90

Chapter-4

RESULTS AND DISCUSSIONS

 91

Chapter 4: Results and Discussions

S.No. Name of the Sub-Title Page

No.

4.1 General 92
4.2 Results and Discussion for the Experiments of Stage- 1 92
4.3 Results and Discussion for the Experiments of Stage- 2 on DMA 94
Investigations
4.4 Results and Discussion for Stage 3 - Investigations using fabricated 99
Rotational Viscometer
4.5 Results and Discussion for Stage 4 – Micro structural Investigations 103
4.5.1 Fractionation tests 103
4.5.2 SEM Analysis 104
4.5.3 X- Ray Diffractograms 104
4.5.4 TG-DSC Studies 110
4.6 Stage 5 - Experimentation: Performance of Mixes 122
4.6.1 Results and Discussion on Marshall Stability Tests for 122
DBM
4.6.2 Results and Discussion on Marshall Stability Tests for BC 127
4.6.3 Results and Discussion on Resilient Modulus Test 132
4.7 Results of stripping value Test 134
4.8 Summary 137
 92

CHAPTER 4
RESULTS AND DISCUSSIONS

4.1 General
The details of experimental studies taken up to fulfill research objectives were
mentioned in the previous chapter. The details of results and discussion are presented
in this Chapter.
4.2 Results and Discussion for the Experiments of Stage- 1
As detailed in Section 3.3.2, the straight run bitumen VG 30 has been considered to
investigate the influence of acid modification in the present study. A minimum of
three samples have been prepared for all the listed fundamental tests and the average
values have been presented in Table 4.1 for ready reference. The results on viscosity
have been mentioned separately. The values presented in Table 4.1 have also been
converted in the form of figures to find out the optimum PPA dosage. These graphs
are presented in Fig 4.1 for ready reference.
Table 4.1 Results of fundamental tests with varying dosage of PPA
Binder PPA Penetration Softening Ductility
Content (1/10 mm) Point (cm)
(w/w %) ( 0C)
VG 30 0 64.00 47.00 100+
0.5 44.00 50.25 82.67
1.0 36.00 53.25 67.47
1.5 34.00 55.50 51.97
2 31.00 59.00 41.00

65 a
60

55
Penetration (dmm)

50

45

40

35

30

0.0 0.5 1.0 1.5 2.0

PPA Content (%)
 93

60
b
58

Softening Point (Degrees Celsius)

56

54

52

50

48

46
0.0 0.5 1.0 1.5 2.0
PPA Content (%)

100 c
90
Ductility at 25 C (cm)

80
0

70

60

50

40

0.0 0.5 1.0 1.5 2.0

PPA Content (%)

Fig. 4.1 Effect of PPA Modification on VG 30 Straight run bitumen

(a) Penetration (b) Softening Point (c) Ductility
It could be observed from the Table 4.1 that the penetration values have reduced from
64 for VG 30 grade bitumen to a value 31 at 2% dosage of PPA and as a result,
bitumen has transformed into hard grade bitumen. The ductility value has also got
reduced from initial value of 100+cm to a value 41cm at 2% dosage of PPA.
Similarly, the softening point value has increased from 470C for VG 30 sample to
590C at 2% dosage for VG 30 sample modified with PPA.
Out of the three fundamental properties mentioned viz., Penetration, Ductility and
Softening point, the ductility value at 2% is recorded as 41cm. Dosages higher than
2% of PPA will eventually result in unacceptable low ductility value according to IS
73 - 2013 [2]. It is with this inference that the investigations beyond 2% dosage of
PPA were not carried out. Also, as detailed in the literature review chapter, the
optimum dosage of PPA required will depend on crude source
 94

[125,130,131,135,136,137], on the extent of stiffness required or the extent of Useful

Temperature Interval (UTI) desired to improve PG grade, without compromising with
the fundamental properties as per requirements of codes for paving grade. In view of
these, in the current research, a dosage of 2% was considered as optimum for other
investigations so that the maximum enhancement in VG 30 grade due to PPA
modification at high temperature can be studied and documented. The observations
made in the current research are matching with observations as suggested in [130,138,
139]. The properties of the modified bitumen at optimum PPA dosage (2%) vis a vis
the unmodified bitumen are summarized and presented in Table 4.2.
Table 4.2 Results of Fundamental Tests
Sample Penetration Soften Ductility PI Type
(dmm) ing (cm)
Point
(0C)
VG30 64 46 100+ -1.41 Sol
VG30RT 36 50 32 -1.88 Sol
VG30PPA2 31 59 41 -0.25 Sol
VG30PPA2RT 30 58 31 -0.51 Sol
PMB 40 55 62 100+ 1.67 Sol-Gel
Legend 1: VG30 - Straight run bitumen with viscosity grade 30, VG30RT-RTFOT aged VG 30, VG30PPA2 - PPA modified
bitumen at a dosage of 2% by weight and VG30PPA2RT - RTFOT aged PPA modified bitumen at 2% dosage
Legend 2: Sol PI<0; Gel PI>2; Sol-Gel 0 to 2; Paving grades -2<PI<2

The penetration index is worked out from the Eqn. 2.1 and has lead to interesting
observations. All samples have PI values in between -2 and 2 indicating that the
binders are suitable for paving purposes [1, 19,125]. The value of PI for VG30 binder
shows that it is of Sol type [1,125, 18]. The PI value has decreased due to short term
RTFOT ageing. It has increased due to PPA modification making the binder to move
towards the Sol-Gel type of bitumen which can be regarded as a desirable
phenomenon as these bitumen will exhibit resistance to temperature. The polymer
modified bitumen had positive PI value which has exhibited Sol-Gel type. Usually,
Sol type bitumen are temperature susceptible and where as gel type are more
temperature resistant.
4.3 Results and Discussion on DMA Investigations
As mentioned in Section 3.4.1, DMA was performed in accordance with ASTM D
6373 [179] using a Bohlin Dynamic Shear Rheometer at an oscillatory frequency of
 95

