into production process. A
comparably successful application of the ultrasonic method for testing mineral building
materials has so far not been attained. The development of such systems starts at the early
50’s of the past century. Two main objectives can be discerned. The first one relates to the
application of ultrasonic testing technique to measure the development of mechanical
properties of concrete during hardening. The second one dealt with the assessment of
damage and cracks. An overview of activities in was reported in Whitehurst, Elvery, Casson,
Byfors and others.

Early reports of iBMB in the field of ultrasonic testing are presented by Eisenmann,
Steinkamp and Odewald between 1951 and 1956. Eisenmann et al. investigated the
influence of cracks in concrete on ultrasonic pulse velocity. In addition, they described the
difference between static and dynamic modulus of elasticity. This work was continued by
Kordina, Waubke and Roy. Kordina et al. investigated the development of ultrasonic pulse
velocity as the velocity of the transverse and longitudinal waves as well as compressive
strength increases. In 1974 Neisecke presented a new method to control the quality of
mineral building materials. He measured the transverse wave velocity, longitudinal velocity
and the intensity of the incoming signal. So in principle, it was possible to measure the
 -2-

development dynamic Poisson’s ratio and dynamic Young’s modulus simultaneously. The
field of application of this method was among other things the observation of hardening
process of concrete in the first 72 hours. Hillger continued this work. In 1983 Hillger
presented an enhanced method by which the spectrum of the incoming signal could be
analysed.

The next chapter gives a short survey of current field of application of the ultrasonic
measuring technique on cement based materials at iBMB. The two above-mentioned
publications form the basis of the further investigations presented this report.

In order to predict mechanical properties of concrete at very early age, non-destructive test
methods are recommended. Combining such methods with destructive tests reliable
information regarding the mechanical properties can be acquired. In this paper a test method
is presented to determine the initial setting time, at which the initial mechanical properties
like Young's modulus and tensile strength can be measured destructively. The method was
developed in order to improve existing models of mechanical properties models and to
obtain a better understanding of hardening concrete.

2 MODELING OF MECHANICAL PROPERTIES OF CONCRETE AT VERY

EARLY AGES

2.1 Problem and scope of interest

During the first hours and days after mixing, properties and compaction of concrete rapidly
change within a wide range of behaviour. Fresh concrete is a more or less Newtonian fluid
which gradually changes into a Bingham type of material, finally attaining solid body
properties with increasing compressive strength and stiffness. The development of
mechanical properties can be described by the degree of hydration. For the prediction of
mechanical properties of early age concrete as well as for the prediction of stresses caused
by differences of temperature and autogenous shrinkage, it is essential to assess the initial
degree of hydration, from which on the development of strength and stiffness begins.

Several researchers have shown, that the development of the mechanical properties can be
modelled with the degree of hydration or equivalent age. Especially the compressive
strength, the tensile strength and the modulus of elasticity shows a significant dependence
on the degree of hydration; e.g. viz. [33] and [36].
 -3-

All models as described in the literature depict the test results well, but they contain
uncertainties, in particular for a low degree of hydration. Figure 1 shows the dependence of
the modulus of elasticity and axial tensile strength on the degree of hydration as described
by models and also their real behavior. With α0 the end of dormant phase and the on-set of
strength and stiffness is denoted. It is determined by regression of measured tensile
strength data with a linear fct - α relationship stipulated a-forehand. But this method exhibits
two main technical difficulties. The first one relates to the fact, that reasonable results for the
tensile strength can only be measured if α is greater ≥ 0.4 , because of necessary
preparation time. Furthermore, the scatter of test results at very early age is rather large.
The second one relates to the difficulty to model the degree of hydration at very early ages
precisely enough. So, to obtain more precise values of α0i, the tests have to be performed
under adiabatic conditions and in- situ measurements of mechanical properties are
necessary. Figure 1 also shows that the evolution of mechanical properties sets on much
earlier than described by α0. From this point of view it is clear, to determinate the
corresponding smaller initial degree of hydration α0i a nondestructive testing method like
ultrasonic measurement is more advantageous.

It is a consequence of these facts that the solution of this problem requires a NDT method
such as the US technique because the velocities of ultrasonic pulses directly depends on the
development of Young’s modulus.

Figure 1: Normalised tensile strength and

modulus of elasticity dependent on degree of
hydration - models and real behaviour
 -4-

2.2 The use of NDT technique

It is well known that the dynamic Young’s modulus dyn E and the dynamic Poisson’s ratio
dyn µ depend on the ultrasonic pulse velocity of shear and compressional waves ([18], [25]).
The properties dyn E and dyn µ are defined by

1 − 2 (v s / vc )
2
dyn µ = , (1)
2 − 2 (v s / v c )
2

and

(1 + dyn µ )(1 − 2 dyn µ ) 2

dyn E = vc ρ c (2)
(1 − dyn µ )

Herein are: ρc , the density of (fresh) concrete; vc , velocity of compression wave; vs , velocity
of shear wave propagation. The velocities can easily be calculated from the measured times
of flight tc and ts with vc = dc / tc and vs = ds / ts , where dc and ds denotes the thickness of
specimen in the corresponding direction. Additional there exists a linear dependence
between the dynamic Young’s modulus and static modulus of elasticity.

stat E = k dyn E . (3)

The factor k depends on degree of hydration and the damping characteristics of concrete.
Using these equations it is possible to determine the initial degree of hydration by measuring
tc and ts (see [23]).

Eq. ( 1 ) and ( 2 ) can be used in to different ways. The first one is to measure the velocity of
both wave forms and derive the development dynamic Young’s modulus and dynamic
Poisson ratio. The experiences with this method are reported in chapter 5 of this paper. The
second one is to use only one wave form and make some assumption about the
development of the correspondent one. This method is used in chapter 4.

3 MATERIAL PROPERTIES OF YOUNG CONCRETE

The degree of hydration is determined on basis of adiabatic calorimetric. For the description
of the results of destructive tests of several material properties the models developed at the
 -5-

iBMB and the models used in Scandinavia are briefly presented. More to these can be found
in Gutsch (1998) and Westman (1999). The heat of hydration is expressed by

Q (t e ) = Q pot ⋅ α (t e ) (4)

with α(te) defined by

c1
  t 
ln α (t e ) = − ln1 + e   .

(5)
  tk 
There-in are

te effective age (see [30]),

Qpot maximum heat of hydration (see [30]),

Q (te) heat liberated until te

tk concrete specific parameter [h],

c1 concrete specific parameter [-].

