See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/275653064

Cyclic Loading Responses of Cement-Stabilised Base Materials: An

Investigation on Moduli for Pavement Design

Conference Paper · February 2015

CITATIONS READS

2 1,690

4 authors:

Korakod Nusit Peerapong Jitsangiam

Naresuan University Chiang Mai University
15 PUBLICATIONS   33 CITATIONS    77 PUBLICATIONS   215 CITATIONS   

SEE PROFILE SEE PROFILE

Jayantha Kodikara Ha H. Bui

Monash University (Australia) Monash University (Australia)
274 PUBLICATIONS   2,344 CITATIONS    95 PUBLICATIONS   1,205 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Unsaturated soil modelling using MPK framework View project

Computational failure modelling of geomaterials with SPH View project

All content following this page was uploaded by Korakod Nusit on 01 May 2015.

The user has requested enhancement of the downloaded file.

 Cyclic loading responses of cement-stabilised base materials: An
investigation on moduli for pavement design
K. Nusit1, P. Jitsangiam1, A.M. ASCE, J. Kodikara2, A.M. ASCE and H. Bui2 A.M. ASCE
1
Department of Civil Engineering, School of Civil and Mechanical Engineering, Curtin University, Kent Street,
Bentley, Perth, Western Australia 6102, Australia; email: [email protected];
[email protected]
2
Department of Civil Engineering, Clayton Campus, Monash University, Victoria 3800, Australia; email:
[email protected]; [email protected]

ABSTRACT

In the general Mechanistic-Empirical pavement design guideline, the design parameter normally used
in representing the response of stabilised pavement material is the elastic modulus measured from a
statically-monotonic compression test. Nevertheless, this elastic modulus may not respond to actual
behaviour of real pavement which is subjected to cyclic loading from moving vehicles. Using the
elastic modulus in pavement analysis could lead to an inaccurate estimation of the stress-strain
relationship which is relevant to a pavement response prediction under traffic loading conditions. In
this research, the dynamic modulus of cement-stabilised material measured from Asphalt Mixture
Performance Tester was examined to be used as a pavement design parameter with consideration of
cyclic (traffic) loading conditions. The laboratory results from this research reveal that the cyclic
response of cement-stabilised material in term of the dynamic modulus is not much affected by a
variation on temperatures and loading frequencies. However, the dynamic modulus is greatly
influenced by cement contents and curing periods. Moreover, the elastic moduli measured based on
different strain rates was also examined. In order to compare the effects of modulus to the pavement
response, finite element analysis was performed in this research by altering the modulus of base
course layer. Flexural modulus of cement-stabilised base material determined by Chummuneerat et al.
(2013) was also included in the finite element analysis. The results of the finite element analysis show
that a tensile strain at a critical location can be reduced by 20% if the elastic modulus of a base course
layer were replaced by the dynamic modulus and the flexural modulus. In addition, stress induced by
the traffic load can be evenly transferred to the subgrade layer by applying the dynamic modulus and
the flexural modulus in the analysis.

Keywords: Cyclic Response, Dynamic Modulus, Finite Element Method, Cement-Stabilised Base

1 INTRODUCTION

The cement-stabilised materials are commonly used in a high traffic road pavement due to its
relatively high stiffness, and its capability to evenly distribute the traffic load to an existing ground
(subgrade) layer. This cement stabilisation technique is also applied to road construction using poorly-
graded materials (i.e., sub-standard materials) in a pavement base layer, due to well-graded materials
being very expensive or unavailable (Berthelot et al. 2010). To date, the road pavement containing
cement-stabilised base material can be designed using various approaches; however, mechanistic
design approach is becoming more preferable (Huang 2004). In mechanistic pavement design,
pavement response in terms of stresses and strains induced by the moving wheels and axle load is
significant. The reason is suitable pavement materials and proper thicknesses are fundamentally
designed based on the pavement response. Moreover, estimated tensile strain is very important to the
fatigue performance of cement-stabilised base material. This can be distinguished from the commonly
used fatigue-life empirical formula adopted in Australia (Austroads 2010) as shown in (1).