10 rad/sec. Based on the results from the fundamental tests, at an optimum dosage of
2%, dynamic mechanical analysis has been carried out to see the effect of acid
modification. The short term aged samples (RTFOT residues) were also subjected to
DMA investigations. The average results of three samples have been presented. Error
bars have been drawn using the parameter Standard error of mean.
A sample output from the DSR software is presented in Appendix K. Based on the
results obtained from dynamic shear rheometer tests, samples with notations VG30,
VG30RT, VG30PPA2, VG30PPA2RT & PMB 40 have shown the similar properties
shown by the corresponding Performance Graded (PG) bitumen viz. PG 64, PG 70,
PG 76, PG 76 and PG 76 respectively. Ideally, the PG graded bitumen are referred
with respect to the high and low temperature regimes at which they are expected to
perform satisfactorily. For this classification, the samples are supposed to be tested
both at high and low temperature regimes. However, in the present study, the
investigations are limited to high temperature regimes with DSR and as a result, only
one temperature has been mentioned along with PG grade specified above.
The complex modulus and phase angles observed for all the samples considered
during the present study are being summarized and presented in Table 4.3.
Table 4.3 DSR Test results
Sample Parameter Temperature (0C)
46 52 58 64 70 76 82
VG30 G*(kPa) 19.83 11.40 3.61 1.48 0.64 - -
δ (Deg.) 84.17 86 87.30 87.50 87.80 - -
G*/sin δ 19.93 11.43 3.61 1.48 0.64
VG30RT G* 111 40 14.90 5.90 2.56 1.18 -
δ 72.20 77.20 81.10 83.90 85.90 87.10 -
G*/sin δ 116.58 41.02 15.08 5.93 2.57 1.18 -
VG30PPA2 G* 97.40 36.40 18.20 8.22 3.59 1.68 0.876
δ 60.90 64 66 68.60 71.60 74.90 76.60
G*/sin δ 111.47 40.50 19.92 8.83 3.78 1.74 0.90
VG30PPA2RT G* 115 62.40 26.40 11.10 5.00 2.34 0.95
δ 60.30 62.40 65.50 68.90 72.60 76 79.40
G*/sin δ 132.39 70.41 29.01 11.90 5.24 2.41 0.97
PMB 40 G* 39 30.80 14.10 6.51 3.24 1.66 0.76
δ 64.60 65.60 65.80 67.40 69.10 71 73.40
G*/sin δ 43.17 33.82 15.46 7.05 3.47 1.76 0.79

It can be observed from the Table 4.3 that VG30 (Equivalent PG 64), when modified
with 2% PPA, has shown the behavior similar to that of PG 76. The G*/Sinδ, which
 96

exemplifies the rutting behavior of bituminous materials, has also been found and the
results of which are presented through Fig. 4.2. It can be seen from the figure that the
Complex Modulus (G*) values have increased while phase angle (δ) values have
decreased, consequently improvising the stiffness and rutting resistance values.
Though the PPA modified VG 30 at 2% dosage and PMB 40 bitumen have been
categorized as equivalent to the PG 76 grade bitumen, a closer observation will reveal
that PPA modified binders had higher complex modulus values for the entire
temperature range from 460C to 580C than that of polymer modified bitumen. Beyond
580C, these two modified binders had almost similar complex modulus values.

Fig. 4.2 Comparison of SHRP Rutting Parameter G*/Sinδ for binders

 97

The DSR test results of PPA modified bitumen and Polymer modified bitumen are
further compared with respect to the performance of VG 30 straight run bitumen with
the following formula.

Percentage Increase in Complex Modulus 

G *PPA/ PMB  G *VG30 x100 Eqn 4.1
G *VG30

Percentage decrease in Phase Angle 

 VG 30   PPA / PMB x100 Eqn 4.2
 VG 30
The improvement with regard to G* and δ values observed with respect to the above formulae
is depicted through Fig 4.3 and 4.4. These figures demonstrate the effectiveness of PPA
modified VG 30 binder vis-a-vis polymer modified bitumen in exhibiting stiffness and elastic
properties.

600

PPA PMB
500
% Increase in Complex Modulus

400

300

200

100

0
40 46 52 58 64 70 76

Temperature (C)

Fig. 4.3 Comparison of VG30PPA2 and PMB 40 bitumen in the improvement of

complex modulus with respect to VG 30 bitumen
 98

35
PPA PMB

30

% Decrease in Phase angle

25

20

15

10
40 46 52 58 64 70 76

Temperature (C)

Fig. 4.4 Comparison of VG30PPA2 and PMB 40 bitumen in the improvement of

phase angle with respect to VG 30 bitumen
Also, the Complex modulus, phase angle and G*/sin δ values have been plotted on
bar graphs for VG30, VG30PPA2 and PMB 40 samples and depicted through Fig 4.5
to 4.7 respectively. These figures demonstrate pictorially, the efficacy of PPA
modification over VG 30 as well as PMB 40.

100
VG30
VG30PPA2
PMB40
Complex Modulus (kPa)

10

40 46 52 58 64 70 76 82 88

Temperature (C)

Fig 4.5 Bar graph for comparison of G* values

 99

VG30
90 VG30PPA2
PMB40
85

80

Phase Angle (Degrees) 75

70

65

60

55

50

45

40
40 46 52 58 64 70 76 82 88

Temperature (C)

Fig 4.6 Bar graph for comparison of phase angle values

100 VG30
VG30PPA2
PMB40

10
G*/Sin (kPa)

1
fail

0.1
46 52 58 64 70 76 82 88

Temperature (C)

Fig 4.7 Bar graph for comparison of G*/ Sin δ values

4.4 Investigations using fabricated Rotational Viscometer
As per the existing grading system, out of 4 grades specified in the code of practice
[2], VG-30 grade has been recommended for use in India based on the prevailing
 100

environmental conditions [178]. For the purpose of grading the bitumen, many
agencies use capillary viscometer, where viscosity is measured based on the flow
under gravity, though this viscometer is not so versatile to capture dynamic and static
states of flow, due to its availability at cheaper price.

A detailed discussion on fabrication of the equipment, drawings, photographs of

working model, methodology to measure viscosity and calibration was presented in
the previous section. A provisional application for a domestic patent has been filed
with the patent title “Measurement of Viscosity of bituminous binders in
rotational mode through indirect measurement of torque”. The acknowledgement
of the patent is presented in Appendix L and the Prior Art Search report is presented
in the Appendix M.

The test results obtained for VG 30 grade bitumen covering the range of temperatures
from 80 to 1500C is presented here so that the property of consistency can be
understood at varying temperatures. The results obtained have been summarized and
presented in Table 4.4 for ready reference. In this Table, first four columns indicate
the observed / set variables which include: The temperature of the test, voltage, stable
angular speed and stable current. Variables in the remaining four columns indicate the
estimated variables. The torque is estimated from the equations developed from the
measured current, speed and the viscosity is computed using Eqn. 4.11.

