An Investigation on Resilient Modulus

of Bituminous Mixtures

S. Deepa and J. Murali Krishnan

Abstract IRC:37-2012 suggests to use resilient modulus as an input parameter in

the design of bituminous pavements in India. IRC:37 specifies the values of resilient
modulus at different temperatures for Bituminous Concrete (BC) and Dense Bitu-
minous Macadam (DBM) mixes with various grades of unmodified and modified
binders. However, IRC:37 recommends identical values of resilient modulus to be
used for the mixes, irrespective of the type of modified binder used. At higher tem-
peratures, it can be noticed that IRC:37 tabulated resilient modulus values for the
mixes with unmodified binder are higher than those with modified binders. Also,
the influence of Poisson’s ratio is not incorporated since a constant value of 0.35 is
used for all the mixes. In this investigation, BC mixtures with an unmodified (VG30)
and three modified binders (CRMB, plastomer and elastomer) were tested at three
different temperatures (25, 30 and 35◦ C) at 1 Hz frequency. Tests were conducted
as per ASTM D7369-11. Samples of diameter 152.4 mm were used for the test.
The sample was subjected to 200 cycles, and the last five cycles were used for the
computation of Poisson’s ratio and resilient modulus. Both instantaneous and total
resilient modulus were computed, and these values were compared with the IRC:37
specified values. The possible reasons for the difference in the modulus values from
the present study and IRC:37 are explored.

Keywords Resilient modulus · Elastomer · Plastomer · Pavement design

S. Deepa · J. Murali Krishnan (B)

Department of Civil Engineering, Indian Institute of Technology Madras,
Chennai 600036, India
e-mail: [email protected]
S. Deepa
e-mail: [email protected]

© Springer Nature Singapore Pte Ltd. 2020 895

T. V. Mathew et al. (eds.), Transportation Research, Lecture Notes
in Civil Engineering 45, https://doi.org/10.1007/978-981-32-9042-6_71
896 S. Deepa and J. Murali Krishnan

1 Introduction

Material characterization has a vital role in the design of bituminous pavements, and
hence, substantial research has been carried out to decide on the material parame-
ter to be used in the bituminous pavement design procedure. Different studies have
adopted different modes of testing to characterize the bituminous mixtures. The most
widely used test methods employ uniaxial compression, indirect tension and beam
flexure for material characterization. The current Mechanistic-Empirical Pavement
Design Guide (M-EPDG) developed by AASHTO [11] recommends dynamic mod-
ulus measured in uniaxial compression to be used in pavement design. However, an
earlier version, AASHTO Design Guide [1], recommended the use of indirect ten-
sion mode to determine the material modulus, termed as resilient modulus. In 1982,
the indirect tension test (IDT) was adopted by ASTM as a standard test method for
measuring resilient modulus of bituminous mixtures [2]. After subsequent revisions,
the test protocol was withdrawn in 2003 and finally standardized under ASTMD7369
[5]. Several studies have used the indirect tension mode of testing to characterize
bituminous mixtures [10, 13, 14].
Pavement design in India is carried out as per IRC:37-2012 [7], wherein resilient
modulus is used as the input parameter for bituminous materials in the stress analysis
procedure. IRC:37-2012 tabulates the resilient modulus of BC and DBM mixes with
modified and unmodified binders. IRC:37-2012 claims that the tests have been carried
out following ASTM: D7369-09 [4]. However, it is clear from the experimental
results that ASTM: D4123-82 [2] has been followed, and the resilient modulus values
were calculated assuming a Poisson’s ratio of 0.35 . It can be observed from the
IRC:37-2012 tabulated values that resilient modulus is the same for all the mixes
irrespective of the type of modifier. Hence, the IRC:37-2012 tabulated values of
resilient modulus do not distinguish between different types of modified binders.
The influence of Poisson’s ratio on resilient modulus is also not captured since a
constant value of 0.35 is used for all the mixes.
ASTM: D7369-11 suggests three gauge lengths that can be used in the repeated
load indirect tension test: one-fourth, one-half and full diameter of the sample [5].
The standard also states in Note 3 that “The results obtained with gauge length
of one-fourth of the diameter of the specimen have the best precision”. However,
IRC:37-2012 [7] seems to have used the largest gauge length for all the tests. This
will result in less precise results since the sample non-homogeneity exerts a greater
influence when the gauge length is large. Small gauge lengths result in smaller mea-
sured deformations and hence lead to issues related to signal capture by the LVDTs.
In the current investigation, tests were conducted following the ASTM: D7369-11
procedure adopting a gauge length of half the sample diameter [5]. A comparison
of the experimental results with IRC:37-2012 [7] values and that computed follow-
ing the ASTM: D4123-82 [2] post-processing has been carried out. The results are
discussed in the further sections.
An Investigation on Resilient Modulus of Bituminous Mixtures 897