Nf = RF[((113,000/E)0.84+191)/]m (1)

which RF is reliability factor for cement-stabilised materials, Nf is allowable number of standard load
repetitions, E is material modulus,  is horizontal micro-tensile strain at the bottom of the base, and m
is the damage exponent. The value of m recommended by Austroads (2010) is 12. According to the
guideline, the application of (1) is limited by the modulus of cement-stabilised material. The equation is
only suitable for cement-stabilised material with the value of elastic modulus ranges from 2,000 to
10,000 MPa (Austroads 2010). The fatigue-life formula shows that slightly change in predicted tensile
strain value may leads to a significant difference in predicted fatigue life. This is because of the
relatively high exponent of fatigue-life equation. Accordingly, an appropriate modulus input is
necessary for mechanistic pavement design. It means that design modulus should satisfactory
represents material response under the traffic loads.

In Australia, the flexural modulus is recommended by Austroads (2010) as a design modulus input due
to its potential to represent the tension failure of pavement behaviour in the field. Theoretically, the
flexural modulus is the modulus of a rectangular beam specimen subjected to a cyclic or monotonic
bending force. It should be noted that the calculation procedure for the stress-strain relationship in
pavement structure was developed based on the theory of elasticity. However, an accurate method for
determining the elastic modulus from a beam under bending forces is not yet available (Iyer 2005).
Moreover, parameters measured from flexural tests are largely influenced by sample preparation,
handling, curing, and sample quality. This explicitly produces a significant degree of test result
uncertainties (Walker and Bloem 1957). On the other hand, the National Cooperative Highway
Research Program (2004) suggests the elastic modulus as a design input for mechanistic-empirical
pavement design of cement-stabilised base course. In order to determine the elastic modulus, the
stress-strain response obtained from the Unconfined Compressive Strength (UCS) test is required.
This test can be performed according to ASTM C469 (2010). Alternatively, the elastic modulus of the
cement-stabilised base material can be estimated from the empirical formula. However, pavement
structure in the field is subjected to cyclic compressive loads from moving vehicles, which is totally
different from the UCS test condition. In addition, it had recently been proved by many works that
dynamic moduli of cement-stabilised base material are considerably greater than their elastic moduli
measured from monotonic loading test (Mindess and Young 1981; Neville 1998; Lee, Kwan, and
Zheng 2013; Kolias and Williams 1980). It should be noted that the dynamic moduli considered in
previous literatures were measured from either ultrasonic pulse velocity or vibration resonance
methods. Apart from the effects of loading condition, Bischoff and Perry (1991) characterised the
behaviours of concrete under the fast compressive-monotonic loading. Based on their study, the
material strength and modulus were also influenced by the strain rate of loading. The strain rates of
monotonic loading fall between 5x10-5 per second and 5x10-4 per second were considered as static
loading range, whereas the strain rates fall between 10-3 per second and 10-2 per second were
corresponded to the earthquake (dynamic) loading range. Xiao, Li and Lin (2008) likewise performed a
monotonic loading test on concrete specimens using different values of strain rate. From their tests,
the initial elastic modulus was increased from 1.23x104 MPa to 1.60x104 MPa when the tested strain
rate was changed from 10-5 per second to 10-1 per second.

From literature reviews, it can be noted that the modulus input for mechanistic pavement design
recommended by various guidelines still contains some limitations. A reliable modulus input is
necessary in order to capture the cyclic response of a pavement structure under repetitive loading
conditions. In this study, an alternative method for modulus measurement was introduced. The testing
machine, namely Asphalt Mixture Performance Tester (AMPT), which has been extensively used for
characterising the dynamic response of asphalt concrete material, was used to determine the dynamic
moduli of cement-stabilised materials in this study. By using AMPT, the cyclic load can be directly
applied to the test specimens. Therefore, same testing scheme as asphalt concrete was adapted and
applied to the study materials. In addition, compressive loading rate and test temperature can be
controlled during the dynamic modulus measurement by AMPT. Accordingly, effects of loading rate,
temperature, curing time and cement content were also examined in this research. At the final stage of
this research, the finite element analysis was performed in order to investigate the effects of modulus
types to the pavement response.

2 MODULUS OF CEMENT-STABILISED BASE MATERIAL

In this research, elastic moduli interpreted from UCS test using different test strain rates was
characterized. Two sets of UCS test were conducted based on different strain rates. The first set of
UCS test was subjected to strain rate recommended by the test standard (AS 5101.2.2 2008), that is
0.0087 per minute (1.45x10-4 per second). An additional strain rate of 0.0667 per minute (1.11x10-3
per second) was selected for the second set of UCS test as it is coincided with earthquake (dynamic)
loading range (Bischoff and Perry 1991). After that, the dynamic moduli were determined by
considering the effects of cement content, curing time, temperature and loading frequency. Moreover,
flexural modulus and relevant information from previous literature (Chummuneerat et al. 2013) were
summarised and compared with the test results from this research.

