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

net/publication/282522397

Use of the Multiple Stress Creep Recovery (MSCR) Test Method to Characterize
Polymer-Modified Asphalt Binders

Article  in  Journal of Testing and Evaluation · January 2016

DOI: 10.1520/JTE20140061

CITATIONS READS

6 571

4 authors, including:

Zahid Hossain Debaroti Ghosh

Arkansas State University - Jonesboro MN Department of Transportation
67 PUBLICATIONS   290 CITATIONS    10 PUBLICATIONS   11 CITATIONS   

SEE PROFILE SEE PROFILE

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

Evaluation of Bio-Fog Sealants for Pavement Preservation View project

Use of Nanoclays as Alternatives of Polymers toward Improving Performance of Asphalt Binders View project

All content following this page was uploaded by Debaroti Ghosh on 23 February 2017.

The user has requested enhancement of the downloaded file.

Journal of
Testing and Evaluation
Zahid Hossain1, Debaroti Ghosh2, Musharraf Zaman3, and Kenneth Hobson4

DOI: 10.1520/JTE20140061

Use of the Multiple Stress Creep

Recovery (MSCR) Test Method
to Characterize Polymer-
Modified Asphalt Binders
VOL. 44 / NO. 1 / JANUARY 2016
 Journal of Testing and Evaluation

doi:10.1520/JTE20140061 / Vol. 44 / No. 1 / January 2016 / available online at www.astm.org

Zahid Hossain1, Debaroti Ghosh2, Musharraf Zaman3, and Kenneth Hobson4

Use of the Multiple Stress Creep Recovery

(MSCR) Test Method to Characterize
Polymer-Modified Asphalt Binders

Reference
Hossain, Z., Ghosh, D., Zaman, M., and Hobson, K., “Use of the Multiple Stress Creep Recovery (MSCR)
Test Method to Characterize Polymer-Modified Asphalt Binders,” Journal of Testing and Evaluation, Vol.
44, No. 1, 2016, pp. 507–520, doi:10.1520/JTE20140061. ISSN 0090-3973

ABSTRACT
Manuscript received February 19, 2014; Use of polymer-modified binder has become much more important in recent years to enhance
accepted for publication February 23,
2015; published online July 6, 2015. the durability and strength of asphalt concrete (AC) pavements in order to sustain higher traffic
1
volumes, heavier loads, and extreme weather conditions. Unlike neat (unmodified) asphalt
Assistant Professor, Department of Civil
Engineering, AR State Univ., PO Box binders, polymer-modified binders are sensitive to the applied stress levels and show a non-
1740, LSW#239, State University, AR linear response in rutting factor and phase angle. The widely used dynamic shear rheometer
72467 (Corresponding author),
(DSR) test (AASHTO T315) is not sufficient to capture viscoelastic properties of polymer
e-mail: mhossain@astate.edu
modified-asphalt binders. Thus, state transportation agencies often need to conduct additional
2
Former Graduate Research Assistant,
expensive and time consuming “PG Plus” tests (e.g., elastic recovery (ER), force ductility,
School of Civil Engineering and
Environmental Science, The Univ. of toughness, and tenacity) for characterizing polymer-modified binders. Multiple stress creep
Oklahoma, 202 West Boyd Street, recovery (MSCR), a recently introduced test method (AASHTO TP 70) for measuring high
CEC #334, Norman, OK 73019,
e-mail: debaroti.ghosh@ou.edu temperature properties of an asphalt binder, is expected to replace the existing AASHTO T315
3
for short-term aged binder. In the current study, MSCR and Superpave tests were conducted on
David Ross Boyd Professor and Aaron
Alexander Professor, School of Civil three commonly used performance grade (PG) binders (PG 64-22, PG 70-28, and PG 76-28)
Engineering and Environmental Science, from 12 different sources throughout Oklahoma. Based on expected traffic loads and service
and Alumni Chair Professor of Petroleum
temperatures, the tested binder samples were graded in accordance with the MSCR grading
and Geological Engineering, College of
Engineering, The University of system. The non-recoverable creep compliance (Jnr) and MSCR % recovery values obtained
Oklahoma, 202 West Boyd Street, from the MSCR tests were then analyzed using the polymer curve and quadrant methods to
CEC #334, Norman, OK 73019,
e-mail: zaman@ou.edu assess the feasibility of the adoption of the MSCR test method for conditions prevailing in
4
Oklahoma. Findings of this study reveal that the MSCR test method can be used to characterize
Bituminous Engineer, Oklahoma
Department of Transportation, 200 NE polymer-modified binders without penalizing suppliers or risking the users.
21st Street, Oklahoma City, OK 73105,
e-mail: khobson@odot.org
Keywords
elastic recovery, rutting, mscr, creep compliance, non-recoverable strain