2 Experimental Investigations

2.1 Materials

Bituminous Concrete Grade 2 (BC-Grade 2) with nominal maximum aggregate size

of 13.2 mm was employed in the current investigation. The gradation as per MoRTH
has been adopted [12]. Four binders were used in this study, an unmodified bitumen
and this base bitumen modified with three types of modifiers—elastomer, plastomer
and crumb rubber. The base bitumen used is an air-blown bitumen of VG30 grade as
per IS: 73-13 [9]. The plastomer, elastomer and crumb rubber modified bitumen used
are PMB(P)40, PMB(E)40 and CRMB-60, respectively [8]. Further, the bituminous
concrete mixes with VG30, plastomer, elastomer and CRMB will be referred to as
BC-VG30, BC-PMB(P), BC-PMB(E) and BC-CRMB.

2.2 Sample Preparation

The bituminous mixtures were short-term aged for 4 hours at mixing temperature and
half an hour at compaction temperature. The mixing temperature for the mix with
unmodified binder was 165 ◦ C, whereas 185 ◦ C was adopted for mixes with modified
binder. Compaction was carried out in the shear box compactor [6] by applying a
constant vertical load of 600 kPa and a constant shear angle of 4◦ . The fabricated
beam was 450 mm long and 150 mm wide, and the height was 169 mm, based on
targeted air voids of 4 ± 0.5%. Two cylinders, each of 150 mm diameter, were cored
from the compacted beam. Further, each cylinder was sliced into two samples of
152.4 mm diameter and 63 ± 0.5 mm height. Thus, a total of four samples were
obtained from a compacted beam as shown in Fig. 1.

2.3 Test Procedure

Prior to the conduct of the resilient modulus test (repeated load diametral indirect
tension test), an indirect tension test [3] was carried out in order to estimate the
failure load for each type of mix. The IDT was conducted on samples with 101.6 mm
diameter and 63 mm thickness. The IDT samples were also cored and sliced from
a beam compacted using the shear box compactor [6]. Since the sample diameter
for the IDT test is small, three cylinders of 100 mm diameter could be cored from a
compacted beam. Further, each cylinder was sliced into two samples, each of 101.6
mm diameter and 63 mm thickness. In the IDT test, the sample is loaded to failure and
10–20% of the failure load was used as the input load in the repeated load diametral
indirect tension test.
898 S. Deepa and J. Murali Krishnan

Fig. 1 Resilient modulus test samples cored and sliced from a compacted beam

Table 1 Input load levels

Temperature (◦ C) Pcyclic (N) and (% loading)
BC-VG30 BC-CRMB BC-PMB(E) BC-PMB(P)
25 2354 (10) 1600 (10) 2170 (10) 2350 (10)
30 1330 (8) 1059 (9) 1458 (9) 1557 (9)
35 740 (6) 900 (9) 948 (9) 1205 (9)