Table 4.4 Test results on VG 30 bitumen using fabricated rotational viscometer

Stable Stable Mean Shear Strain Viscosity
Temperature Voltage speed current torque stress rate (Pa.S)
(0C) (V) (rpm) (A) (N cm) (Pa) (rad/sec)
80 6 280 0.31 5.83 4680.80 233.87 20.01
90 6 307 0.28 2.34 1881.24 256.42 7.34
100 9 450 0.26 1.87 1504.73 375.87 4.00
120 12 620 0.14 0.87 696.53 517.86 1.35
135 12 622 0.12 0.66 526.49 518.70 1.02
150 12 935 0.12 0.69 554.21 780.97 0.71
A graph between temperature and viscosity has been prepared and is shown in Fig 4.8
for VG 30 bitumen and an exponential relation was observed to have good R 2 value.

A computational GUI based tool using C# is developed where the user can input the
data in the window and will be able to calculate the viscosity. The source code for C#
 101

is presented in Appendix N. A screen shot while running the program and after
getting the output is presented in Fig 4.9. This tool becomes quite handy after
summarizing the data of the experiment.

20 VG 30 Bitumen
18

16 -x/t1
y = y0+ Ae
Dynamic Viscosity (PaS) 14
2
12
R =0.996
y0 = 0.9721
10
A= 69664.23
8
t1= 9.74555
6

0
80 100 120 140
0
Temperature ( C)

Fig 4.8 Temperature Vs Viscosity Plot for VG 30 Binder

The experimentation on the fabricated viscometer is continued further on VG30,

VG30RT, VG30PPA2, VG30PPA2RT samples to investigate for the adequacy in
workability at 1350C for adopting PPA modification in actual construction practice
and these results have been compared with that of Brookfield viscometer. The test
results are presented through Table 4.5.

Table 4.5 Comparison of test results of viscosity using fabricated (Spindle 1) and
Brookfield Rotational Viscometer
Sample Code Voltage Stable Stable Mean Viscosity @ Viscosity @
(V) (RPM) Current Torque 1350C using 1350C using
(amp) (Ncm) Fabricated Brookfield
viscometer (DV II Pro)
viscometer
VG30 12 619 0.15 1.17 1.42 1.04
VG30RT 12 617 0.17 1.60 2.49 2.38
VG30PPA2 6 310 0.15 0.78 2.41 2.27
VG30PPA2RT 6 309 0.14 0.81 2.51 2.65
PMB 40 12 615 0.15 1.17 1.82 1.87
 102

Fig. 4.9 Sample output

utput of the C# Programming tool for computation of viscosity
using fabricated rotational viscometer
 103

4.5 Stage 4 – Discussion on Micro structural investigations

4.5.1 Discussion on Fractionation tests
The results of fractionation tests are summarized in Table 4.6 and depicted through
Fig 4.10 for varying dosages of PPA. It could be observed from the table that the solid
fraction of bitumen, asphaltenes, has increased substantially due to ageing and
modification. Corresponding to a dosage of 2% of PPA, the asphaltenes content was
observed to be 28.1 % (w/w %) while it was 14.4% (w/w %) for unmodified VG 30
bitumen, recording an increase of 13.7 % asphaltene content.

Table 4.6 Results of Fractionation of bitumen

Sample Asphaltenes Maltenes
(%) (%)

VG30 14.4 85.6

VG30RT 19.3 80.7
VG30PPA2 28.1 71.9
VG30PPA2 RT 33.2 66.8
PMB 40 13.2 86.8
Maltenes Asphaltenes
100
e
80

75.3 71.9
60 82.5 79.1
85.6

40

20
24.7 28.1
17.5 20.9
14.4
0
0.0 0.5 1.0 1.5 2.0
PPA Content (%)

Fig 4.10 Fractions of bitumen with varying dosages of PPA

The recovered asphaltenes were subjected to three types of micro structural
investigations viz. Scanning Electron Microscopy, X-Ray Diffraction Analysis and
 104

Thermo gravimetric and Differential Scanning Calorimetry (TG-DSC) to understand

micro structural changes due to aging and modification. The details of these
investigations are presented below.

4.5.2 Discussions on SEM Analysis

The micro structure of asphaltenes has been investigated using Scanning Electron
Microscope. Since the raw samples cannot be used directly, it requires the elimination
of oil phase from the bitumen and as such, several procedures have been suggested in
the literature [107]. The procedure adopted in the present study has been detailed in
Section 3.6.3.1. The images were presented in Fig 4.11 with 100X, 1000X and 5000X
magnification levels for asphaltenes extracted from straight run bitumen (VG 30) and
PPA modified (1% and 2%) bitumen. These images reveal the changed morphology
with PPA modification. However, all the SEM images indicate that the asphaltenes
extracted from both modified and unmodified bitumen are amorphous in nature. It can
be noted here that the crystalline nature is not revealed by SEM images and hence the
investigations are carried forward with X-Ray Diffraction studies, the details of which
are presented below.

4.5.3 Discussions on X-Ray Diffractograms

The micro structure of the extracted asphaltenes (from unmodified, PPA modified,
unmodified aged & modified aged) has been investigated using X-Ray Diffraction
(XRD) technique. Earlier, limited studies have been reported by researchers [88-90]
on micro structure of asphaltenes using XRD technique and the patterns of
diffractograms for asphaltenes have been published. The process followed in
obtaining the diffractograms has been presented in Section 3.6.4.
It has been reported by the earlier studies that there are four peaks that can be
observed in XRD patterns of petroleum asphaltenes in particular [88,89]. The first is γ
band which arises from scattered X-rays by aliphatic chains or condensed saturated
rings [88,89]. The second peak, a graphene band, comes from the diffraction of X-
Rays from the stacks of aromatic molecules. The remaining two peaks are observed to
 105

(a) (b) (c)

4.11.1. Images for asphaltenes of VG 30

(a) (b) (c)

4.11.2 Images for asphaltenes of VG 30 with 1% PPA

(a) (b) (c)