The axial tensile strength fct, compressive strength fc and the Young’s modulus Ect can be
expressed by a degree of hydration dependent form function.

ni  12 for E ct
 α − α0  
X i (α ) = X i1   ; X i (α ) ∈ { E ct , f ct , f c }; ni = 1 for f ct . (6)
 1 − α0  3
2 for fc
Here-in are: Xi1, hypothestical end value of property at α=1; ni, property specific exponent.
Eq. ( 6 ) shows the values ni used by iBMB. The values Xi1 and ni are determined by
regression of test results. The Scandinavian models expresses the heat of hydration, degree
of hydration, strength and stiffness by a similar expression.

 t  bi 
Yi (t e ) = Yi1 exp  −  e  (7)
  ai  

The parameters Yi1, ai and bi must be determined on basis of tests.
 -6-

4 DETERMINATION OF INITIAL DEGREE OF HYDRATION WITH THE

COMPRESSIONAL WAVE METHOD (CWM)

4.1 Introduction

The use of only one wave-form, esp. compressional (longitudinal) waves, to determine the
elastic moduli of concrete structures is an established method. For the description of the
hardening of concrete -as described in [1] - several assumptions have to be made. With
ultrasonic measurements the attempt is to be undertaken to point out the connection
between the degree of hydration and the dynamic as well as static modulus of elasticity.
Concrete compositions out of practice are investigated. The object of the investigation was
to prove the influence of substantial variables on the development of stiffness and hardening
of concrete.

It has to be mentioned that this research was supported by VDZ e.V. in 1996. A presentation
of this work is reported earlier in [31].

4.2 Compressional wave method

The tests for the determination of α0i by the analysis of hydration-dependent processes of
the dynamic modulus of elasticity run in two steps. On the one hand the non-linear involution
coefficients are to be determined for the calculation of the degree of hydration with
specification of a temperature history. This test takes place under semi-adiabatic curing
conditions. On the other hand the pulse velocity and the temperature development of the
sample as a function of the concrete age are measured. The continuous temperature entry
by means of thermocouple serves the transformation of the real concrete age into the
effective concrete age. Afterward the degree of hydration can be calculated afterwards
according to equation ( 5 ), so that the process of the ultrasonic rate is present as function of
the degree of hydration. And last but not least, the development of dyn E acc. to equation ( 2
). Equation ( 2 ) shows, that the use of a time independent and constant Poisson ratio is not
advantageous. In the context of the investigations represented here the development of the
Poisson ratio from the mechanical measurements of Paulini et.al 1994 was determined (viz.
Figure 2).
 -7-

Figure 2 : Development of
Poisson´s ratio vs. degree of
hydration acc. to Paulini etal.
1994

4.2.1 Application of the sound penetration technique and testing

Concrete is tested by sound penetration with separate transmitting and receiving heads (see
Figure 3). During the sound penetration the development of stiffness is recognized by a
decreased transient time tp (pulse transit time method) and by a smaller receiving amplitude
A (intensity method). In the context of this work the executed investigations had the target of
the testing this procedure. The test system Sonograf 1000 and test heads of the type S9204
are used [12]. The low-frequency absorbed transducers S9204 have a diameter of 22 mm
and a pulse width up to 60 micro seconds with an operating frequency from 80 kHz to
180 kHz. The transient characteristic of these heads is represented in Figure 4. Due to their
diameter the transducers have no directional characteristic. With the small dimensions of the
transducers easy coupling at the concrete with Vaseline and the following adhesion without
mounting plate are favourable due to cohesiveness. The used measuring instrument
Sonograf 1000 was conceived particularly for ultrasonic testing in the frequency range from
10 kHz to 1 MHz at strongly sound-weakening materials like cement based. A spectrum
analyser divided besides each sound impulse into fifteen frequency ranges width of 20 kHz.
The sketch of a Figure 5 represents the measuring system as well as the measuring
instrument schematically. Sound-weakening materials affect like a filter the frequency
spectrum. Hillger 1983 observes a filter shift during the confirmation phase. High frequencies
are filtered briefly after mixing, high intensities are in the low-frequency area to be found. A
maximum of the higher frequencies occurs seven hours after mixing. The frequency
characteristic does not change with progressive hydration any longer considerably.
 -8-

Figure 3 : Ultrasonic
testing of mineral
building materials
according to the
sound penetration
technique (Hillger &
Neisecke 1993).

Figure 4 :
Transmission
characteristics of
test heads S9204
acc. To Hillger
(Sonograf).

The tests were performed for 11 concrete compositions. Table 1 shows their composition. In
view of the specimen size, the maximum grain diameter of aggregate had to be limited to
16 mm. Specimen size was 10/20/30 cm. The temperatures of fresh concrete is listed in
Table 2. The specimen size was selected in such a way that the dimensions are maximised
and so that the concrete could be penetrate as soon as possible. A wooden formwork
insulated with 40 mm PS-foam plates was used. In order to ensure coupling of heads to the
fresh and hardening concrete throughout test, a hold-down attachment was developed. The
test heads were positioned in holes of the thermal insulation and were coupled to concrete
with vaseline. For the optimisation of coupling at the confirming concrete a hold-down
attachment was compiled, which worked against chemical shrinking. In a copper tube a
stamp is led and a spring presses the probe to the concrete with a strength of 50 N. The
spring travel is from the outside adjustable by an appropriate screw, so that the distance
between the test heads can be adjusted exactly on 1 mm. After placing the transducers
could be positioned justified in the isolation. The set-up performed every 10 minutes a
measurement. At this moment the sender was taken in operation. This measure was seized,
in order to avoid, that the concrete absorbs sound energy permanently. The impulse was
 -9-

transmitted until the spectrum analyser determined the run time of the signals and the
intensities in all 15 filters.

Figure 5 : Sketch of
used test set up in
chapter 4.