Copyright V
C 2015 by ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959. 507
508 Journal of Testing and Evaluation
Introduction
In the current Superpave specifications, complex modulus (G*)
and phase angle (d) are measured in the linear visco-elastic
range. For neat binders, the flow is linear and not sensitive to
the stress level applied in the widely used AASHTO T315 [1]
using a dynamic shear rheometer (DSR). For polymer modified
asphalt (PMA) binders, however, the response is nonlinear and
sensitive to the stress level of the test. The polymer chains can
be rearranged substantially with the increase of stress. As a con-
sequence, many state departments of transportation (DOTs)
require additional tests that are referred to as “PG Plus” tests
(e.g., elastic recovery (ER), force ductility, toughness, and tenac-
ity) in their quality assurance process. Drawbacks of these “PG
Plus” tests include the absence of standards across the states
and variations in the test methods. Furthermore, the “PG Plus”
tests are empirical in nature and do not consider in-service
temperatures and stress conditions of the pavements.
The multiple stress creep recovery (MSCR) (AASHTO TP
70 [2]) test method incorporates the well-established creep and
recovery test concept in order to evaluate the binder’s potential
for permanent deformation. Traditionally, the MSCR test is
conducted at two stress levels (0.1 and 3.2 kPa); at each stress
level, asphalt binder specimens are subjected to ten cycles of
loading and recovery, as shown in Fig. 1. After applying the load
for 1 s, it is removed and the sample is allowed to recover for
9 s. The non-recoverable creep compliance (Jnr) and MSCR %
recovery are the desired parameters to be reported from this
test. The Jnr is a measure of the amount of residual strain left in
the specimen after repeated creep and recovery, relative to the
amount of stress applied. The MSCR % recovery is a measure of
how much the sample returns to its previous shape after being
repeatedly stretched and relaxed. The Asphalt Institute (AI)
researchers [3–5] have taken leadership roles in establishing the
FIG. 1 Examples of modified asphalt binder response to repeated loading [1].
MSCR testing protocol(s) and guidelines so that state DOTs are
on-board with such initiatives.
D’Angelo and Dongre [6] reported findings of MSCR test
results while investigating the optimum blending of polymers in
a base binder. In particular, the effects of modification parame-
ters such as polymer content, compatibility of the base binder,
cross linker content, and the amount of polyphosphoric acid
(PPA) were studied. These researchers reported that the disper-
sion, along with the blending time and temperature of styrene-
butadiene-styrene (SBS), had noticeable effects on the perform-
ance of the PMA binder, and the ER test fails to address this
issue since it is performed at a low temperature (25 C). These
researchers concluded that the MSCR parameters (Jnr and
MSCR % recovery) indicated larger differences within the dif-
ferent processing methods of PMA binders depending upon the
added polymer/crosslink type and content than the ER test
data. Thus, the MSCR parameters showed more sensitivity to
optimum blending, which was also verified by florescence
microscopy.
Li et al. [7] investigated PG, MSCR, mix dynamic modulus,
and mix fatigue resistance of asphalt binders modified with
PPA, PPA plus Elvaloy, SBS, and SBS plus PPA. The binder
modified with PPA showed the worst performance (the highest
Jnr and the lowest MSCR % recovery), whereas the binders
modified with PPA plus Elvaloy and with SBS plus PPA were
the best (the lowest Jnr and the highest MSCR % recovery), irre-
spective of their aging conditions (laboratory aged and recov-
ered from asphalt mixes). It was concluded that all four types of
binder under the rotational thin film oven (RTFO)-aging condi-
tion met the AI introduced Jnr criteria for heavy traffic
(Jnr < 2 kPa1), whereas the recovered binders from asphalt
mixes were satisfactory for very heavy traffic (Jnr < 1.0 kPa1).
Zeleew et al. [8] presented a comparative evaluation of
MSCR test results of four warm mix asphalt (WMA) technolo-
gies: three foaming processes (Advera, Low Emission Asphalt
(LEA), and Gencor) and an organic additive (Sasobit). In this
study, a PG 64-22 binder was used as the control binder.
The WMA technologies exhibited lower stiffness (i.e., high
compliance) when tested at higher stress levels and test temper-
atures. This study concluded that the MSCR test results of
WMA additive modified binders are highly dependent on test
temperatures and shear stress levels, and MSCR is a suitable
tool for ranking deformation properties of WMA.
Gibson et al. [9] reported poor performance of existing
Strategic Highway Research Program (SHRP) specifications in
predicting rut of polymer modified binders. On the other hand,
the MSCR test showed an excellent correlation with field
rutting. At 64 C, the correlation of the binder’s rutting factor
(G*/sind) to the mix’s field rutting provided a coefficient of
determination (R2) value of only 0.13, whereas the correlation
between the Jnr and the filed rutting provided a significantly
high R2 value of 0.82.