4.11.3 Images for asphaltenes of VG30 with 2% PPA
Fig 4.11 SEM Images of Asphaltnes for
(a) 100x (b) 1000x (c) 5000x magnification levels

be (10) and (11) bands, which are resulted due to the reflections in X-ray pattern from
the in plane structure of aromatics.
A theoretical representation of the band peaks, as suggested by Mohammed Nahid
Siddiqui et al and Ebert L B et al [88,89], is presented in Fig. 4.12.
In the present study, the diffractograms have been obtained for asphaltenes extracted
from VG30, VG30RT, VG30PPA2 and VG30PPA2RT bitumen through the XRD
studies. VG30PPA1 has not been investigated for XRD studies as 1% PPA was found
to be a non-optimum solution.
The diffractograms have been presented through Fig 4.13 ((a) to (d)) for asphaltenes
of VG30, VG30RT, VG30PPA2 and VG30PPART respectively.
 106

γ Graphene (002)
(a)

Intensity

(10) (11)

(b)
2θ

[010] Y Y Y
(10) (11)
(002)

[100]
X X X
[001] Z
Z Z

(c)

Fig 4.12. Bands (a) and Planes (b) of X-Ray Diffractogram (c) cross section of
asphaltene model
Source: M.N.Siddique et al [17]
 107

900
800
700
600
500
Count

400
300
200
100
0
0 10 20 30 40 50 60 70 80 90 100
2

Fig 4.13 (a) Diffractograms of asphaltenes of VG 30 bitumen

1200

1000

800

600
Count

400

200

0 10 20 30 40 50 60 70 80 90 100
2

Fig 4.13 (b) Diffractograms of asphaltenes of VG30RT bitumen

 108

1200

1000

800
Count

600

400

200

0 10 20 30 40 50 60 70 80 90 100
2
Fig 4.13 (c) Diffractograms of asphaltenes of VG30PPA2 bitumen

1200

1000

800

600
Count

400

200

0 10 20 30 40 50 60 70 80 90 100
2
Fig 4.13 (d) Diffractograms of asphaltenes of VG30PPA2RT bitumen
 109

It can be observed from the diffractogram of asphaltenes of VG 30 bitumen that all

the four peaks detailed above were exhibited indicating that they depict semi
crystalline structure. However, for asphaltenes extracted from aged and modified
bitumen, no specific peaks could be observed (Fig. 4.13 (d)) and hence the
computation of crystalline parameters was not carried out.
The XRD data has been analyzed with Origin Software using quick peak gadget for
parameters peak X (2θ), peak Y (count). Area under the peak and full width at half
maximum has been found to assess the crystalline parameters for each case. The
parameters analyzed include: the layer distance between aromatic sheets (dm), the
distance between the saturated portions of the molecules (dγ), the average diameter of
aromatic sheets (La) on the basis of Scherrer formula, the average height of stack of
aromatic sheets perpendicular to the plane of sheet (Lc) and the number of aromatic
sheets in a stacked cluster (M) as presented in M.N.Siddiqui et al [Fig 3 of [83]]. The
formulae used to compute the crystalline parameters are:
 Eqn. 4.3
dm  sin 
2
5 Eqn. 4.4
d  sin 
8
1 . 84  0 .92 Eqn. 4.5
La  cos  
w B1 / 2

0 .45 Eqn. 4.6

L c  0 . 9 w cos  
B1 / 2

Lc Eqn. 4.7
M 1
dm
where λ is the wave length of Cu Kα radiation and θ is the Bragg’s Angle of

incidence. B1/2 is full width at half maximum (FWHM) (for La, (11) band and for Lc,

grapheme band) and w is the band width.

The procedure adopted to find the crystalline parameters is described in Appendix O
for complete understanding.
The parameters viz. dm, dγ, Lc and M are computed for all the peaks as detailed in
Appendix O and a summary of these parameters is presented in Table 4.13 for
asphaltenes extracted from the different binders considered for XRD studies. Further,
 110

as suggested in literature [88-89], the Parameters dm, Lc with the respective peak as
Graphene, dγ from Gamma Band and the parameter La, found from FWHM of (11)
band, have been extracted from Table 4.8 and presented in Table 4.7.

It could be observed that the crystalline parameters; the average diameter of aromatic

sheets La, the average height of aromatic sheets Lc and the number of aromatic sheets

M have shown significant change due to aging and modification when compared to

the parameters of straight run bitumen. Thus it can be concluded that the aging and

modification brings clear changes in crystalline structure of asphaltenes with almost

similar impact.

Table 4.7 Summary of Crystalline Parameters of Asphaltenes

Asphaltenes dm dγ La Lc M
(Graphene) (Gamma) (11) (Graphene)
(computed from
Eqn. 4.7)
VG30 3.26 4.10 11.19 5.09 2.56
VG30RT 3.30 4.09 - 10.28 4.12
VG30PPA2 3.31 4.08 - 8.86 3.68
VG30PPA2RT - - - - -

4.5.4 Discussions on TG-DSC Studies

This study is carried out to investigate the thermal stability of bitumen and its solid
fraction; asphaltenes and also for the possible changes in glass transition, crystallanity
and melting phases. The source data of the experiments for bitumen and asphaltenes
were presented in Appendix P to W.
From the point of view of Pavement Engineering, the transitions of bitumen for
temperatures below the expected lowest freezing temperature of pavement and
beyond 2000C is of no relevance. The Superpave specifications do not consider any of
these parameters pertaining to the transitions. However, these investigations are taken
up for investigative purposes to understand the decomposition patterns of bitumen,
asphaltenes and their thermal transitions. This study could probable to lead towards
linking the decomposition patters of bitumen, asphaltenes and thermal transitions with
rheology of bitumen at varying temperatures.
 111

Table 4.8 Peak-wise crystalline parameters of Asphaltenes

Sample Code Peak Peak X Peak Y Area FWHM dm dγ La Lc M

Name
VG 30 Gamma 21.64 829 16.1562 0.0468 4.10 5.13 19.65 9.61 3.34
Graphene 27.32 474 10.8809 0.0884 3.26 4.08 10.41 5.09 2.56
(10) 39.61 538 33.7771 0.0852 2.27 2.84 10.80 5.28 3.32
(11) 50.41 381 11.3356 0.0822 1.81 2.26 11.19 5.47 4.03
VG30RT Gamma 21.70 1202 47.8678 0.2200 4.09 5.12 4.18 2.05 1.50
Graphene 26.96 944 12.0143 0.0438 3.30 4.13 21.02 10.28 4.12
(10) 39.67 603 239.7200 0.1497 2.27 2.84 6.15 3.01 2.32
VG30PPA2 Gamma 21.74 1182 28.5976 0.1009 4.09 5.11 9.12 4.46 2.09
Graphene 26.92 913 24.4208 0.0508 3.31 4.14 18.11 8.86 3.68
VG30PPA2RT Gamma 17.23 1080 2.0096 0.0188 5.14 6.43 49.07 24.00 5.67
 112

The experiments were carried out in the range from room temperature to 1400 0C with
a heating rate of 100C/min. Thermo gravimetric analysis will help in measuring the
mass loss with heating while the differential scanning calorimetry will help in
understanding the thermal transitions that occur at changing temperatures.