Table 1: Concrete compositions used in CWM

concrete composition C1 C2 C3 C4
cement CEM I 32.5 R CEM I 42.5 R CEM III/A 32.5 CEM I 32.5 R
Plant A Plant A Plant A Plant B
w/c – ratio 0.65 0.50 0.50 0.37
cement content [kg/m³] 270 330 330 260
fly ash content [kg/m³] 60 - - 60
silica suspension - - - 120
additives [kg/m³] 2.67 - - 15.00
workability (spread table) KR KR KR KP

concrete composition C5 C6 C7 C8
cement CEM I 32.5 R CEM I 32.5 R CEM I 32.5 R CEM I 32.5 R
Plant B Plant B Plant B Plant B
w/c - ratio 0.55 0.50 0.55 0.60
cement content [kg/m³] 260 330 330 330
fly ash content [kg/m³] 120 - - -
silica suspension - - - -
additives [kg/m³] 15.00 33.30 33.30 33.30
workability (spread table) KR KR KR KR

concrete composition C9 C10 C11

cement CEM I 42.5 R CEM III/A CEM III/B 32.5 - NW / HS / NA
Plant B 32.5 Plant B
w/c - ratio 0.60 0.60 0.60
cement content [kg/m³] 330 330 330
fly ash content [kg/m³] - - -
silica suspension - - -
 - 10 -

additives [kg/m³] 33.30 33.30 33.30

workability (spread table) KR KR KR
CEM I, OPC; CEM III A; moderate heat GBFSC, slag content 56 M.-%, CEM III/B low heat
GBFSC, slag content 68 M.-%

Table 2: Temperatures of fresh concrete

sample S1,1 S1,2 S1,3 S2,1 S2,1 S3,1 S3,2 S4,1
composition C1 C1 C1 C2 C2 C3 C3 C4
temperature of fresh concrete 18.3 21.0 18.8 18.5 19.0 18.9 16.3 15.6
[°C]

sample S5,1 S6,1 S6,2 S6,3 S7,1 S7,2 S8,1 S8,2

composition C5 C6 C6 C6 C7 C7 C8 C8
temperature of fresh concrete 24.8 22.0 22.5 21.2 20.8 22.2 21.5 22.0
[°C]

sample S9,1 S9,2 S10,1 S10,2 S11,1 S11,2

composition C9 C9 C10 C10 C11 C11
temperature of fresh concrete 23.2 22.4 22.4 20.5 20.8 21.0
[°C]

4.3 Test Results of CWM

4.3.1 Velocities of compressional waves

The pulse velocities were measured as a function of the real concrete age (time after
mixing). Figure 6 presents the results. The graphs show the known phenomenon that
impulses are transferred only after a certain age. This is surely to explain with the very high
absorption of the material in the liquid - mash state. Hillger indicates the speed of sound as
300 m/s. Such a value is nearly the speed of sound in water (340 m/s).

The representations of the transferred sound rates show all an almost identical process over
the duration of test: In one period of 2 to 4 hours after mixing the concrete no modifications
of the mechanical characteristics are provable. The solid skeleton, which would enable a
faster sound propagation, does not appear yet to be structured. After the beginning of
solidification, which becomes apparent by starting the pulse transmission, a steep rise of the
measuring curves is to be determined. Sound rates up to 3500 m/s are achieved in 3 - 4
hours. This point in time corresponds for instance with the hydration phase, in which the first
CSH phases of the still young hardened cement paste bridge the water-filled gaps between
 - 11 -

the cement grains. After Paulini [27] the transverse contraction ratio of approx. 0.5
(transverse contraction ratio of a liquid) falls on approx. 0.25.

After approximately 8 - 9 hours the sound parameters change only slightly. With the end of
the solidification a solid skeleton is available for the sound transmission, whose
microstructure (the pore space) does not have considerable influence more on the sound
away line.
Ultraschallgeschwindigkeit [m/s]

5000
S1,1 Figure 6: Pulse velocities
S1,2
4500 S1,3
vs. real age
S2,1
4000 S2,2
S3,1
S3,2
3500 S4,1
S5,1
3000 S6,1
S6,2
S6,3
2500 S7,1
S7,2
2000 S8,1
S8,2
S9,1
1500 S9,2
S10,1
1000 S10,2
S11,1
S11,2
500

0
2 4 6 8 10 20 40 60 80 100
USSH | 12.2.2001 Betonalter [h]

4.4 Development of dyn E acc. to CWM

In similar way as pulse velocity on time behave also the dynamic modulus of elasticity The
degree of hydration dependent development of the elastic module is represented in Figure 7.
The calculation of the dynamic elastic module took place on the assumption of a constant
concrete density. This acceptance is not accurate surely. As well known the concrete
shrinks, so that the gross density does not represent a constant size directly over the
duration of the hydration. Neisecke 1974 reports on the pursuit of the density development
of the concrete during the solidification and confirmation. Up to a degree of hydration in the
magnitude of 0.003 to 0.08 dyn E does not take significant values. Afterwards dyn E steeply
rises, and reaches values up to 40 GPa. The curves exhibit three regions: very steep rising
of the modulus of elasticity to 25 GPa with immediate flattening, in order to rise only slightly
to 30 GPa, rising with somewhat a slighter slope, but the final value of dyn E achieves
approx. 40 GPa and fewer steep rising relatively high α0 and relatively small final values of
25 GPa.
 - 12 -

Figure 7: Dynamic
Edyn [N/mm2] 45000 Young’s modulus vs.
S1,1

40000
S1,2 degree of hydration (all
S1,3
S2,1 concrete compositions) .
S2,2
35000 S3,1
S3,2
30000 S4,1
S5,1
S6,1
25000 S6,2
S6,3
S7,1
20000 S7,2
S8,1
S8,2
15000
S9,1
S9,2
10000 S10,1
S10,2
S11,1
5000 S11,2

0
0 0.2 0.4 0.6 0.8 1
USEA | 12.2.2001 Hydratationsgrad α [-]

4.4.1 Influence of cement type

The cement type influences on the hardening process of concrete. For this reason test
series executed with differentiated cement places and identical other concrete-technological
parameters. The results of these measurements are presented in Figure 8. The water –
cement ratio was kept constant with 0.6. Basically, the OPC indicates an earlier solid
behavior. This is understandable because of the composition of the cement. The smaller
granulated blast-furnace slag proportion in the OPC compared to the GBFSC leads to a
more rapid hydration, which becomes apparent not only temporally, but also into senses of
earlier degrees of hydration. The GBFS cement leads accordingly to concrete compositions,
which indicates with same α as a OPC - concrete a smaller strength and with a smaller
modulus of elasticity. The general tendency is preserved.