FIG. 2 Plot of neat and modified binders showing the relationship between
stress and non-recoverable compliance [8].
D’Angelo [10] performed MSCR tests on a few neat and
PMA binders to determine if a relationship existed between the
existing SHRP grading and the Jnr at 3.2 kPa1. For neat bind-
ers, properties obtained from the MSCR specifications were
found to be similar to those obtained for the original SHRP
specifications; i.e., flow had linear behavior beyond a stress level
of 3.2 kPa. However, PMA binders had non-linear properties, as
shown in Fig. 2.
IMPLEMENTATION OF THE MSCR METHOD
According to Horan [4], there are two ways to implement the
MSCR test: partial implementation (PI), and full implementa-
tion (FI). The PI includes the use of the MSCR test method in
conjunction with AASHTO M320 [11]. The FI includes the
adoption of the MSCR test method (AASHTO TP 70 [2]) and
specification (AASHTO MP 19 [12]), a revised grading system
strictly based on climate and loading, which eliminate grade
bumping and “PG Plus” tests. It is further mentioned that some
transportation agencies may be reluctant to accept the FI of the
MSCR method and prefer the adoption of the PI. In the FI pro-
cess, the names of the binder types will change. A possible
example of this is “PG 64-22H” instead of “PG 70-22” because
MSCR tests are performed at 64 C. The PI of the MSCR
method could be done as outlined in AI’s “Guidance on the Use
of the MSCR Test with AASHTO M320 Specification” [5].
According to AI guidelines, if the evaluation of the delayed elas-
tic response of binders is the target for MSCR test, other “PG
Plus” tests with a similar purpose should be eliminated. Further-
more, it is reported that “PG Plus” tests are expected to provide
simplified results with a much higher degree of error than
MSCR. Thus, a strong correlation between “PG Plus” tests and
MSCR results may not exist.
In an ongoing study, Kabir [13] illustrated the steps neces-
sary for adoption of the MSCR test method for conditions
prevailing in Louisiana. This study intends to test three types
HOSSAIN ET AL. ON MSCR ASSESSMENT 509
(PG 70-22, PG 76-22, and PG 82-22) of local binders from
multiple sources and set acceptable specifications for the state of
Louisiana. Off partially reported test results of 36 binder sam-
ples, the MSCR grades of 20, 9, and 7 samples were found to be
PG 64E-XX, PG 64V-XX, and PG 64H-XX, respectively. None
of the tested binders in this study was found to be stress sensi-
tive up to a stress level of 3.2 kPa.
Mehta et al. [14] also presented the on-going activities of
providing recommendations on the MSCR test method to
expand the use of different polymers for conditions prevailing
in New Jersey. These researchers observed that binders with the
Jnr value less than 0.5 kPa1 showed improved high temperature
performance. Furthermore, it was reported that an MSCR %
recovery at 3.2 kPa greater than 40 % would ensure that the
binder would be above the MSCR curve, which could serve as a
criterion to decide whether the binder would withstand heavy
and extreme levels of traffic.
The MSCR test method has been or is being implemented
in the United States by various user groups. Such initiatives
have led a few state agencies to verify the reproducibility of
MSCR tests and the specification criteria [15,16]. Research is
being conducted by the AI on an inter-laboratory study to
determine the precision of AASHTO TP 70 [2] for the South-
eastern Asphalt User/Producer Group (SEAUPG) and for the
North East Asphalt/User Producer Group (NEAUPG). Four
states (Maine, New Hampshire, Rhode Island, and Florida)
have already adopted the FI of the MSCR test method with
modified MSCR grades [17]. Recently, the DOTs of Louisiana
and New Jersey have decided to adopt the MSCR test method
after conducting multi-year projects [13,14]. Both of these
studies concluded that it was possible to replace the currently
used ER and force ductility test data with the MSCR test data.
Some other relevant studies performed MSCR tests on
binders modified with acid [7], elastomers [18], and emulsifiers
[19] and reported the Jnr as a prime indicator of the MSCR
grade and pavement rutting. Other studies also suggested that
the MSCR test method is a satisfactory tool for estimating
permanent strains in asphalt mixes with different polymer
modified binders [20–22]. Besides hot mix asphalt (HMA),
binders using WMA technologies such as Sasobit, Advera,
surfactant, tensoactive additives, and Evotherm have been char-
acterized by the MSCR test method [23–26].
In recent years, a few other studies have been conducted
to evaluate the relationship between the Jnr and measured field
rutting. Among others, three field studies, namely, I-55 in
Mississippi, the Federal Highway Administration (FHWA)
Accelerated Loading Facility (ALF) polymer study, and the
Minnesota Road Hamburg mix study, have been widely refer-
enced in the literature [15,27]. These studies observed that a
reduction of the Jnr value by half typically reduced the corre-
sponding field rutting by half. These studies further reported
that the standard RTFO environmental grade in the existing
510 Journal of Testing and Evaluation

SHRP binder specification, |G*|/sin(d) of 2.2 kPa, was found to

be equivalent to a Jnr value of 4.0 kPa1measured at 3.2 kPa. FIG. 3 Analyses of MSCR test results: (a) curve representing stepped scale
of minimum % recovery for measured Jnr and (b) quadrant plot for
Since no analysis has been done for binders and condi- MSCR % recovery vs. % elastic recovery or phase angle.
tions (climate and traffic) prevailing in Oklahoma, no guide-
lines are currently available for Oklahoma. Like many other
state DOTs, the Oklahoma Department of Transportation
(ODOT) is in the process of implementing the MSCR test
method in its quality assurance process, and it needs actual
MSCR test data of local binders and pavement temperatures
for a better understanding and confidence. With a reliability
of 98 %, the high PG temperature of pavements throughout
Oklahoma varies from 64 to 70 C [5], which have been con-
sidered as MSCR testing temperatures of the current study.
The present study intends to produce MSCR data for different
types of PMA binders in Oklahoma and recommends neces-
sary guidelines for ODOT.