A brief description about the DSC experimentation and associated thermal transitions
were presented through Appendix X. Out of the three transitions discussed in
Appendix X, crystallization and melting are considered to be first order transitions
while the glass transition is considered as a second order transition. Even though
theoretical postulations have been presented about the three important events in a
thermogram, not all materials exhibit similar trends in reality. Often these transitions
are very complex for some materials like bitumen. In the present research, the
following samples have been considered for the TG-DSC analysis. Appropriate codes
were assigned to each of these samples for clarity and are presented in Table 4.9.

Table 4.9 Sample Coding for TG-DSC Analysis

Sample Sample Abbreviation Description
Code
A VG30 Straight run Bitumen
B VG30RT RTFOT aged bitumen
Bitumen
C VG30PPA2 PPA modified bitumen
D VG30PPA2RT RTFOT aged PPA modified bitumen
E VG30 Asphaltenes of Straight run Bitumen
F VG30RT Asphaltenes of RTFOT aged bitumen
G Asphaltenes VG30PPA2 Asphaltenes of PPA modified bitumen
H VG30PPA2RT Asphaltenes of RTFOT aged PPA
modified bitumen

The results of TG-DSC for each of these sample codes are presented through figures
Fig. 4.14 to 4.21. Figure (a) describes the mass loss behaviour while Figure (b)
describes thermal transitions in DSC plot for all the samples listed. The observations
for each of these samples with associated temperature of transition are presented in
Table 4.10.
 113

Table 4.10 TG-DSC Observations

Sample Magnitude of Sample Magnitude of
(Bitumen) maximum mass loss (Asphaltenes) maximum mass loss
and and
correspondingTemper correspondingTemper
ature ature
A 78.95% at 463.70C E 43.92 % at 457.20C
B 79.28 % at 462.50C F 38.54 % at 459.20C
C 73.93 % at 467.10C G 44.14 % at 457.50C
D 74.61 % at 465.20C H 38.41 % at 457.90C

The following observations were made with reference to TG-DSC studies.

 For all bitumen samples and asphaltenes, no base line (Refer Fig G-3) could
be observed in the test range of temperatures and hence the glass transition is
expected below room temperature.
 The PPA modified bitumen with sample code VG30PPA2 is observed to have
higher thermal stability (73.93 % mass loss at 467.10C) than VG 30 bitumen
(78.95% mass loss at 463.70C) and similar kind of behaviour was observed
from DMA studies, wherein the modulus values of modified bitumen are
observed to be better than straight run bitumen.
 No transitions are observed for samples B and D (below 2000C) whereas for
samples A and C, endothermic transitions are observed with peaks at 109.50C
and 1000C respectively for VG30 and PPA modified VG30 bitumen. This
results in increase in heat capacity consequently resulting in intensification of
ageing process at the above mentioned temperatures.
 No transitions were observed (below 2000C) for the asphaltenes except for the
asphaltenes of VG 30 for which an endothermic transition was noticed at
125.5 0 C with insignificant mass loss as observed from TG curves.
 In the temperature zone of Short (1630C) and Long term (1100C) ageing
experiments of Superpave using RTFOT & PAV, the bitumen is observed to
exhibit high thermal stability without any significant mass loss.
 114

1.5

Exo
TG
100
1.0 DSC

0.5
80
DSC (mW/mg)

0.0

TG (Mass %)
60
-0.5

-1.0
40

-1.5

20
-2.0

0 200 400 600 800 1000 1200 1400 1600

Temperature (Degrees Celsius)

Fig 4.14 (a) TG-DSC Thermogram for VG 30 Bitumen

Fig 4.14 (b) DSC Thermogram for VG 30 Bitumen

 115

1.5

Exo
TG
1.0
100
DSC
0.5
0.0
-0.5 80
DSC (mW/mg)

-1.0
-1.5
60
-2.0
-2.5
-3.0 40
-3.5
-4.0
20
-4.5
-5.0
0 200 400 600 800 1000 1200 1400
Temperature (Degrees Celsius)

Fig 4.15 (a) TG-DSC Thermogram for RTFOT aged VG 30 Bitumen

Fig 4.15 (b) DSC Thermogram for RTFOT aged VG 30 Bitumen

 116

1.5
Exo TG
1.0 100
DSC
0.5
80
0.0
DSC (mW/mg)

-0.5
60
-1.0

-1.5
40
-2.0

-2.5 20

-3.0
0 200 400 600 800 1000 1200 1400
Temperature (Degrees Celsius)

Fig 4.16 (a) TG-DSC Thermogram for PPA modified VG 30 Bitumen

Fig 4.16 (b) DSC Thermogram for PPA modified VG 30 Bitumen

 117

3.5
Exo TG
3.0 100
DSC
2.5

2.0 80
DSC (mW/mg)

1.5

TG (Mass %)
1.0 60

0.5

0.0 40

-0.5

-1.0 20

-1.5
0 200 400 600 800 1000 1200 1400
Temperature (Degrees Celsius)

Fig 4.17 (a) TG-DSC Thermogram for RTFOT aged PPA modified VG 30
Bitumen

Fig 4.17 (b) DSC Thermogram for RTFOT aged PPA modified VG 30 Bitumen
 118

1.5

Exo
TG
1.0
DSC 100
0.5
0.0
-0.5
DSC (mW/mg)

-1.0

TG (Mass %)
80
-1.5
-2.0
-2.5
-3.0
60
-3.5
-4.0
-4.5
-5.0 40
0 200 400 600 800 1000 1200 1400
Temperature (Degrees Celsius)