2.1 Elastic modulus of CTB

Korakod et al. (2014) determined the Optimum Moisture Content (OMC) and other index properties of
cement-stabilised crushed rock base material. The percentage of cement by aggregate weight greater
than 3% was selected since their research focuses on characterising the behaviours of stabilised
material. According to Austroads (2008), at least 3% of cement content is required to achieve the
structural characteristics of stabilised or bound pavement. Therefore, the specimens with cement
content equal to 4%, 5% and 6% were tested in their research. After the OMC values were determined
from the compaction test, the UCS specimens were prepared based on AS 5101.2.2 (2008). Table 1
summarises the values of OMC, MDD, UCS and elastic moduli at different curing duration.

An additional UCS test on 6 specimens was conducted in this research. The same specimen
preparation and test procedures as performed by Korakod et al. (2014) were employed, except the
strain rate was changed. Three specimens with 5% cement content and three specimens with 6%
cement content were cured at 7 days mark and tested at strain rate equal to 0.0667 per minute.
Comparison between the stress-strain curves from the tests with different strain rate are shown in
Figure 1.

Table 1: Summary of UCS test results and compaction test (by Korakod et al. 2014)
Cement OMCa MDDb 7 Days 28 Days
content UCSc Ed UCSc Ed
(MPa) (MPa) (MPa) (MPa)
(%) (%) (g/cm3) 0.0087/min 0.0677/min 0.0087/min 0.0677/min 0.0087/min
5 2.30 6.10 8.0 8.5 455 1,060 10.4 1,770
6 2.33 6.40 9.6 10.6 680 1,200 12.5 2,350
a
Optimum moisture content.
b
Maximum dry density.
c
Unconfined compressive strength averaged from three UCS specimens.
d
Secant elastic modulus averaged from three UCS specimens.

Figure 1 shows the similar finding by Xiao, Li and Lin (2008). UCS and the elastic moduli from
monotonic compression tests increase with respect to the test strain rates. Average elastic moduli of
5% cement content specimen increased from 455 to 1,060 MPa when test strain rates were changed
from 0.0087 per minute to 0.0667 per minute. Same observation was found from the specimen with
6% cement content that is average elastic moduli rose up almost two times if the test strain rate was
increased. It should be noted that, the calculation method for elastic modulus in this research is based
on the procedure recommended by ASTM C469 (2010).

Figure 1. Strain rate effect to the stress-strain curve of CTB with 5% and 6% cement content

It can be seen from the test results that, for monotonic compression test, elastic modulus of the
cement-stabilised base is influenced by the strain rate. This finding is important because of strain rate
in the field caused by traffic loads may not equal to the test value; in fact, loading condition in the field
is usually complex, hence in-situ strain rate would vary greatly. Accordingly, the elastic modulus of
cement-stabilised base material is influenced by the strain rate, which may result in imprecise
estimation of stress-strain response by the mechanistic pavement design.

2.2 Dynamic modulus of CTB

The dynamic modulus of cement-stabilised base material measured from AMPT was also analysed in
this research. The configuration and component of AMPT are illustrated in Figure 2. AMPT was
developed by the National Cooperative Highway Research Program (NCHRP) to characterise the
asphalt concrete performances under cyclic loading condition (Bonaquist, Christensen, and Stump
2008). To determine the critical strain of asphalt concrete material for mechanistic-empirical design,
the dynamic modulus was recommended to be the design parameter. Basically, dynamic modulus is
the modulus of a visco-elastic material under sinusoidal loading conditions. In general, dynamic
modulus can be measured from laboratory testing or calculated from the predictive models (Robbins
2009). Mathematical definition of dynamic modulus is the absolute value of a complex modulus;
therefore it can be calculated by (2).

|E*| = |*/*| = |(0/0)ei| = 0/0 (2)

which  is stress amplitude,  is strain amplitude,  is angular frequency, and  is the phase angle.
For elastic material, the phase angle is equal to 0°, while it becomes 90° for viscous materials. Figure
3 shows the sinusoidal stress () and sinusoidal strain () plotted against time (t) with the phase lag
of  degree.