MSCR Data Analysis Approach

POLYMER METHOD
As mentioned in the preceding section, in the case of PMA
binders, the strain response is non-linear and sensitive to the
stress level of the test. The Jnr value obtained from the MSCR
test provides a good correlation with the high temperature rut-
ting for both neat and polymer-modified binders. The MSCR %
recovery measurement can identify and quantify how the poly-
mer is working in the binder. In the current practice, no specifi-
cations have been developed for any state DOT yet for the
MSCR % recovery but recommendations have been made per-
taining to a minimum value, as described in AASHTO TP 70
[2] and presented in Fig. 3(a) [28]. A standard MSCR curve,
relating the MSCR % recovery and Jnr at 3.2 kPa shown in
Fig. 3(a), can be expressed by Eq 1 [3]. The MSCR curve based
on Eq 1 is used to examine whether the binder passes or fails
the MSCR % recovery. Data points above the MSCR curve are
considered to have sufficient delayed elastic response for an
elastomeric-modified asphalt binder.
y ¼ 29:37x0:2633 (1)
where:
y ¼ MSCR % recovery, and
x ¼ Non-recoverable creep compliance at 3.2 kPa
(Jnr@3.2 kPa).
The MSCR curve stops at Jnr ¼ 2 kPa1. It is known that Jnr
values greater than 2 kPa1 are not required to have any mini-
mum value of MSCR % recovery [3]. The MSCR curve marks a
range of elasticity. Data points above the MSCR curve indicate
binders of high elasticity or binders that are modified with elas-
tomeric polymers. Data points below the MSCR curve indicate
binders of poor elasticity or binders that are not modified with
enough elastomeric polymers.
QUADRANT METHOD
Four quadrant plots (Fig. 3(b)) can be used to analyze the
MSCR and ER test data at the same time [3]. The quadrant
analysis is a simple way to organize customer satisfaction
data to express how and where to improve an organization’s
operation. Fig. 3(b) shows a quadrant plot in which the ER is
plotted along the x-axis and the MSCR % recovery is shown
along the y-axis. The four quadrants (first, second, third, and
fourth, in this order) are named as follows: pass (meets both ER
and MSCR % recovery targets), user risk, fail (fails both ER and
MSCR % recovery targets), and supplier risk. The term “user
risk” is used to indicate a situation where the MSCR % recovery
value meets the proposed specification while not meeting the
ER or phase angle (d) criterion. By contrast, the term “supplier
risk” indicates a situation where the current ER or d criterion is
met, but the MSCR % recovery value does not meet the pro-
posed specification. In both cases, the term “risk” applies only
to the comparison between the MSCR % recovery and ER test
results and does not imply any increased risk of reduced per-
formance. The limit for dividing the quadrants are as follows:
the x-axis is equal to the minimum % ER requirement, and the
y-axis is equal to the minimum MSCR % recovery requirement.
 HOSSAIN ET AL. ON MSCR ASSESSMENT 511

The minimum MSCR % recovery varies depending on the

binder grades. The minimum % ER requirement depends on FIG. 4 Flow chart for the performed study.

the agency’s minimum ER requirement. The limit for the phase

angle is the maximum phase angle that an agency practices.
Anderson [3] also reported that when evaluating the data
graphically in four quadrant plots, the minimum MSCR %
recovery requirement should be established in a manner where
the “user risk” is approximately equal to the “supplier risk.” As
per AI [5], at 64 C, an appropriate MSCR % recovery at 3.2 kPa
is 15 % less than the current ER criterion if the ER value at
25 C is being used. The minimum ER values for PG 70-28 and
PG 76-28 binders are 65 and 70 %, respectively. Thus, the
MSCR % recovery values of PG 70-28 and 76-28 binders would
be 50 % and 55 %, respectively. Based on the MSCR test
data and ODOT’s current ER criteria, the proposed study inves-
tigates whether the four quadrant method is applicable and
helpful in establishing specification limits of MSCR % recovery
at 3.2 kPa.