After sample preparation, the next step involved was the fixing of LVDTs. The
marking of the gauge positions, fixing of studs and mounting of LVDTs were carried
out as per ASTM: D7369-11 [5]. Gauge length of one-half the sample diameter
was used in the current investigation. Once the samples are mounted on the test
setup, ASTM: D7369-11 suggests testing the sample in two orientations (0 and 90◦
rotation) and measure the deformations from both the planes of the sample. The load
waveform is haversine compression with 0.1 s load duration followed by 0.9 s rest
period. Table 1 shows the load levels used in the current study for repeated diametral
indirect tension test.
In order to check the repeatability of results, the current investigation introduced
slight modifications to the ASTM: D7369-11 procedure [5], and they are described
further. The standard recommends applying 100 preconditioning cycles to the sample
and using the last five cycles for modulus calculation. This is to ensure that the
material response would attain a steady state for the purpose of modulus computation.
However, there is no certainty on whether 100 cycles are sufficient to yield a steady
An Investigation on Resilient Modulus of Bituminous Mixtures 899

Fig. 2 Test procedure

state of response. Hence, in the current study, 200 cycles were applied to the sample,
and the last five cycles were used for modulus determination.
The procedure adopted for the test in the current investigation is described briefly
as follows. Initially, the sample was mounted with LVDTs on a single plane (Plane
1) and was subjected to 200 load cycles. The resulting deformations in the vertical
and horizontal directions were measured using the LVDTs. This was followed by a
rest period of 15 min after which a second trial was conducted. After a rest period
of 15 min, the sample was rotated through 90◦ and the test was repeated. Further,
LVDTs were mounted on Plane 2 of the sample, and the same procedure adopted for
Plane 1 was repeated. Generally, as per ASTM: D7369-11, only one trial is performed
in each orientation. However, this modification was introduced for the purpose of
checking the repeatability between the trials. Also, since the applied loads are only
a small percentage of the failure load (<10%), the sample is not damaged. The test
procedure is schematically represented in Fig. 2.

2.4 Post-processing

Post-processing of the experimental data was conducted as per the ASTM: D7369-
11 standard [5]. ASTM:D7369-11 suggests the calculation of two resilient modulus
values based on the nature of recoverable deformation. First one is an instantaneous
resilient modulus computed using the instantaneous recoverable deformation mea-
sured immediately after the load pulse, and the next is the total resilient modulus
900 S. Deepa and J. Murali Krishnan

computed using the total recoverable deformation measured at the end of a load–
unload cycle. Poisson’s ratio (μ) and resilient modulus (MR) were determined using
Eqs. 1 and 2.
I4 − I1 ( δδhv )
μ= (1)
I3 − I2 ( δδhv )

Pcyc (I1 − I2 .μ)

MR = (2)
δh t

where δv and δh are the recoverable deformations, t is the sample thickness, and Pcyclic
is the cyclic load applied. The values of the constants are I4 = −1.0695, I1 = 0.2339,
I2 = −0.7801 and I3 = 0.3075, and these have been specified in ASTM:D7369-11
[5] for the gauge length adopted in the current study. The computation of total and
instantaneous recoverable deformations was carried out as per the regression pro-
cedure suggested by ASTM:D7369-11. Since the current investigation employed
200 cycles against 100 cycles specified in the standard, horizontal and vertical de-
formations for the last five cycles prior to both 100th and 200th cycles have been
compared here. These deformations were used to compute the total and instantaneous
resilient modulus corresponding to the 100th and 200th cycles. For demonstration,
the instantaneous and total recoverable deformations and the corresponding Pois-
son’s ratio and resilient modulus for BC-VG30 (P-1 and 0◦ orientation) at 25 ◦ C are
shown in Tables 2 and 3.

Table 2 Instantaneous recoverable deformations for BC-VG30 at 25 ◦ C

Deformation Pulse 96 Pulse 97 Pulse 98 Pulse 99 Pulse 100 Mean
(µm)
H V H V H V H V H V
1.968 −4.240 1.948 −4.201 1.965 −4.143 1.937 −4.225 1.96 −4.206
Load (N) 2364 2354 2331 2344 2360
μ 0.41 0.41 0.43 0.40 0.42 0.41
MR (MPa) 11061 11113 11218 10974 11144 11102