Fig 4.18 (a) TG-DSC Thermogram for asphaltenes of VG 30 Bitumen

Fig 4.18 (b) DSC Thermogram for asphaltenes of VG 30 Bitumen

 119

110

Exo
TG
4 DSC
100

2
DSC (mW/mg)

90
0

80
-2

-4 70

-6
60
0 200 400 600 800 1000 1200 1400
Temperature (Degrees Celsius)

Fig 4.19 (a) TG-DSC Thermogram for asphaltenes of RTFOT aged VG 30

Bitumen

Fig 4.19 (b) DSC Thermogram for asphaltenes of RTFOT aged VG 30 Bitumen
 120

110
Exo
TG
2 DSC
100

90
DSC (mW/mg)

TG (Mass %)
80

70
-2

60

-4 50

0 200 400 600 800 1000 1200 1400

Temperature (Degrees Celsius)
Fig 4.20 (a) TG-DSC Thermogram for asphaltenes of PPA modified VG 30
Bitumen

Fig 4.20 (b) DSC Thermogram for asphaltenes of PPA modified VG 30 Bitumen
 121

Exo
TG
100
2 DSC

90
0
DSC (mW/mg)

TG (Mass %)
80
-2

70
-4

60

-6

50
0 200 400 600 800 1000 1200 1400
Temperature (Degrees Celsius)

Fig 4.21 (a) TG-DSC Thermogram for asphaltenes of RTFOT aged PPA
modified VG 30 Bitumen

Fig 4.21 (b) DSC Thermogram for asphaltenes of RTFOT aged PPA modified
VG 30 Bitumen
 122

4.6 Results and Discussion on Marshall Stability Tests

In lieu of the above discussions made in Section 3.7.2, the efficacy of modification
needs to be tested with respect to Marshall’s stability values. It is in this direction, that
the current study is extended by investigating Marshall’s parameters for 2 types of
mixes viz. Bituminous Concrete (BC) and Dense Bituminous Macadam (DBM), the
details of which are presented in the next article.
4.6.1 Results of Marshall’s test for DBM
The dense bituminous macadam (DBM) layer is regarded as binder course in flexible
pavements. The investigations are confined to Grade 2 of DBM as suggested by
MoRTH [162]. The details of gradation chosen as discussed in the section 3.7.1.5.3
are presented in Table 4.11 for reference. The selected gradation with Nominal
Maximum Particle Size, 25mm (NMPS) with superpave control points with restricted
zone is depicted in Fig 4.22.
The specific gravities of materials used in the Marshall Stability test have been found
in accordance with IS 2386 (Part III) [180]. The test results on aggregate were
presented through Table 4.12 & 4.13. The results indicate the aggregates from local
lime stone quarry are suitable for mixes. The specific gravity of mineral filler and
bitumen is found to be equal to 2.42 and 0.97 respectively. The proportion of mineral
filler is taken as 2% and has fulfilled grading requirements of MoRTH.

Table 4.11 Gradation of DBM Layer for Grade 2 with Thickness 50-75 mm
Sieve Size Cumulative % by Sieve Size Cumulative % by
(mm) weight of total (mm) weight of total
aggregate passing aggregate passing
Range Selected Range Selected
Suggested (% finer) Suggested (% finer)
(% finer) (% finer)
37.5 100 100 2.36 28-42 40
26.5 90-100 95 1.18 - -
19 71-95 83 0.6 - -
13.2 56-80 68 0.3 7-21 14
9.5 - - 0.15 - -
4.75 38-54 46 0.075 2-8 5
Binder Content Min 4.5%
 123

100

80

Percent Passing (%)

60

40
MoRTH Lower
MoRTH Upper
20 Max. Densilty Line
Gradation Adopted
Superpave
0 Restricted Zone

0 1 2 3 4 5
Sieve Size Raised to 0.45

Fig 4.22 Selected gradation for DBM with maximum density line and Superpave
control points
Table 4.12 Result of Tests on Aggregates – Specific Gravity and Water
absorption (IS 2386 Part III)

Aggregate size Bulk specific Water

(mm) gravity absorption
37.5 - 26.5 2.646 0.16
26.5 - 19.0 2.644 0.19
19.0 - 13.2 2.643 0.22
13.2 - 4.75 2.640 0.23
4.75 - 2.36 2.657 0.34
2.36 - 0.30 2.660 0.49
0.30 - 0.075 2.652 1.06
0.075mm 2.639 1.78
passing
Acceptability 2.5 – 3.0 <2%
criteria [162]

Table 4.13 Result of Tests on Aggregates – Strength and Shape Tests

(IS 2386 Part I &IV)
Name of the Test Result Acceptability
Criteria
Aggregate Impact Value (%) 12.25 <30%
Los Angeles abrasion Value (%) 13.56 <30%
Aggregate Crushing Value (%) 14.52 <30%
Flakiness Index (%) 12.26 Combined
Elongation Index (%) 13.82 Value <30%
 124

The required quantity of aggregates and filler based on trial binder content is heated
to a temperature of 175-1900C. Bitumen is simultaneously heated to a temperature of
135-1500C. The heated aggregates and bitumen are thoroughly mixed at a temperature
of 154 -1600C. The mix is placed in a preheated mould (1400C) and compacted by a
rammer with 75 blows on either side at temperature of 1380C to 1490C. The prepared
mould is loaded in the Marshall Stability test setup after immersing the extracted
sample for a period of 30-45minutes as per the test protocol ASTM D 6927-15 [163].