The rises of the curves do not run however parallel That the dyn E with OPC manufactured
concrete compositions more steeply rises, as the correspondent value of a GBFS concrete is
to be marked. This slowing down results from the delayed hydration of the granulated blast-
furnace slag. The latter takes place only after suggestion via lime, which is set free with the
hydration of the Portland cement clinkers. Tests are performed with OPC and GBFSC with a
constant cement content (s. Table 1) and a constant water-cement ratio of 0.6 . OPC-
concrete show an early solidification than GBFS - concrete.

Further it is to be stressed that a curvature of the measuring curves of GBFS - concrete is

related to the calorimetric influence of the degree of hydration. GBFS cements indicate a
smaller proportion of C3A. The specific heat of hydration of C3A is however very high,
whereby and thus at the strength contribute latter no significant proportion of the stiffness.
 - 13 -

From this the almost horizontal process of the dynamic elastic module explains itself with
degree of hydration around approx. 0.15. The unusually strong distinction between the
CEM III/A and CEM III/B cements is to be explained with the same argumentation. The
cement analysis of the manufacturer indicates a granulated blast-furnace slag content of
56 m% for the CEM III/A, while the analysis of the CEM III/B shows a proportion of 68 m% of
granulated blast-furnace slag. The normal compressive strength after two days amounts to
interestingly enough 15.3 MPa or 8.8 MPa. These values clarify the delaying influence of the
additional 12 m% quantities granulated blast-furnace slag.

Figure 8 :Influence of
cement type on dynamic
Young’s modulus vers.
degree of hydration.

4.4.2 Influence of water - cement ratio

Figure 9 shows the dependence of dyn E on degree of hydration for OPC- and GBFS -
concrete with a constant binder content of 330 kg/m3 and variable wcr. The initial steep rise
of dyn E is for OPC more pronounced than for GBFCS; it increases with decreasing w/c
ratio.

The study of the influence of w/c ratio is presented in Figure 9. The figure shows the
hydration-dependent process of the dynamic modulus for the examined concrete
compositions, which were manufactured with a OPC. Basically it is to be stated that dyn E
rises more steeply with falling w/z value. Furthermore the stress transmission is registered
later with rising w/c - values. Obviously with falling w/c - ratios structure bridges rather
established between the cement grains, which increase the stiffness of the respective
concrete. These processes are interpretable with the past realisations over the influence of
the w/c value on the development of the void structure With a w/c - ratio of approx. 0.4
(exception in this case is the mixture C4, which was mixed with addition of a silica
 - 14 -

suspension) the remaining water forms capillary pores in the hardened cement paste. The
attenuation of the structure caused by these pores and the sound damping lead to smaller
stiffness developments.

Figure 9 :Influence of
45000 water – cement ratio
Edyn [N/mm2]

on dynamic Young’s
40000
modulus vs. degree of
35000 hydration.
30000

25000

20000

15000

10000 270 kg CEM I 32.5 R, w/z=0.65

330 kg CEM III/A 32.5, w/z=0.50
5000 330 kg CEM I 32.5 R, w/z=0.50
330 kg CEM I 32.5 R, w/z=0.55
0
0 0.2 0.4 0.6 0.8
α0i
Hydratationsgrad α [-]

The temperature of fresh concrete Tc0 also plays a role regarding the process of the stiffness
development. It is known that high values of Tc0 are connected with a lower compressive
strenght at higher age. This statement also hold for the modulus of elasticity. With the
increase of Tc0, the initial degree of hydration α0i decreases.

4.5 Determination of initial degree of hydration

The extrapolation of the hydration-dependent process of the dynamic modulus of elasticity

enables a quantifying of α0i when disappearing the stiffness. Table 3 represents the results
of these determinations.

These values are clearly lower values for α0, determined so far. In Laube (1994) α0 was
calculated from the extrapolation of the straight lines for fct vs. α. Since serious axial tensile
tests succeed only with a degree of hydration starting from approx. 0.40 and the scatter of
this test results is large, an regression analysis is strongly faulted. This error can be
quantified only with difficulty. A further problem in the determination of α0 is to be found in
the fact, that the determination of the degree of hydration and the handling of the samples
(protection from drying shrinkage, exact temperature storage etc..) represent identifiable
 - 15 -

sources of error. In the next section one points out, how these sources of error can be
considered.

Table 3: Determinate values of initial degree of hydration acc. to CWM

sample S1,1 S1,2 S1,3 S2,1 S2,1 S3,1 S3,2 S4,1 S5,1 S6,1 S6,2
composition C1 C1 C1 C2 C2 C3 C3 C4 C5 C6 C6
α0i [-] 0.079 0.071 0.071 0.031 0.031 0.040 0.020 0.029 0.009 0.020 0.034

sample S6,3 S7,1 S7,2 S8,1 S8,2 S9,1 S9,2 S10,1 S10,2 S11,1 S11,2
composition C6 C7 C7 C8 C8 C9 C9 C10 C10 C11 C11
α0i [-] 0.034 0.020 0.029 0.025 0.030 0.030 0.040 0.030 0.040 0.050 0.050

5 DETERMINATION OF INITIAL DEGREE OF HYDRATION WITH COMBINED

SHEAR AND COMPRESSIONAL WAVE METHOD (SCWM)

Figure 10: Test set-

up of SCWM

5.1 Introduction

The CW Method of chapter 4 can be decisively improved, if the following points are
considered: Parallel to the rate of the longitudinal waves the rate of speed of the shear
waves is measured. The rates are determined under adiabatic conditions. Thus, a direct
 - 16 -

relationship with the degree of hydration is established and the model uncertainties of Eq. ( 5
) are minimised. Conventional mechanical tests for the assessment of the static modulus of
elasticity on specimen cured under almost adiabatic conditions (destructive testing in axial
tension).

5.2 Test set-up

Figure 10 shows the test set-up. It consists of the adiabatic calorimeter, controlling unit,
ultrasonic measurement device and controlling unit for the external water bath. All controlling
units are integrated in a standard PC. A proprietary software steers all controlling tasks in
this apparatus. The water bath is used to store the concrete specimens for destructive
testing under adiabatic curing conditions.