Materials and Methodology

In the present study, three types of commonly used ODOT- Results and Discussions
certified binders (PG 64-22, PG 70-28, and PG 76-28) from 12
different sources throughout Oklahoma were tested at 64 C and SUPERPAVE TEST RESULTS

at two stress levels (0.1 and 3.2 kPa). Specifically, MSCR tests
were conducted on 59 PG 64-22 samples from eleven different
sources (A1 through A11), 47 PG 70-28 samples from five dif-
ferent sources, and 48 PG 76-28 samples from four different
sources. The three types of binders from two selected sources
(A3 and A7) were evaluated for viscosity measurements and
Superpave PG grading for quality assurance purposes. The
Superpave binder tests include viscosity determination using a
Brookfield Rotational Viscometer (RV) (AASHTO T316 [29]),
short term aging using an RTFO (AASHTO T240 [30]), long
term aging using a Pressure Aging Vessel (PAV) (AASHTO
R28 [31]), determination of rutting and fatigue factors of
un-aged, RTFO-aged and PAV-aged samples using a DSR
(AASHTO T315 [1]), determination of low-temperature
thermal cracking creep stiffness, and rate of stress relaxation
(slope of m-value) using a bending beam rheometer (BBR)
(AASHTO T313 [32]). A high level plan of the current study
is presented in Fig. 4. The current study evaluated binders
modified with elastomers, which enhance strength at high
temperatures as well as elasticity at low temperatures. Three
types of commonly used elastomeric copolymers are styrene-
butadiene (SB), styrene-butadiene-styrene (SBS), and styrene-
butadiene-rubber (SBR). However, the type and amount of
elastomeric copolymers used in modifying the binders were
unknown to the research team. Plastomers, on the other hand,
enhance the strength but not elasticity of binders and were not
considered in this study as they are not generally used in many
states such as Oklahoma, Arkansas, and Illinois.
The PG of each binder was determined in accordance with
AASHTO M320 [11]. A summary of the verification results of
the continuous PG grades of the selective binders is presented
in Table 1.
ANALYSIS OF MSCR TEST DATA
The AASHTO MP 19 [12] test specification states that the
MSCR test should be performed on RTFO-aged binders at
64 C. Tables 2–4 present the MSCR test data and analysis of
tested asphalt binders. As presented in these tables, the stress-
sensitivity (column 5), based on Eq 2, validates whether the
binder meets the requirement of Jnrdiff, in accordance with
AASHTO MP 19 [12]. Equation 2 is used to examine whether
the asphalt binder is stress sensitive. If the value of Jnrdiff does
not exceed 75 %, then the binder meets the AASHTO MP 19
[12] requirement and a value of “Yes” is shown in column 5.
Otherwise, a value of “No” is shown in column 5.
Jnr;3:2kPa  Jnr;0:1kPa
Jnrdiff ¼  100  75% (2)
Jnr;0:1kPa
The MSCR % recovery (column 9) validation is essentially
to verify if the binder meets the AASHTO TP 70 [2] criterion,
as shown in Eq 3. Based on the outcome (True or False) of this
verification a value of “Yes” or “No” is shown in column 9. Fur-
thermore, the value of column 9 is “N/A” if the condition
Jnr@3.2 kPa > 2 kPa1 is true, indicating an insignificant recovery.
0:2633
29:371  Jnrð3:2kPaÞ < R3200 (3)
512 Journal of Testing and Evaluation

TABLE 1 Superpave test result and actual grading of selective binders.

Superpave Test Results (Performance Grading Verification)

DSR Test BBR Test

Binder Temperature Unaged Temperature RTFO aged Temperature m-value Temperature S-value at Actual
Source Type ( C) G*/sin d (kPa) ( C) G*/sin d (kPa) ( C) at 60 s ( C) 60 s (MPa) Grade
A3 PG 64-22 61 2.86 61 7.84 9 0.40 9 95.50 PG 67-28
64 1.61 64 5.19 12 0.38 12 143.33
67 1.12 67 3.58 15 0.35 15 213.22
67.76 1a 69.57 2.2b 20.53 0.3c 18.7 300d
30.53 28.7
A3 PG 76-28 73 2.16 73 4.45 15 0.36 15 83.34 PG 80-36
76 1.62 76 4.15 18 0.36 18 114.82
79 1.21 79 3.35 21 0.34 21 181.15
80.5 1a 83.3 2.2b 29.4 0.3c 26.4 300d
39.4 36.4
A3 PG 70-28 67 2.23 67 6.17 15 0.36 15 99.55 PG 74-31
70 1.53 70 4.12 18 0.33 18 154.63
73 1.16 73 3.06 21 0.30 21 237.36
74.3 1a 74.5 2.2b 21 0.3c 23.3 300d
31 33.3
A7 PG 64-22 61 2.74 61 3.62 9 0.32 9 110.52 PG 67-26
64 1.44 64 4.02 12 0.33 12 150.04
67 1.11 67 2.76 15 0.31 15 207.54
67.96 1a 68.35 2.2b 16.45 0.3c 19.8 300d
26 29.0
A7 PG 76-28 73 1.99 73 6.23 15 0.37 15 76.00 PG 86-32
76 1.56 76 4.82 18 0.40 18 156.32
79 1.40 79 4.23 21 0.33 21 168.22
86.5 1a 89.26 2.2b 22.48 0.3c 54.2 300d
32.48 64.2
A7 PG 70-28 67 3.68 73 5.85 15 0.34 15 72.45 PG 76-35
70 2.46 76 4.15 18 0.33 18 108.68
73 2.32 79 3.23 21 0.32 21 151.49
101.3 1a 76.36 2.2b 25.22 0.3c 30.4 300d
35.22 40.4
a
AASHTO T315 original binder performance criteria: |G*|/sin(d) > ¼ 1.00 kPa.
b
AASHTO T315 RTFO binder performance criteria: |G*|/sin(d) > ¼ 2.20 kPa.
c
AASHTO T313 PAV binder m-value @ 60 s: 3.
d
AASHTO T313 PAV binder stiffness @ 60 s: 300 MPa.