The following properties have been worked out as per the procedure [163].

a) Theoretical specific gravity ,Gt

b) The bulk specific gravity of the mix ,Gm
c) Percent air voids ,Vv
d) Percent volume of bitumen ,Vb
e) Percent void in mixed aggregate ,VMA
f) Percent voids filled with bitumen ,VFB.
The parameters have been worked out for various trial binder contents and
summarized in Table 4.14.
Table 4.14 Summary of Marshall’s Stability Test Results

Binder Stability Flow Vv Vb VMA VFB Gm

Content Value (mm) (%) (%) (%) (%)
(%) ( KN )
4.0 11.21 1.40 5.31 10.19 16.81 60.63 2.43
4.5 14.64 2.70 4.74 11.05 15.80 69.98 2.44
5.0 15.85 4.60 3.10 12.39 15.49 80.05 2.46
5.5 8.48 6.20 1.54 13.73 15.27 90.22 2.45
6.0 7.92 7.90 1.17 14.92 16.08 92.79 2.44

For each parameter listed above, the average value of three samples for a trail dosage
is computed and presented. The graphical plots for the above parameters have been
prepared and presented in Fig. 4.23 and from the guidelines, the optimum binder
content for each of the parameter is worked out. By considering the specifications
given by MoRTH [162], the value of optimum binder content is found to be 4.75 % as
can be observed from the Table 4.15.
 125

Table 4.15 Calculation of OBC for DBM

Sl.No. Parameter Binder content
required
(%)
1 Volume of voids i.e. @4% Vv 4.75
2 Voids filled in bitumen i.e. @70% VFB 4.5
3 Maximum stability 15.85 kN 5
Optimum Binder Content – Average 4.75

16 Stability

14

12
Stability (kN)

10

3.5 4.0 4.5 5.0 5.5 6.0 6.5

Binder Content (%)

(a) Binder Content Vs Stability

5
Flow (mm)

0
3.5 4.0 4.5 5.0 5.5 6.0 6.5
Binder Content (%)

(b) Binder Content Vs Flow

 126

5
Volume of Voids (%)

0
3.5 4.0 4.5 5.0 5.5 6.0 6.5
Binder Content (%)

(c) Binder Content Vs Volume of Voids (Vv)

100

95

90
Voids filled with Bitumen (%)

85

80

75

70

65

60

55
3.5 4.0 4.5 5.0 5.5 6.0 6.5
Binder Content (%)

(d) Binder Content Vs VFB

 127

2.470

2.465

2.460

2.455

Bulk Specific Gravity

2.450

2.445

2.440

2.435

2.430

2.425

2.420
3.5 4.0 4.5 5.0 5.5 6.0 6.5
Binder Content (%)

Fig 4.23 (e) Binder Content Vs Bulk Specific Gravity (Gm)

Fig. 4.23 Results of Marshall’s Test for DBM
4.6.2 Results of Marshall Stability test for BC
The bituminous concrete (BC) layer is regarded as wearing course in flexible
pavements. The investigations are confined to Grade 2 of BC as suggested by
MoRTH. The details of gradation chosen as discussed in section 4.7.1.6.2 is presented
in Table 4.16 for reference and a graph showing superpave control points for Nominal
Maximum Particle Size (NMPS) 12.5mm for BC Grade 2 is presented in Fig. 4.24.

Table 4.16 Gradation of BC Layer for Grade 2 with Thickness 30-45mm

Sieve Cumulative % by weight Sieve Cumulative % by weight
Size of total aggregate passing Size of total aggregate
(mm) (mm) passing
Range Selected Range Selected
Suggested (% finer) Suggested (% finer)
(% finer) (% finer)
19 100 100 1.18 34-48 41
13.2 79-100 93 0.6 26-38 30
9.5 70-88 81 0.3 18-28 22
4.75 53-71 60 0.15 12-20 16
2.36 42-58 51 0.075 4-10 8
Binder 5 – 7%
Content
 128

The average of three samples of the above properties are determined for each mix
with varying bitumen content. The results are presented through Table 4.17 and Fig
4.25.
Table 4.17 Summary of Marshall’s Stability Test Results

Binder Stability Flow Vv Vb VMA VFB Gm

Content Value (mm) (%) (%) (%) (%)
(%) ( KN )
4.0 12.52 2.56 6.15 9.43 15.58 60.51 2.33
4.5 15.00 3.45 5.74 10.66 16.41 64.99 2.36
5.0 15.40 4.25 5.19 11.83 17.02 69.49 2.37
5.5 13.21 7.73 4.60 11.80 16.40 71.95 2.36
6.0 12.33 7.95 3.91 14.96 18.87 79.28 2.35

100

80
Percent Passing (%)

60

40

Max. Density Line

20 MoRTH Lower
MoRTH Upper
Gradation Adopted
Superpave
0 Restricted Zone

0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0

Sieve Size raised to 0.45

Fig 4.24 Selected gradation superimposed over maximum density line and
Superpave control points

For each parameter listed above, the average value three samples for a trail dosage is
computed and presented through graphical plots in Fig. 4.45 and from the guidelines
the optimum binder content for each of the parameter is worked out. By considering
 129

the specifications given by MoRTH, the value of optimum binder content is

calculated and presented in Table 4.18.

Table 4.18 Optimum binder content for BC

Sl.No. Parameter Binder

content
required
(%)
1 Volume of voids i.e. @ 4.5% Vv 6
2 Voids filled in bitumen i.e. @70% VFB 5.5
3 Maximum stability 15.4 kN 4.8
Optimum Binder Content – Average 5.40

After finding the OBC for both DBM and BC mixes, three more samples in each of
the categories were cast with the OBC and the samples were tested and it was found
that the OBC and other parameters were almost matching with the earlier values
found at OBC.

20
Stability
18

16
Stability (kN)

14

12

10

8
3.5 4.0 4.5 5.0 5.5 6.0 6.5
Binder Content (%)

Fig 4.25 (a) Binder Content Vs Stability

 130

8 Flow

Flow (mm)
5

2
3.5 4.0 4.5 5.0 5.5 6.0 6.5
Binder Content (%)

Fig 4.25 (b) Binder Content Vs Flow

Volume of Voids

6
Volume of Voids (%)

3
3.5 4.0 4.5 5.0 5.5 6.0 6.5
Binder Content (%)

Fig 4.25 (c) Binder Content Vs Volume of Voids

 131

80

75

70
VFB

65

60

3.5 4.0 4.5 5.0 5.5 6.0 6.5

Binder Content (%)

Fig 4.25 (d) Binder Content Vs VFB

2.390
2.385
2.380
2.375
2.370
2.365
2.360
2.355
Gm

2.350
2.345
2.340
2.335
2.330
2.325
2.320
3.5 4.0 4.5 5.0 5.5 6.0 6.5
Binder Content (%)

Fig 4.25 (e) Binder content Vs Bulk specififc gravity (Gm)

Fig 4.25 Results of Marshall’s Test for BC
 132

It can be clearly observed that the Marshall’s Stability values are passing the MoRTH
requirements for both DBM and BC with PPA modified binder at 2% dosage. This
can be attributed to stiffness enhancement of the VG 30 binder due to modification at
2% dosage. DBM mix has recorded a stability value of 15.85 kN while it is 15.40 kN
for BC. The other requirements viz. flow, Vv and VMA and VFB have also been
checked at the optimum binder content of 4.75 % for DBM and 5.40 % for BC.
Resilient Modulus has been the defining parameter to qualify bituminous concrete for
its applications. In the present study, resilient modulus has been found in the
laboratory with repeated load test. (GMR Pavement Research Centre, Shamshabad,
Hyderabad). In order to determine the resilient modulus of the mixes prepared by
PPA modified binder, 9 numbers of samples were cast at the OBC for each layer and
3 of them in each group were used to find the indirect tensile strength and remaining
six samples were used for resilient modulus test, three each at 250 C and 350 C. As per
the codal provisions, 10 to 15% of the Indirect Tensile strength has been adopted as
the magnitude of peak to peak load in Resilient Modulus Test.