5.3 Adiabatic calorimeter

Immediately after mixing of the concrete the measurements start. The steel bucket
(Vol. = 11.6 dm3, ∅ = 205 mm) contains a tube to protect the temperature sensor (hand
selected Pt 100) in the centre of specimen. This tube is filled with 3 ml silicon oil. The sample
is insulated by an air filled chamber surrounded by a metal jacket. Multiple layers of
insulation are inserted between the jacket and external shell. A ventilator at the top of the
jacket is installed to guarantee a good circulation of air in the chamber. The air temperature
is measured by temperature sensor (Pt 100, placed directly in the air stream of the
ventilator) and heated by a silicon-oil-filled heat exchanger. The temperature difference
between specimen and air (∆T < 0.01 K, thermal loss < 0.02 K/h) is controlled by PC. The
weight of specimen is measured before and after testing.

5.4 Ultrasonic measurement

The ultrasonic transducers are integrated in the steel bucket (viz. Figure 10). The
transducers are protected by a plastic tube. A rubber seal ring fixes the sensors securely.
This construction guarantees that no moisture is lost during testing. The transducers are
coupled directly. As receiver and transponder a commercial transient time measurement
system is used. Two types of commercial ultrasonic shear wave and compressional wave
transducers are used. They are able to work with a frequency range of 50 kHz or 100 kHz.
 - 17 -

The dimensions dc and ds were evaluated for each single test. Pitches from 90 mm up to
110 mm are measured.

Usually the transient time of compressional wave is defined as the first significant change of
measured signal. This procedure is acceptable for non-mineral and well-hydrated building
materials. But for young concrete it may cause some problems. The main reason for this
phenomenon is the excellent damping property of fresh concrete. The results are flat and
small impulses, so the transient time is difficult to identify. The main aggravating
circumstances are the disadvantageous signal - noise ratio and the time dependent
transducer - specimen system. For this measurements all evaluations were automatically
performed by a personal computer. Every 300-3600 seconds ten evaluations are made and
the average value is stored. To ensure unchangingly measures some declarations have to
be made. The first is, if the maximum amplitude is detected, the incoming signal is amplified
to 100 % of its intensity. The second is, the wanted transient time is defined as point of time,
were the first amplitude reaches 5 % of the maximum one. To control this procedure, all
measured a-scans are evaluated additionally by hand (viz. Figure 11). The identification on-
set time of shear pulses complicated. In this case, the transient time can not be evaluated as
the first significant change of signal from zero-line. The reason is that the shear wave
transducer can not generate a natural shear wave. But if the ratio of shear wave signal to
compressional wave ratio is big enough, it is possible to identify ts ([29]). In Figure 12 the
procedure is shown. For the calculation of transient time the point of time is depicted, who
shows the first significant change of the compensated ultra-sonic-signal. All signals are
evaluated in the same way as described in the paragraph above.
 - 18 -

60
amplitude
40 amplified
to 100%
20
A [%]

-20

-40
evaluated transient time
-60
40 80 120 160 200
tc [µs]

Figure 11: Identification of transient time of Figure 12: Identification of transient time of s-
c-wave wave

5.5 Testing of axial tensile strength and static modulus of elasticity

The specimen with the dimensioned ratio of h / d = 2 is shown in Figure 5. The tests were
performed with a constant strain rate of 0.02 ‰ h-1. In order to relate the mechanical
properties of concrete at very early age to the degree of hydration, tests started as soon as
possible. The Young's modulus is defined as secant modulus between 5 and 55 % of tensile
strength.

5.6 Concrete compositions

Table 4 shows the investigated concrete compositions. Four types of parameter variations
were made. The task is to evaluate the influences of cement type, cement content, water -
cement-ratio and workability at the end of dormant phase. As cements the typical German
cement types OPC (CEM I 32.5 R), GBFS-PC cement class B (CEM III 32.5 NWHSNA) and
rapid hardening PC (CEM I 42.5 R, CEM I 52.5 R) are considered. These cement types are
selected regarding their maximum heat of hydration. Additional two HPC - compositions are
tested (CO21/R1-R2, CO11/I1-I5). All these compositions were investigated extensively.
Their major thermodynamic properties and the correspondent mechanical properties were
tested at iBMB. The comprehensive results are reported formerly (viz. [34]).

Table 4: Concrete compositions

concrete No. cement type cement content w/c Tc0 remarks
3
[kg/m ] [-] [°C]
CO1/P10 OPC 270 0.65 9.5 60 kg/m3 FA*1
 - 19 -

3
CO1/P11 OPC 270 0.65 15.3 60 kg/m FA*1
CO1/P12 OPC 270 0.65 20.1 60 kg/m3 FA*1
CO9/P13 GBFS-PC 325 0.60 20.0
CO2/P14 GBFS-PC 390 0.47 20.0
CO4/P15 OPC 390 0.40 21.0
CO6/P16 OPC 390 0.51 20.0
CO7/P17 OPC 300 0.51 20.0
CO8/P18 GBFS-PC 390 0.51 20.0
3
CO21/R1-R2 HPC 370 0.37 22.2 20.0 [kg/m ] SF*2 German Mix
3
CO11/I1-I5 HPC 370 0.40 20.4 18.4 [kg/m ] SF*2 Norwegian Mix
*1 *2
FA = Fly ash SF = silica fume

Figure 13: Specimen for the tensile

strength test

5.7 Models and Test results

5.7.1 Test results

From this comprehensive test work only few representative results can be presented. For
this, the results for the high-performance concrete compositions CO21/R1-R2 and CO11/I1-
I5 (viz. Table 1) are chosen. The latter is commonly used in Norway for structures in very
aggressive environment.

5.7.2 HPC

As mentioned above in the Brite Euram research program "Improved Production of

Advanced Concrete Structures" the concrete composition CO11/I1-I5 was tested
extensively. All ingredients of this mix came from Norway, so this is called "Norwegian HPC".
 - 20 -

This recipe is a typical HPC and used in the Øeresund Project. Some results of these tests
are reported. Additional to that a "German HPC" was affiliated to this test program.