Tables 2–4 also present the MSCR grades of tested binders
depending on the following criteria, as described in AASHTO
MP 19 [12]:
If Jnr ¼ < 4.0 then PG 64-XXS,
If Jnr ¼ < 2.0 then PG 64-XXH,
If Jnr ¼ < 1.0 then PG 64-XXV, and
If Jnr ¼ < 0.5 then PG 64-XXE.
where, S, H, V, and E denote traffic conditions of standard,
heavy, very heavy, and extreme, respectively.
From Tables 2–4, it is seen that all tested PG 76-28 binder
samples from all sources have been graded as PG 64-XXE, in
accordance with AASHTO TP 70 [2]. In the case of PG 70-28
binders, a majority (78 %) of the samples (35 out of 45) have
been graded as PG 64-XXE, while the remainder of the (22 %)
binder samples have been graded as PG 64-XXV. In the case of
PG 64-22 binders, a majority of samples (34 out of 37) have
been graded as PG 64-XXS, which is followed by PG 64-XXE
(2 out of 37), and PG 64-XXV (1 out of 37). D’Angelo [6]
also indicated that a Superpave PG 64-22 binder would be
equivalent to a PG 64-22S binder in the MSCR grading system.
Findings of the current study suggest that the Jnr criteria perti-
nent to traffic load recommended by the AI are applicable for
 HOSSAIN ET AL. ON MSCR ASSESSMENT 513

TABLE 2 MSCR test data and analysis for PG 76-28 binders.

1 2 3 4 5 6 7 8 9 10 11
Source Jnr@0.1 kPa Jnr@3.2 kPa Jnrdiff Stress Sensitivity R100 R3200 Rdiff % Recovery (Meets MSCR ER
ID (kPa1) (kPa1) (%) (Meets AASHTO MP 19) (%) (%) (%) AASHTO TP 70) GRADE (%)
A1 0.01 0.02 3.33 YES 97.73 97.73 0.0 YES PG64-28E 95
A1 0.01 0.01 7.21 YES 97.92 98.11 0.2 NO PG64-28E 97.5
A1 0.01 0.01 0.74 YES 98.03 98.08 0.1 YES PG64-28E 97.5
A1 0.02 0.02 8.46 YES 97.62 97.53 0.1 YES PG64-28E 97.5
A1 0.03 0.03 22.89 YES 95.61 96.50 0.9 YES PG64-28E 97.5
A1 0.02 0.02 5.39 YES 97.02 96.91 0.1 YES PG64-28E 95
A1 0.04 0.04 1.31 YES 94.48 94.51 0.0 YES PG64-28E 97.5
A1 0.02 0.02 13.24 YES 96.72 97.01 0.3 YES PG64-28E 95
A1 0.02 0.02 13.48 YES 97.40 97.69 0.3 YES PG64-28E 97.5
A1 0.02 0.02 1.69 YES 97.28 97.33 0.1 YES PG64-28E 97.5
A1 0.04 0.03 21.68 YES 95.6 96.4 0.8 YES PG64-28E 97.5
A1 0.02 0.02 4.33 YES 96.8 96.8 0.0 YES PG64-28E 95
A1 0.03 0.03 0.10 YES 96.2 96.1 0.0 YES PG64-28E 100
A1 0.02 0.02 12.40 YES 96.8 97.1 0.3 YES PG64-28E NA
A1 0.03 0.02 15.55 YES 96.4 96.8 0.4 YES PG64-28E NA
A3 0.03 0.03 3.80 YES 92.1 92.0 0.2 YES PG64-28E 92.5
A3 0.10 0.09 12.02 YES 86.2 87.9 1.9 YES PG64-28E 95
A3 0.10 0.10 1.45 YES 85.9 86.0 0.1 YES PG64-28E 95
A3 0.04 0.03 22.65 YES 94.9 95.9 1.0 YES PG64-28E 97.5
A3 0.08 0.07 17.32 YES 90.9 91.4 0.5 YES PG64-28E 95
A3 0.08 0.08 0.34 YES 88.6 88.0 0.7 YES PG64-28E 95
A3 0.15 0.14 3.17 YES 81.84 82.34 0.6 YES PG64-28E 92.5
A3 0.13 0.14 2.79 YES 81.57 80.93 0.8 YES PG64-28E NA
A3 0.11 0.10 6.59 YES 83.96 84.36 0.5 YES PG64-28E 95
A3 0.12 0.11 1.96 YES 85.72 85.71 0.0 YES PG64-28E NA
A3 0.11 0.11 3.40 YES 84.63 84.24 0.5 YES PG64-28E 95
A3 0.08 0.08 6.10 YES 85.28 85.49 0.2 YES PG64-28E NA
A3 0.08 0.08 0.90 YES 42.23 87.50 87.3 YES PG64-28E NA
A4 0.06 0.05 12.07 YES 88.51 89.28 0.9 YES PG64-28E 95
A4 0.11 0.10 10.46 YES 84.73 86.27 1.8 YES PG64-28E 97.5
A4 0.03 0.03 20.66 YES 95.6 96.49 0.9 YES PG64-28E 97.5
A4 0.08 0.09 1.90 YES 86.94 86.93 0.0 YES PG64-28E 97.5
A4 0.10 0.08 14.37 YES 87.08 88.58 1.7 YES PG64-28E NA
A4 0.15 0.13 12.95 YES 84.12 85.65 1.8 YES PG64-28E NA
A4 0.12 0.12 0.67 YES 85.29 85.08 0.2 YES PG64-28E 95
A7 0.03 0.03 3.05 YES 93.74 93.63 0.1 YES PG64-28E 92.5
A7 0.02 0.02 1.85 YES 95.42 95.11 0.3 YES PG64-28E 95
A7 0.05 0.04 15.83 YES 94.29 94.78 0.5 YES PG64-28E 95
A7 0.02 0.02 11.09 YES 95.69 95.72 0.0 YES PG64-28E 92.5
A7 0.03 0.03 9.97 YES 95.11 95.27 0.2 YES PG64-28E 97.5
A7 0.01 0.01 0.49 YES 96.79 96.75 0.0 YES PG64-28E NA
A7 0.00 0.01 — NO 99.99 96.44 3.6 NO PG64-22E NA