4.6.3 Results and Discussion on Resilient Modulus Test

The resilient modulus test was conducted at 1500N peak load for DBM samples and
BC layer, at 25 and 350C temperatures. The test was conducted using repeated load
equipment with cyclic loading for an estimated poisson’s ratio as per ASTM D 4123-
82 with loading pulse width of 0.1sec and 0.9sec rest period. A seating of force of 10
% of peak load was applied in accordance with AASHTO TP 31 [181]. The sample is
conditioned for 50 cycles and the data from subsequent 5 cycles is recorded to
determine the resilient modulus. The expression used to calculate Mr is given below.

P (0.27  )
Mr  Eqn. 4.24
t h

Where Mr - Resilient Modulus (MPa)

P – Peak load intensity in N
µ - Poisson’s ratio
 h  sum of horizontal deformations measured from two LVDTs (microns)
 133

The raw data of the resilient modulus tests results are presented in Appendix from Y
to AF. The test results of DBM and BC mixes have been presented in Table 4.19.
Here, the minimum, maximum and representative resilient modulus values have been
tabulated.

Table 4.19 Results of Resilient Modulus

Temperature Mix Binder Resilient Modulus Values (MPa)
(0C) Minimum Maximum Representative
25 DBM VG30 3002 3100 3100
VG30+PPA2 4021 4413 4021
35 DBM VG30 1682 1930 1806
VG30+PPA2 2036 2289 2289
25 BC VG30 3038 3069 3054
VG30+PPA2 3769 4006 3859
35 BC VG30 1651 1773 1712
VG30+PPA2 2161 2321 2161

It can be observed that the resilient modulus values have also been enhanced due to
PPA modification at 2% dosage.
The percentage enhancement in resilient modulus values of DBM and BC mixes
prepared with PPA modified binder with respect to VG 30 bitumen and recommended
by IRC [182] were calculated and summarized in Table 4.20 for ready reference.
Table 4.20 Comparison of Resilient Modulus values
Mix Binder Representative Enhancement
type Resilient Modulus w.r.t. VG 30
(MPa) (%)
(found in the
present work)
250C 350C 250C 350C

DBM VG30 3100 1806 - -

DBM VG30+ 4021 2289 29.71 26.74
PPA 2%
BC VG30 3054 1712 - -
BC VG30+ 3859 2161 26.38 26.23
PPA 2%

It is also observed that the resilient modulus values at 350C for mixes prepared with
2% PPA were enhanced by 26.74 % and 26.23 % respectively for DBM and BC.
 134

Also, the enhancement in resilient modulus values at 250C for mixes prepared with
2% PPA were 29.71 % and 26.38 % respectively for DBM and BC.
It can be observed that the resilient modulus value for DBM and BC at 250C is
recommended as 3000 MPa while the same value at 350C is recommended as 1700
MPa [182]. By examination of experimental results, it can be observed that the values
are higher than recommended values. Though the codal provision has the same value
of Resilient Modulus at a given temperature for both DBM and BC, it was observed
that the DBM at a given temperature is performing slightly better than BC at the same
temperature. This trend is common both unmodified VG 30 sample as well as PPA
modified VG 30 samples. The corresponding numerical differences of resilient
modulus values of DBM and BC; with and without PPA modification are summarized
and presented in Table 4.21 for ready reference.

Table 4.21 Comparison of Resilient Modulus values w.r.t. DBM

Sample and Condition Resilient Modulus Percentage
(MPa) Difference by
which DBM is
DBM BC above BC

@ 250C VG 30 3100 3054 1.50

@ 250C VG 30 + 2% PPA 4021 3859 4.03
@ 350C VG 30 1806 1712 5.20
@ 350C VG 30 + 2% PPA 2289 2161 5.59

Improved performance of DBM in lieu with BC at higher temperatures can be

attributed to lesser bitumen content in DBM which gets affected with higher
temperatures.

4.7 Results of Stripping value Test

This test does not give any quantitative measure. However, the visual inspection of
the aggregates will demonstrate the extent of stripping taking place for the aggregates.
Photographs showing the extent of stripping for the following cases were presented
through the Table 4.22 and 4.23 for ready reference.
 135

Table 4.22 Stripping test on aggregates – Stage 1

Binder Without stripping agent after With stripping agent after

0
immersing in water at 40 C for immersing in water at 400C for
one day one day
VG 30

VG 30 + PPA
2%
 136

Table 4.23 Stripping test on aggregates – Stage 2

Binde Without stripping agent after With stripping agent after

r boiling test boiling test
VG 30

VG 30
+ PPA
2%
 137

 VG 30 binder without the stripping agent

 VG 30 binder with the stripping agent
 VG 30 + PPA 2% binder without the stripping agent
 VG 30 + PPA 2% binder with the stripping agent

It was observed from visual inspection that the extent of stripping both for the
immersion test in stage 1 and boiling test in stage 2 was very minimal for PPA
modified bitumen and can be observed the aggregates could retain the shiny lustre
even after two stages of stripping test while a notable amount of stripping was
observed for VG 30 bitumen which can be noticed clearly from photographs.

4.8 Summary
The detailed discussion on experimental investigations mentioned in research
methodology of Chapter 1 pertaining to five stages of research activity viz.
fundamental tests, dynamic mechanical analysis, experiments related to indigenously
developed rotational viscometer, micro structural investigations using SEM, XRD &
TG-DSC and experiments on bituminous mixes were presented in this Chapter.