5.7.3 Material properties at early ages of HPC

0.6 5
Norwegian HPC Norwegian HPC
German HPC German HPC
0.5
4

0.4
3

fct [MPa]
meas α [-]

0.3

2
0.2

1
0.1

0 0
0 4 8 12 16 20 10 50 100 500 1000
te [h] te [h]
Figure 14: Development of degree of Figure 15: Development of axial tensile
hydration strength

The next four figures (Figure 14 - Figure 17) shows the development of the investigated
material properties vs. equivalent time at early ages. The calculated form parameters are
documented in Table 5. The heat development of the German Mix starts earlier. The reason
is the higher amount of C3A. The development of the mechanical properties are nearly the
same under adiabatic curing conditions. The maximum compressive strength for both mixes
reaches values from 68 to 75 MPa. For the maximum values of axial tensile strength values
between 3.1 MPa and 4.1 MPa are measured and last but not least for the statically Young's
modulus under tension values has been established between 28 GPa and 34 GPa after
672 h. The most interesting point of these investigations are the mechanical properties at
very early ages. The Figures also shows, that all properties of the Norwegian Mix starts
nearly at the same age. The initial setting time for te0i is about 9 h. From Figure 14 one can
derive a degree of hydration of 0.1.
 - 21 -

35 80

70
30
60
25
50
Ect [GPa]

fc [MPa]
20 40

30
15
20
10
Norwegian HPC 10 Norwegian HPC
German HPC German HPC
5 0
10 50 100 500 1000 10 50 100 500 1000
te [h] te [h]
Figure 16: Development of Young’s modulus Figure 17: Development of compressive
vs. equivalent age strength vs. equivalent age

Table 5: Parameters of shape functions for Norwegian Mix and German Mix
“Norwegian Mix” “German Mix”
3
α iBMB Qpot kJ/m 174864 192831
tk h 9.141 6.87
c1 - -0.853 -0.816
3
Skan Qpot kJ/m 174864 192831
a h 22.21 15.80
b - 0.456 0.352
fct iBMB fct1 MPa 5.71 4.49
α0 - 0.17 0.12
Skan fct1 MPa 3.99 3.27
a h 36.32 14.33
b - 0.832 0.978
Ect iBMB Ect1 GPa 36.63 34.22
α0 - 0.17 0.12
Skan Ect1 GPa 33.85 28.49
a h 16.11 0.427
b - 0.603 0.464
fcc iBMB fc1 MPa 124 117
α0 - 0.17 0.12
Skan. fc1 MPa 88.97 -
a h 94.16 -
b - 0.444 -
 - 22 -

5.7.4 Application of NDT on HPC

In Figure 18 and Figure 19 the evaluation of dynamic Poisson’s ratio and dynamic Young’s
modulus vs. equivalent age for very early ages is presented. It can be noticed, that the end
of the dormant phase for the Norwegian HPC is achieved after 8 hours and in 5 hours for the
German HPC. This result fit very well to the observations of the tests presented in section
5.7.3. From this results the proposition can be derived that the development of the
mechanical properties of the two mixes starts earlier then predicted from destructive testing.
In Figure 20 and Figure 21 a comparison of the test results for the Norwegian Mix are
shown. Both pictures underline the fact, that the dynamic E-modulus is larger than the static
E-modulus. The ratio dyn E/stat E varies from 1.7 to 1.5 for equivalent ages greater than
50h. The development of both properties start nearly at the same time (6.5h for dyn E and
11h for stat E). Figure 13 give another point of view. The dyn E - α plot shows, that the
mechanical properties starts earlier than predicted from destructive testing. The reason for
that observation is, that the degree of hydration is much more sensitive on outwardly
influences than equivalent age. To model the dynamic measurements also the equations
from section 4.1.2 and 4.2.1 are used. These models predict the test results well, as Figure
12 and Figure 13 shows. The calculated form Parameters are given in the figure. In the next
step the gap in the development of mechanical properties between α0i and α0 has to be
closed.

0.5 45

36
0.4
dyn E [GPa]

27
dyn µ [-]

0.3

18

0.2
9
Norwegian HPC Norwegian HPC
German HPC German HPC
0.1 0
5 10 50 100 5 10 50 100
te [h] te [h]
Figure 18: Development of Poisson’s ratio Figure 19: Development of dynamic Young’s
equivalent age modulus vs. equivalent age
 - 23 -

50 50 50 50
Norwegian HPC Norwegian HPC

40 40 40 40
stat Ect [GPa]

dyn Ect [GPa]

dyn Ect [GPa]

30 30 30 30

Ect [GPa]
20 20 20 20

meas stat
10 10 10 meas stat 10
meas dyn
meas dyn
mod stat mod stat
mod dyn mod dyn
0 0 0 0
0 50 100 150 200 0 0.2 0.4 0.6 0.8 1
te [h] α [-]
Figure 20: Comparison of dyn. Young’s Figure 21: Comparison of dyn. Young’s
modulus and stat. mod. of elasticity vs. modulus and stat. Mod. of elasticity vs.
equivalent age degree of hydration

5.7.5 Improvement of iBMB – model

Figure 22: Improvement of

model for the prediction of
mechanical properties of
early age concrete

In order to obtain a simple way to improve Eq. ( 6 ), the following procedure is suggested.
Figure 22 shows the procedure. For pre-determined values of α0i and α0 (Figure 21).
 - 24 -

Additional assume a value γ ≥ 1. Evaluate Eq. ( 6 ) with the value α =γ ⋅α0. Then connect this
point with a straight line to α0i. The improved model can expressed by:

  α − α 0i 
 X i (γ ⋅ α 0 ) ⋅   for α 0i ≤ α ≤ γ ⋅ α 0
~   γ ⋅ α 0 − α 0i 
X i (α ) =  ni .
 α −α0  (8)
X i1 ⋅   for α 0i < α ≤ 1
  1−α0 


5.7.6 Improvement of Scandinavian model

If the dormant phase te0i is known, Eq. ( 7 ) can revisited easily. Substitute the argument te
with te - te0i and the detention of the development can predicted precisely.

Table 6: Shape parameters of improved models for Norwegian and German HPC
Parameters of iBMB Parameters of
model Eq. ( 8 ) Scandinavian
model Eq. ( 7 )
Norwegian Ect1 GPa 36.44 Ect1 GPa 26.26
HPC
ni - 0.5 a h 11.94
α0 - 0.176 b - 1.539
α0i - 0.05 te0i h 11
γ - 1.4 - - -
German HPC Ect1 GPa 34.22 Ect1 GPa 28.59
ni - 0.5 a h 0.427
α0 - 0.12 b - 0.464
α0i - 0.03 te0i h 6
γ - 1.3 - - -

5.8 Parameter studies regarding end of dormant phase

The methods described in section 5.7.2 are used to investigate the OPC- and GBFSC mixes
defined in Table 4. As mentioned above especially the influence of cement-type, cement
content, w/c-ratio and temperature of fresh concrete on α0i are studied.

Lower initial temperatures have the same effect on the development of hydration as the use
of retarder. As Table 7 shows low values of Tc0 causes higher values of α0i.