Oklahoma. This means, for “extreme” (more than 30  106
ESALs and standing traffic loading), “very heavy” (more than
30  106 ESALs or standard traffic loading), “heavy” (from 10
to 30  106 ESALs or slow moving traffic loading), and
“standard” (less than 10  106 ESALs and standard traffic load-
ing) traffic, the corresponding Jnr value should not exceed 0.5,
1.0, 2.0, and 4.0 kPa1, respectively.
The MSCR % Recovery and Jnr values of the tested binder
samples are presented in Fig. 5(a). According to AASHTO TP
70 [2], the binder position under the MSCR curve indicates
insufficient presence of elastomeric modifier in the binder
(Fig. 5(a)). The binder position above the MSCR curve indicates
the presence of polymer in the binder. As expected, both PG
70-28 and PG 76-28 binders are well above the MSCR curve,
514 Journal of Testing and Evaluation

TABLE 3 MSCR test data and analysis for PG 70-28 binders.

1 2 3 4 5 6 7 8 9 10 11
Source Jnr@0.1 kPa Jnr@3.2 kPa Jnrdiff Stress Sensitivity R100 R3200 Rdiff % Recovery (Meets MSCR ER
ID (kPa1) (kPa1) (%) (Meets AASHTO MP 19) (%) (%) (%) AASHTO TP 70) Grade (%)
A1 0.05 0.05 0.88 YES 94.95 94.99 0.0 YES PG64-28E 95
A1 0.05 0.05 5.48 YES 95.81 95.15 0.7 YES PG64-28E 95
A1 0.09 0.07 22.97 YES 91.90 93.27 1.5 YES PG64-28E 95
A1 0.21 0.14 32.09 YES 86.63 90.18 4.1 YES PG64-28E 95
A1 0.10 0.09 9.67 YES 90.93 91.10 0.2 YES PG64-28E 92.5
A1 0.11 0.08 27.63 YES 90.48 92.45 2.2 YES PG64-28E 95
A1 0.12 0.09 30.63 YES 88.04 90.65 3.0 YES PG64-28E 92.5
A1 0.08 0.06 28.04 YES 92.87 94.33 1.6 YES PG64-28E 92.5
A1 0.08 0.06 26.25 YES 92.33 93.59 1.4 YES PG64-28E 92.5
A1 0.11 0.07 35.94 YES 90.50 92.70 2.4 YES PG64-28E 95
A1 0.15 0.13 11.75 YES 87.1 87.2 0.2 YES PG64-28E 95
A1 0.12 0.08 32.74 YES 89.8 92.7 3.3 YES PG64-28E 95
A1 0.14 0.12 12.12 YES 87.0 87.7 0.9 YES PG64-28E NA
A1 0.05 0.05 1.43 YES 94.8 94.9 0.1 YES PG64-28E NA
A1 0.07 0.06 17.97 YES 92.1 93.1 1.1 YES PG64-28E NA
A1 0.04 0.04 1.38 YES 95.5 95.5 0.0 YES PG64-28E NA
A3 0.54 0.78 44.97 YES 63.8 50.2 21.3 YES PG64-28V 90
A3 0.47 0.58 23.63 YES 65.5 59.1 9.7 YES PG64-28V 82.5
A3 0.30 0.35 14.64 YES 70.7 67.1 5.2 YES PG64-28E 95
A3 0.46 0.64 38.26 YES 66.3 54.1 18.5 YES PG64-28V 87.5
A3 0.37 0.50 33.97 YES 64.0 51.5 19.6 YES PG64-28E 85
A3 0.57 0.82 43.33 YES 60.27 45.14 25.1 YES PG64-28V 85
A3 0.50 0.70 40.07 YES 60.34 46.39 23.1 YES PG64-28V NA
A3 0.49 0.66 34.57 YES 58.28 44.52 23.6 YES PG64-28V 87.5