Table 3 Total recoverable deformations for BC-VG30 at 25 ◦ C

Deformation Pulse 96 Pulse 97 Pulse 98 Pulse 99 Pulse 100 Mean
(µm)
H V H V H V H V H V
2.720 −5.953 2.683 −5.898 2.712 −5.876 2.68 −5.948 2.737 −5.943
Load (N) 2364 2355 2333 2344 2360
μ 0.40 0.39 0.41 0.39 0.41 0.40
MR (MPa) 7303 7339 7319 7230 7307 7300
An Investigation on Resilient Modulus of Bituminous Mixtures 901

3 Analysis of Experimental Data

The experimental data for all the four mixes was collected at 25, 30 and 35 ◦ C. A total
of eight resilient modulus values were computed for each mix at each temperature
(two planes, two rotations and two recovery types) following ASTM: D7369-11
post-processing [5]. Resilient modulus was computed corresponding to each plane,
rotation and mode of recovery for all the mixes used in the study.

3.1 Instantaneous Resilient Modulus

The instantaneous resilient modulus was computed using the instantaneous recov-
erable deformation measured immediately at the end of loading pulse. Since the
time-dependent recovery is not captured, the instantaneous recoverable deformation
is smaller compared to the total recoverable deformation measured at the end of the
load–unload cycle. The average of the instantaneous resilient modulus over the last
five cycles prior to the 100th and 200th cycles are shown in Table 4. From Table 4, it
can be observed that the resilient modulus values do not vary more than 10% between
the 100th and 200th cycles for any mix at any temperature.

3.2 Total Resilient Modulus

Total resilient modulus was calculated from the total recoverable deformations, and
since the total recoverable deformation is larger than the instantaneous deformation,
the corresponding modulus values are lower. The average of the instantaneous re-
silient modulus over the last five cycles prior to the 100th and 200th cycles is shown
in Table 5.
From Table 5, it is clear that for mixes such as BC-VG30 and BC-PMB(P), there is
as much as 30% reduction in total modulus compared to the instantaneous modulus at
25 ◦ C. For BC-CRMB at 30 and 35 ◦ C, the total modulus is less than the instantaneous
modulus by more than 40%. This is true for all mixes at all the temperatures though
the margin of variation is different.

Table 4 Instantaneous resilient modulus (MPa) as per ASTM: D7369-11

Temperature (◦ C) BC-VG30 BC-CRMB BC-PMB(E) BC-PMB(P)
100th 200th 100th 200th 100th 200th 100th 200th
25 11102 11151 9277 9301 8678 8723 20572 21163
30 6193 6041 7643 7690 5455 5978 8374 8463
35 3952 3954 7279 7272 5263 5406 6113 6111
902 S. Deepa and J. Murali Krishnan

Table 5 Total resilient modulus (MPa) as per ASTM: D7369-11

Temperature (◦ C) BC-VG30 BC-CRMB BC-PMB(E) BC-PMB(P)
100th 200th 100th 200th 100th 200th 100th 200th
25 7300 7344 6449 6417 5905 5868 15079 15641
30 5372 5366 4703 4725 4530 4472 5984 6055
35 2586 2551 3810 3805 3156 3231 3869 3898

Comparing the modulus calculated using ASTM D7369-11 [5], it can be observed
that, at 25 and 30 ◦ C, the mix with plastomer modified binder has the highest resilient
modulus followed by BC-VG30, BC-CRMB and BC-PMB(E). At 35 ◦ C, BC-VG30
has the least modulus followed by BC-PMB(E), BC-CRMB and BC-PMB(P).