The influence of w/c-ratio on the setting characteristics is clearly defined by the amount of
 - 25 -

water used in the concrete mix. A reduction in the water/cement-ratio gives a definite
increase in the setting and stiffening. This can easily noticed in the development of pulse
velocities ([39]). In spite of that, a significant influence of w/c-ratio on α0i can’t established in
this investigations. The reason is only ratio’s greater than 0.37 have been used. The
development of heat of hydration of concrete depends on the chemical composition of
cement, the fineness of cement and amount of cement. So cement-type and cement content
have a direct influence on the end of dormant phase. An increase of cement content causes
an enhancement of α0i. For the influence of fineness of cement the correlation is similar.

Table 7: Influence of concrete composition on end of dormant phase

parameter unit OPC GBFS-PC HPC
Tc0 °C 9 - 21 20 19 - 21
3
C kg/m 300 - 400 300 - 400 370
w/c - 0.4 - 0.65 0.47 - 0.6 0.37 - 0.4
α0i - 0.04 - 0.1 0.08 0.03 - 0.05
γ - 1.5 1.6 1.4
α0 - 0.15 - 0.19 0.36 0.14 - 0.17
te0i h 10 - 16 18 5-9

6 CONCLUSIONS

A realistic prediction of mechanical properties plays important role in the calculation of

stresses due to heat of hydration. The assessment of the development of mechanical
properties like modulus of elasticity, axial tensile strength and compressive strength includes
many uncertainties. One of the fact, that the end of dormant phase cannot be correctly
assessed by ordinary destructive tests correctly.

With a non-destructive test method this problem can be solved. By measuring ultrasonic
pulse velocities of compression and shear waves it is possible to determine the end of the
dormant for arbitrary concrete compositions. Best results can be attained when these
measurement are performed under adiabatic curing conditions.

Investigations presented in this paper shows, that the end of the dormant phase depends on
the same parameters as degree of hydration. Additional destructive test were performed at
early ages and very early ages are needed to transform dyn E into stat E. The NOT results
show good correlation with the non-destructive tests. Using on basis of the results of both
tests, it is possible to improve existing models for the prediction of mechanical properties of
early age concrete.
 - 26 -

7 REFERENCES

[1] Bunke, N. (ed.); 1991. Testing of Concrete - Recommendations and notes as

supplement to DIN 1048 (in German). Deutscher Ausschuß für Stahlbetonbau, Heft
422, Beuth Verlag, Berlin 1991.

[2] Byfors, J.; 1982. Pulse velocity measurements for prediction of the compressive
strength at early ages. RILEM Paris, 6. – 8. April 1982.

[3] Eisenmann, K.; 1951. Untersuchung der Rißbildung des Betons mit Ultraschall. Beton-
und Stahlbetonbau 46. S. 19.

[4] Eisenmann, K.; Odewald, G.; 1959. Untersuchung der elastischen Eigenschaftenvon
Beton mittels Ultraschall. Beton- und Stahlbetonbau 54. S. 156-158.

[5] Eisenmann, K.; Steinkamp, G.; 1954. Messung der Tiefen und Weiten von Betonrissen
mittels Ultraschall. Beton- und Stahlbetonbau 49. S. 36-38.

[6] Elvery, R. H.; Ibrahim, L. A. M.; 1976. Ultrasonic assessment of concrete strength at
early ages. Magazine of concrete research, vol. 28, nr. 97, December 1976.

[7] Emborg, M .; 1998. Development of mechanical behavior at early ages. Published in

RILEM 1998. Prevention of thermal cracking in concrete at early ages. RILEM Report
15. R.Springenschmid ed. FN SPON, London

[8] Gaunaurd, G. C.; Überall, H.; 1981. Resonance theory of bubbly liquids . Journal of the
Acoustics Society of America 69.

[9] Gutsch, A.-W.; 1999. Stoffeigenschaften jungen Betons - Versuche und Modelle. Doct.
Th. TU Braunschweig.

[10] Härig, S.; Teodoru, G.; 1988. Kritische Gesamtauswertung der auf internationaler
Ebene gewonnenen Erfahrungen und Erkenntnisse über die zerstörungsfreien
Prüfmethoden von Beton (Ultraschall). DFG-Abschlußbericht zu den
Forschungsvorhaben HA 1414/1-1 und 1-2.

[11] Hariri, K.; Krauß. M.; Onken P.; 1997. Ultraschallmessungen an jungem Beton,
erschienen in: Festschrift zum 65. Geburtstag von Prof. Dr.-Ing. F. S. Rostásy:
Baustoffe in Praxis, Lehre und Forschung, Institut für Baustoffe, Massivbau und
Brandschutz, Amtliche Materialprüfanstalt für das Bauwesen, Heft 128, Braunschweig
1997. S. 73-75.

[12] Hillger, W.; Sonograph. Benutzerhandbuch des Sonograph 1000. Modulares

Niederfrequenz Ultraschall Prüfgerät. Ingenieurbüro Dr.- Ing. W. Hillger, Braunschweig.

[13] Hillger, W.; 1983; Verbesserungen und Erweiterungen von Ultraschallprüfverfahren zur
zerstörungsfreien Fehlstellen- und Qualitätskontrolle von Betonbauteilen, Institut für
Baustoffe, Massivbau und Brandschutz der TU Braunschweig, Heft 60, Braunschweig
1983.

[14] Hillger, W.; 1987. Ultraschallimpulstechnik zur zerstörungsfreien Prüfung von Beton,
Institut für Strukturmechanik der DFVLR, Braunschweig, erschienen in: DFVLR-
Nachrichten, Nr. 51, S. 34-36, Juni 1987.

[15] Hillger, W.; 1993. Impuls-Echo-Technik an mineralischen Baustoffen , Braunschweig,

erschienen in: Materialprüfung, Nr. 35 (1993) 5, S. 133-136, Carl Hanser Verlag,
München 1993.
 - 27 -

[16] Hillger, W.; Neisecke, J.; 1993. Qualitätssicherung mineralischer Baustoffe durch neue
Ultraschall-Impuls-Echo-Technik, Quality control of Mineral Building Materials by means
of the Novel Ultrasonic Pulse-Echo Technique, Braunschweig, erschienen in: Betonwerk
+ Fertigteil-Technik, Heft 6/1993, S. 82-87.