Figure 2. Asphalt Mixture Performance Tester (AMPT)

The same standard practice as used in asphalt concrete testing (AASHTO PP 61 2009) was employed
to guide the dynamic modulus measurement of cement-stabilised base material in this research.
According to the standard of test, the dimensions of the test specimen are a height of 150 mm and a
diameter of 100 mm. At least two specimens are required for each test, therefore six specimens
represent different cement content and curing duration were prepared (see Table 2). Besides its
potential to characterise the cyclic response of a material, the AMPT also has a capability to control
and vary test temperatures and loading frequencies. Accordingly, the cement-stabilised base
specimens were tested at 4 °C, 20 °C and 40 °C. At each test temperature, the test frequencies were
10 Hz, 1 Hz, 0.1 Hz, plus an additional 0.01 Hz for the test at 40 °C. The test dynamic strain ranges
from 45 to 85 microstrains were employed during the test. More detail on specimen preparation and
test procedure can be found from Korakod et al. (2014). The average dynamic moduli (from 2 samples
per set) determined from AMPT are summarised in Table 2.

Table 2 indicates that, temperatures and loading frequencies slightly affect the dynamic moduli of the
cement-stabilised base material. However, the dynamic modulus notably relies on the cement content
and curing duration of the specimen. Moreover, the value of dynamic modulus is much higher than the
elastic modulus measured from monotonic compression test (see Figure 4). It should be highlighted
that, elastic modulus from monotonic compression test with the strain rate falls within earthquake
loading range (strain rate = 0.0667 per minute) is considerably smaller than the dynamic modulus
measured by AMPT.
Table 2: Dynamic moduli of cement-stabilised base specimens
Cement Curing Dynamic Modulus (MPa)
content 4 °C 20 °C 40 °C
(%) (Days) 10 Hz 1 Hz 0.1 Hz 10 Hz 1 Hz 0.1 Hz 10 Hz 1 Hz 0.1 Hz 0.01 Hz
5 28 19,689 18,719 18,112 18,269 17,678 17,321 17,582 16,855 16,411 16,301
6 7 19,192 18,820 18,496 18,432 18,043 17,678 17,840 17,356 17,106 17,165
6 28 24,545 24,134 23,612 23,515 22,933 22,451 22,776 22,085 21,533 21,327

Figure 3. Sinusoidal stress, sinusoidal strain and phase angle

2.3 Flexural modulus of CTB

Chummuneerat et al. (2013) conducted the beam fatigue test on cement-stabilised base specimens
with the cement content of 3% and 5%. The curing duration for every specimen was 28 days. The
tests were conducted using various strain magnitude which fall within a range of 50 to 200
microstrains. Prior to the fatigue test, the flexural modulus was measured and assigned as the initial
flexural modulus for every individual specimen. According to the test standard (AG:PT/T233 2006), the
fatigue test is performed under a continuous haversine loading frequency of 10 Hz. The fatigue test is
continued until the flexural modulus of the specimen is reduced to half of the initial value or the one
million loading cycle is obtained. The test temperature was 25 °C for every test specimens. Figure 4
shows the initial flexural modulus of specimen with 5% cement content plotted against the elastic
moduli and dynamic moduli determined in this research.

Figure 4. Elastic moduli, dynamic moduli, and initial flexural modulus of cement-stabilised base
specimens with 5% cement content and cured at 28 days

Based on available information shown in Figure 4, the elastic modulus measured from monotonic-
compression test is much lesser than those of the dynamic modulus and initial flexural modulus.
Nevertheless, the initial flexural modulus of the specimen with 5% cement content approximately
equals to the dynamic modulus determined by the AMPT. It should be noted that, this comparison was
made by using the dynamic modulus and initial flexural modulus measured at 20 °C and 25 °C
respectively. The applied dynamic strain by the AMPT was 45 to 85 microstrains, therefore the initial
flexural modulus measured at 50 microstrain was chosen. However, the sinusoidal load of 10 Hz
frequency was used by the AMPT, whereas haversine load of 10 Hz was employed during the beam
fatigue test.

In the next section, the stress-strain response of pavement structure with a cement-stabilised base
layer was analysed by the Finite Element Method (FEM). The analysis was performed based on
different moduli which were characterised in the previous section. Only the moduli of cement-stabilised
base material of 5% cement content and 28 days curing point were used in the finite element
calculation. It should be noted that, the purpose of finite element analysis is to identify the effects of
modulus types to the pavement response and fatigue performance of the cement-stabilised base
course.