3.3 Total Resilient Modulus Calculated Using an Assumed

Poisson’s Ratio

Since IRC:37-2012 [7] has calculated the resilient modulus as per ASTM: D4123-82
[2], Poisson’s ratio of 0.35 was adopted for the calculation of resilient modulus. Since
ASTM: D4123-82 [2] does not clearly mention as how to measure the instantaneous
recoverable deformation, it is understood that IRC:37-2012 [7] used values of total
recoverable deformations and computed total resilient modulus. For the purpose of
comparing the results with that of IRC:37-2012 [7] tabulated data, the collected
experimental data was further analysed using the post-processing method suggested
in ASTM:D4123-82 [2], adopting a Poisson’s ratio of 0.35. Table 6 shows the total
resilient modulus values computed as per ASTM: D4123-82 [2]. As observed from
the Table 6, the resilient modulus computed as per ASTM: D7369-11 [5] is higher
than that computed using ASTM: D4123-82 [2]. The difference is due to the fact that
the actual Poisson’s ratio (computed using both horizontal and vertical deformations)
of the material is greater than the assumed Poisson ratio of 0.35.
It can also be observed that when resilient modulus is calculated using ASTM:
D4123-82 [2], assuming a Poisson’s ratio, the influence of the type binders on resilient

Table 6 Total resilient modulus (MPa) as per ASTM: 4123-82

Temperature (◦ C) Instantaneous resilient modulus (MPa)
BC-VG30 BC-CRMB BC-PMB(E) BC-PMB(P)
100th 200th 100th 200th 100th 200th 100th 200th
25 6790 6825 5521 5649 5014 5041 10772 11019
30 3824 3738 3701 3681 3710 3779 5263 5251
35 1729 1839 2915 2880 2182 2146 3633 3489
An Investigation on Resilient Modulus of Bituminous Mixtures 903

modulus cannot be effectively captured. For instance, at 30 ◦ C, BC-VG30, BC-

PMB(E) and BC-CRMB are observed to have an average resilient modulus value of
nearly 3700 MPa when a constant Poisson’s ratio is used for all the mixtures.

4 Comparison of Experimental Results with IRC:37-2012

The experimental data collected has been analysed using the post-processing meth-
ods suggested in ASTM D4123-82, and ASTM D7369-11 [2, 5]. Both instantaneous
and total resilient modulus were calculated using ASTM D7369-11 [5]. Since ASTM
D4123-82 [2] does not explain the regression procedure for finding instantaneous
deformation, only total resilient modulus was computed assuming a Poisson’s ratio
of 0.35. The next task is to compare these values with the IRC:37-2012 stipulated
values. Table 7 shows the IRC:37-2012 stipulated values of resilient modulus for
BC and DBM mixes with unmodified and modified binders. As observed from Ta-
ble 7, IRC:37-2012 does not effectively distinguish between the resilient modulus
of mixes with different modified binders. Also, IRC:37-2012 assigns lower resilient
modulus to the mixes with modified binder at higher temperatures (30 and 35 ◦ C) as
compared with the ones with unmodified binder. Figure3 provides a comparison of
the resilient modulus values specified by IRC:37-2012 and that determined by the
current investigation.
The reason for the difference in the IRC:37-2012 tabulated values with the values
calculated in the current study could be due to either the difference in the gauge
length used or the sample dimensions. IRC:37-2012 results are based on the use of
full sample diameter as the gauge length for measuring the horizontal deformations
whereas the current study used a gauge length of half sample diameter. Since shorter
gauge lengths give better precision, it is recommended to use short gauge lengths
compared to full diameter gauge length.

5 Conclusions

To summarize, the current study compared the resilient modulus of bituminous mix-
tures listed by IRC:37-2012 and that computed using ASTM: D7369-11 and ASTM:

Table 7 Resilient modulus of bituminous mixes (MPa) as per IRC:37-2012

Mix type Temperature (◦ C)
20 25 30 35 40
BC and DBM for VG30 3500 3000 2500 1700 1250
BC and DBM for modified bitumen 5700 3800 2400 1650 1300
904 S. Deepa and J. Murali Krishnan

(a) BC-VG30 (b) BC-CRMB

(c) BC-PMB(E) (d) BC-PMB(P)