[17] Hillger, W.; 1995. Bildgebende Ultraschall-Prüftechnik für Sandwich-Bauteile,

Braunschweig, erschienen in: Materialprüfung, Nr. 37 (1995) 11-12, S. 469-472, Carl
Hanser Verlag, München 1995.

[18] Keating, J.; Hannant, D. J. ; Hibbert, A. P.;1989. Comparison of shear modulus and
pulse velocity techniques to measure the build-up of structure in fresh cement pastes
used in oil well cementing. Cement and Concrete Research. Vol. 19, pp. 554-566.

[19] Kordina, K.; Roy, V.; Waubke, N. V.; 1967. Ultraschallmessungen an bewehrten
Kiesbetonen. Materialprüfung 9. Nr. 3. S. 81-85.

[20] Krauß, M.; Rostásy, F. S.; 2000. Determination of initial degree of hydration by means
of ultrasonic-measurements. Institut für Baustoffe, Massivbau und Brandschutz der
Technischen Universität Braunschweig. Published in: International Workshop on Control
of Cracking in Early-Age Concrete, Tohoku University, Sendai Japan, August 2000, S.
39-48 (in preparation).

[21] Landwehrs, K.; Neisecke. J; Rostásy. F. S.; 1981. Entwicklung von Prüftechnologien für
die Ermittlung des Festigkeits-, Verformungs- und Temperaturdehnverhaltens von
Betonstahl, Spannstahl und Beton im Bereich tiefer Temperaturen von 0°C bis –180°C ,
Institut für Baustoffe, Massivbau und Brandschutz, Universität Braunschweig 1981.

[22] Morabito, P.; 2000. Round Robin Testing Program: Equipment, testing methods, test
results. Research report Brite-Euram BRPR-CT-97-0437. ENEL SPA.

[23] Nagy, A.; Thelandersson, S.; 1994. Material Charakterization of Young Concrete to
Predict Thermal Stresses. Thermal Cracking in Concrete at Early Ages. Proceedings of
the International RILEM Symposium held in Munich, October 10-12. E&FN Spon
London.

[24] Neisecke J.; Rehm. G.; Waubke, N.-V.; 1970. Ultraschall-Untersuchungsmethoden in

der Baupraxis – Literatursichtung - , Lehrstuhl und Institut für Baustoffkunde und
Stahlbetonbau, TU Braunschweig, Braunschweig 1970.

[25] Neisecke, J.; 1974. Ein dreiparametriges, komplexes Ultraschall - Prüfverfahren für die
zerstörungsfreie Materialprüfung im Bauwesen, Dissertation an der Fakultät für
Bauwesen der Technischen Hochschule zu Braunschweig, Braunschweig 1974.

[26] Onken, P.; 1997. Untersuchungsmethoden zur Bestimmung des Hydratationsverlaufs

von Beton im Rahmen von Eignungsprüfungen , Methods to Examine the Hydration
Evolution of Concrete During Suitability Tests. Published in: Festschrift zum 65.
Geburtstag von Prof. Dr.-Ing. F. S. Rostásy: Baustoffe in Praxis, Lehre und Forschung,
Institut für Baustoffe, Massivbau und Brandschutz, Amtliche Materialprüfanstalt für das
Bauwesen, Heft 128, Braunschweig 1997, S. 131-137.

[27] Paulini, P.; Gratl, N.; 1994. Stiffness Formation of Early Age Concrete. Published in
RILEM 1994. Thermal Cracking in Concrete at Early Ages. RILEM proceedings 25.
R. Springenschmid ed. FN SPON, London.

[28] Rehm, G.; Waubke, N.-V.; Neisecke. J.; 1973. Ultraschall-Untersuchungsmethoden in

der Baupraxis (Literatursichtung), Institut für Baustoffkunde und Stahlbetonbau der TU
Braunschweig, erschienen in: Berichte aus der Bauforschung, Heft 84, Verlag von
Wilhelm Ernst & Sohn, Berlin 1973.
 - 28 -

[29] Rehm, G.; Waubke, N.-V.; Neisecke. J.; 1975. Ultraschall-Impulstechnik für Fertigteile,
TU Braunschweig. Published in: Deutscher Ausschuß für Stahlbeton, Heft 243, Verlag
von Wilhelm Ernst & Sohn, Berlin 1975.

[30] RILEM; 1994. Thermal cracking in concrete at early ages. Proc. Int. RILEM symposium,
R Springenschmid ed. FN SPON, London.

[31] Rostásy, F. S.; Hariri. K.; Onken. P.; Krauß, M.; Sprenger, K.-H.; 1996.
Ultraschallmessungen zur Ermittlung der Steifigkeit von jungem Beton ,
Forschungsbericht gefördert durch den Verein Deutscher Zementwerke e.V.. Institut für
Baustoffe, Massivbau und Brandschutz, Amtliche Materialprüfanstalt für das Bauwesen,
September 1996.

[32] Rostásy, F. S.; Laube, M.; Onken, P.; 1993. Zur Kontrolle früher Temperaturrisse in
Betonbauteilen. Bauingenieur 68, 5.

[33] Rostásy, F. S.; Onken, P.; 1994. Konstituitives Stoffmodell für jungen Beton.
Forschungsbericht. TU Braunschweig.

[34] Rostásy, F. S.; Krauß, M.; 1999. Modeling degree of hydration on basis of adiabatic
heat release. Research report Brite-Euram BRPR-CT-97-0437. TU Braunschweig.

[35] Steinkamp, G.; 1953. Ultraschallimpulsgeräte für Beton. Dissertation. TH Braunschweig.

[36] v.Breugel, K. 1994. Numerical simulation of the effect of curing temperature on the
maximum strength of cement based materials. Published in RILEM 1994. Thermal
Cracking in Concrete at Early Ages. RILEM proceedings 25. R. Springenschmid ed. FN
SPON, London.

[37] Westman, G.; 1995. Concrete Creep and Thermal Stresses. Doct. Th. Luleå University
of Technology.

[38] Whitehurst, E. A.; 1951. Use of the Soniscope for Measuring setting time of concrete.
Proceedings of the American Society for Testing Materials 51. S. 1166-1175.
(Philadelphia)

[39] Winden, van, N. G. B.; 1990.Ultrasonic measurements for setting control. Published in
RILEM 1990. Testing during concrete construction. RILEM proceedings 11. H.W.
Reinhardt ed. Chapman and Hall, London