3 FINITE ELEMENT ANALYSIS

Three dimensional (3D) finite element simulations for each modulus types were carried out using
ABAQUS 6.12. The typical pavement structure and the finite element model employed in the
simulation are illustrated in Figure 5. In order to compare the effects of modulus types, three models
with different moduli of a cement-stabilised base layer were established. The element type for the
pavement structure model was C3D20R (Continuum 3-Dimensional 20 node elements with reduced
integration) brick element. A standard axle load recommended by Austroads (2010) was used in the
calculation as the applied load. This means the pavement structure was subjected to a dual-wheeled
single axle with the applied load of 80 kN. Figure 6 shows a rectangle and two semicircles shape of
contact area between single tyre and pavement surface. Tyre pressure of 750 kPa is recommended
by the guideline (Austroad 2010). To model the load contact area, Huang (2004) recommended to
transform a rectangle and two semicircles shape to be only single rectangle shape as shown in Figure
6. Therefore, a rectangle with dimension of 0.196 m x 0.135 m was used in this research. However,
half wheel load was considered in the calculation due to the symmetry (Figure 5). The boundary
conditions of finite element model shown in Figure 5 are established as follow; (1) vertical
displacements of the nodes on the plane ABCD are fixed, (2) orthogonal displacements to the plane
AEHD and DHGC are fixed, and (3) orthogonal displacements to the planes of symmetry (BCGF and
ABFE) are prevented.

Figure 5. Finite element model, material properties and loading condition

Figure 6. Contact area between single tyre and pavement surface (Left) and equivalent contact area
(Right)

In the finite element calculation, Model 1 represents the pavement structure with cement-stabilised
base course analysed based on elastic modulus, whereas the dynamic modulus and initial flexural
modulus were used in the calculation of Model 2 and 3 respectively (Figure 5). The properties of
asphalt concrete and subgrade were kept constant throughout the analysis of the three models. By
applying the same traffic load to all the models, deformed mesh and horizontal stress generated by
the traffic load only were presented in Figure 7. The figure clearly shows that induced horizontal stress
can be evenly transferred to the subgrade layer by applying the stiffer material as a pavement base
course (Model 2 and 3).

Figure 7. Deformed mesh and horizontal stress distribution induced by the wheel load

Figure 8 shows the horizontal strains induced by the traffic load along the axis a-a’ and b-b’ (see
Figure 5). Positive value indicates tensile strains and vice versa. The critical tensile strain of cement-
stabilised base layers at point X reduced from 1.26x10-4 to 0.98x10-4 and 0.99x10-4, when the elastic
modulus was replaced by dynamic modulus and initial flexural modulus in the analysis, respectively.
Similarly, critical tensile strain at point Y from the analysis using elastic modulus, dynamic modulus
and initial flexural modulus are 1.08x10-4, 0.86x10-4 and 0.87x10-4 respectively. However, the critical
tensile strains of Model 2 and 3 are almost identical because of the magnitudes of modulus used for
both models are not much different. Assuming (1) is applicable to the cement-stabilised base material
with elastic modulus greater than 10,000 MPa, the reduction in tensile strain approximately of 25
microstrain leads to an increasing in predicted Nf of 2x1011 cycles. This approximation was made
based on RF = 1 (95% of desired project reliability) and E = 20,500 MPa.

Figure 8. Induced horizontal strain along the section a-a’ (Left) and b-b’ (Right) of the three models

4 CONCLUSION

The modulus of the cement-stabilised material is important to an estimation of its fatigue performance
in the new mechanistic pavement design and analysis. In this research, the elastic modulus and the
dynamic modulus of a cement-stabilised base material were determined and characterised, whereas
the flexural modulus obtained from Chummuneerat et al. (2013) is also examined. The finite element
analysis of typical pavement section was also performed based on those three different modulus types
of a cement-stabilised base course layer. Results from the finite element analysis show that analyses
using the dynamic modulus and the flexural modulus produced lower values of a tensile strain at the
critical location in pavement than that analysed using the elastic modulus. The reduction in a tensile
strain value results in an increasing in fatigue life of the cement-stabilised pavement structure
according to Austroads empirical formula (Austroads 2010). The magnitudes and performance trends
of strains computed using the dynamic and the flexural modulus are almost identical. According to the
 test procedure, both types of modulus are determined based on the cyclic response of material.
However, the cyclic response of cement-stabilised base material characterised by AMPT is the
simplest in term of sample preparation, handling and testing.