Fig. 3 Comparison of resilient modulus determined using different approaches

D4123-82 post-processing. It was observed that the use of a computed value for
Poisson’s ratio rather than using an assumed value resulted in considerably higher
values of resilient modulus for all the mixtures. It was also found that the modified
binders have a significant influence on resilient modulus of the mixes, and hence, the
same value should not be employed for all the mixes as suggested in IRC:37-2012.
As per the experimental observations, at 25 and 30 ◦ C, BC-VG30 had higher re-
silient modulus than BC-CRMB and BC-PMB(E). However, as test temperature was
increased to 35 ◦ C, all the mixes with modified binders had higher resilient modulus
as compared to BC-VG30. On the other hand, IRC:37-2012 had reported lower mod-
ulus to the mixes with modified binders as compared to BC-VG30. It is observed that
IRC:37-2012 reported values are too low as compared to those obtained from the
experimental investigations. This could be either due to the difference in the gauge
length used or the sample dimensions. The current study employed one-half sample
diameter as the gauge length, while IRC:37-2012 values are based on full sample di-
ameter as gauge length. Current study employed 150-mm-diameter samples, whereas
the details on sample dimensions are not furnished by IRC:37-2012. To conclude,
An Investigation on Resilient Modulus of Bituminous Mixtures 905

the current study recommends using small gauge lengths for measuring deformation.
It is also recommended to compute Poisson’s ratio using measured horizontal and
vertical deformations rather than assuming a value of 0.35.

Acknowledgements The authors thank Department of Science and Technology, Govt. of India,
for funding this investigation. The grant number is DST/TSG/STS/2011/46. The authors acknowl-
edge the technical assistance provided by M/s IPC Global, Australia, during the conduct of the
experiments.
The authors acknowledge the opportunity provided by the 4th Conference of the Transportation
Research Group of India (4th CTRG), held at IIT Bombay, Mumbai, India, between 17 December
2017 and 20 December 2017 to present the work that forms the basis of this manuscript.

References

1. AASHTO (1993) Guide for design of pavement structures. American Association of State
Highway and Transportation Officials, Washington DC, USA
2. ASTM D4123-82 (1995) Standard test method for indirect tension test for resilient modulus
of bituminous mixtures. ASTM International, West Conshohocken, Pennsylvania, USA
3. ASTM D6931-12 (2012) Standard test method for indirect tensile (IDT) strength of bituminous
mixtures. ASTM International, West Conshohocken, Pennsylvania, USA
4. ASTM D7369-09 (2009) Standard test method for determining the resilient modulus of bitumi-
nous mixtures by indirect tension test. ASTM International, West Conshohocken, Pennsylvania,
USA
5. ASTM D7369-11 (2011) Standard test method for determining the resilient modulus of bitumi-
nous mixtures by indirect tension test. ASTM International, West Conshohocken, Pennsylvania,
USA
6. ASTM D7981-15 (2015) Standard practice for compaction of prismatic asphalt specimens by
means of the shear box compactor. ASTM International, West Conshohocken, Pennsylvania,
USA
7. IRC:37-2012 (2012) Guidelines for the design of flexible pavements. Indian Roads Congress,
New Delhi, India
8. IS 15462-04 (2004) Indian standard specification for polymer and rubber modified bitumen.
Bureau of Indian Standards, New Delhi, India
9. IS: 73-13 (2013) Specification for paving bitumen. Bureau of Indian Standards, New Delhi
10. Kim J, Lee HS, Kim N (2010) Determination of shear and bulk moduli of viscoelastic solids
from the indirect tension creep test. J Eng Mech 136(9):1067–1075
11. M-E PDG (2008) Mechanistic empirical pavement design guide: a manual practice. American
Association of State Highway and Transportation Officials, Washington, DC, USA
12. MoRTH (2013) Specification for road & bridge works, 5th rev. Ministry of Road Transport
and Highways, Indian Roads Congress
13. Roque R, Buttlar W (1992) The development of a measurement and analysis system to accu-
rately determine asphalt concrete properties using the indirect tensile mode. J Assoc Asphalt
Technol 61:304–332
14. Von Quintus HL, Rauhut JB, Kennedy TW (1982) Comparisons of asphalt concrete stiffness
as measured by various testing techniques. J Assoc Asphalt Paving Technol 51:35–52