5 ACKNOWLEDGEMENTS

The authors wish to express their gratitude to the Australian Research Council (ARC) for financial
support for this research, under the ARC Linkage Scheme (LP 130100884), conducted at the
Department of Civil Engineering, Curtin University.

REFERENCES

AASHTO PP61 (2010). “Standard practice for developing dynamic modulus master curves for hot mix asphalt
(HMA) using the asphalt mixture performance tester (AMPT).“ American Association of State and Highway
Transportation Officials, Washington, D.C.
AS 5101.2.2 (2008). “Methods for preparation and testing of stabilized materials.“ Method 2.2: Sampling -
Preparation of Stabilized Pavement Materials, Standards Australia, Sydney, Australia.
ASTM C469 (2010).“Static modulus of elasticity and Poison’s ratio of concrete in compression.“ ASTM
International, West Conshohocken, PA.
Austroads (2006). “Fatigue life of compacted bituminous mixes subject to repeated flexural bending.“ AG:
PT/T233, Austroads, Australia.
Austroads (2008). "Guide to pavement technology part 4A: Granular base and subbase materials." Austroads,
Australia.
Austroads (2010). "Guide to pavement technology part 2: Pavement structural design." Austroads, Australia.
Berthelot C., Podborochynski D., Marjerison B., and Saarenketo T. (2010). “Mechanistic characterization of
cement stabilized marginal granular base materials for road construction.” Canadian Journal of Civil
Engineering, 37, 1613–1620.
Bischoff, P. H., and Perry, S. H. (1991). “Compressive behaviour of concrete at high strain rates.“ Materials and
Structures, 24, 425-450.
Bonaquist, R. F., Christensen, D. W., and Stump, W. (2008). "Simple performance tester for Superpave mix
design: First-article development and evaluation." Washington, D.C., USA.
Chummuneerat, S., Jitsangiam, P., Nikraz, H., and Patel, R. "Fatigue characteristics of cement treated base for
Western Australian roads." Proc., The Second International Conference on Engineering and Applied
Sciences.
Huang, Y. H. (2004).“Pavement analysis and design.” Pearson Education, INC., Upper Saddle River, N.J.
Iyer, H. (2005). "The effects of shear deformation in rectangular and wide flange sections." Master of Engineering,
Virginia Polytechnic Institute and State University, Blacksburg, Virginia.
Kolias, S., and Williams, R. I. T. (1980). "Relationships between the static and the dynamic moduli of elasticity in
cement stabilized materials." Materials and Structures, 13(74), 99-107.
Lee, P. K. K., Kwan, A. K. H., and Zheng, W. (2013). "Tensile strength and elastic modulus of typical concrete
made in Hong Kong." HKIE Transactions, 7(2), 35-40.
Mindess, S., and Young, J. (1981). “Concrete.“, Prentice-Hall, Englewood Cliffs, N.J.
National Cooperative Highway Research Program (2004). "Guide for mechanistic-empirical design of new and
rehabilitated pavement structures." Natinal Cooperative Highway Research Program, USA.
Neville, A. M. (1998). “Properties of concrete.“ Addison Wesley Longman Limited.
Nusit, K., Jitsangiam, P., Nikraz, H., and Hewa Thalagahage, R. D. (2014). “Dynamic modulus characteristics of
bound cement-treated crushed rock base course.“ International Conference on Advances in Civil
Engineering for Sustainable Development, 27-29 August 2014, Nakhon Ratchasima, Thailand.
Robbins, M. M. (2009). "An Investigation into dynamic modulus of hot-mix asphalt and Its contributing factors."
Master of Science, Auburn University, Alabama, USA.
Walker, S., and Bloem, D. L. (1957). “Studies of fleural strength of concrete - Part 3: Effects of variations in testing
procedures.“ Paper presented at the ASTM Proceedings, Philadelphia.
Xiao, S., Li, H., And Lin, G. (2008). “Dynamic behaviour and constitutive model of concrete at different strain
rates.“ Magazin of Concrete Research, 60(4), 271-278.

View publication stats