A3 0.39 0.52 32.65 YES 67.05 56.05 16.4 YES PG64-28V NA
A3 0.40 0.55 36.66 YES 64.75 51.42 20.6 YES PG64-28V 92.5
A3 0.34 0.42 22.63 YES 69.08 61.30 11.3 YES PG64-28E 85
A3 0.35 0.44 26.69 YES 65.31 55.53 15.0 YES PG64-28E NA
A3 0.08 0.07 2.47 YES 82.67 82.08 0.7 YES PG64-28E NA
A4 0.13 0.12 8.82 YES 80.61 81.48 1.1 YES PG64-28E 92.5
A4 0.15 0.13 9.60 YES 79.17 79.97 1.0 YES PG64-28E 92.5
A4 0.20 0.23 18.00 YES 71.48 68.14 4.7 YES PG64-28E 92.5
A4 0.04 0.03 16.62 YES 94.9 95.3 0.4 YES PG64-28E 95
A4 0.15 0.14 2.45 YES 79.45 78.85 0.8 YES PG64-28E 87.5
A4 0.29 0.34 14.86 YES 69.45 65.21 6.1 YES PG64-28E 90
A4 0.19 0.20 5.28 YES 80.51 79.35 1.4 YES PG64-28E 92.5
A4 0.21 0.19 8.04 YES 78.5 80.47 2.5 YES PG64-28E NA
A4 0.42 0.50 17.28 YES 63.95 59.16 7.5 YES PG64-28E NA
A5 0.20 0.21 6.11 YES 73.97 71.74 3.0 YES PG64-28E 87.5
A5 0.28 0.34 21.90 YES 71.35 65.96 7.6 YES PG64-28E 90
A5 0.35 0.38 11.09 YES 69.49 66.13 4.8 YES PG64-28E 90
A10 0.17 0.17 1.18 YES 84.44 84.03 0.5 YES PG64-28E 92.5
A10 0.39 0.53 34.19 YES 61.2 51.13 16.5 YES PG64-28V NA
A10 0.17 0.17 0.29 YES 77.96 78.01 0.1 YES PG64-28E NA

representing significant recoverable strains of the PMA binders.
On the other hand, PG 64-22 binder samples are below the
MSCR curve as the MSCR % recovery is expected to be low for
unmodified binders. Wasage et al. [22] also noted a strong
recovery feature of PMA binders in their corresponding study.
The four-quadrant plot (MSCR % recovery versus % ER) of
the tested PG 76-28 binders are presented in Fig. 5(b), which
reveals whether these binders are categorized as “user risk,”
“supplier risk,” “both at risk,” or “none at risk.” As mentioned
earlier, if a binder meets MSCR % recovery, but fails the % ER
 HOSSAIN ET AL. ON MSCR ASSESSMENT 515

TABLE 4 Partial MSCR test data and analysis for PG 64-22 binders.

1 2 3 4 5 6 7 8 9 10
Source Jnr@0.1 kPa Jnr@3.2 kPa Jnrdiff Stress Sensitivity R100 R3200 Rdiff % Recovery (Meets MSCR
ID (kPa1) (kPa1) (%) (Meets AASHTO MP 19) (%) (%) (%) AASHTO TP 70) GRADE
A1 3.28 3.27 0.34 YES 3.37 0.42 112.5 N/A PG64-22S
A1 2.18 2.47 13.34 YES 7.64 1.48 80.7 N/A PG64-22S
A1 3.07 3.15 2.87 YES 1.47 0.80 154.8 N/A PG64-22S
A1 2.95 3.14 6.45 YES 1.90 0.77 59.2 N/A PG64-22S
A1 2.86 3.14 9.85 YES 4.47 0.81 81.9 N/A PG64-22S
A1 2.97 3.33 12.12 YES 6.30 0.73 88.3 N/A PG64-22S
A1 2.49 2.63 5.71 YES 1.68 1.65 1.9 N/A PG64-22S
A2 4.04 4.22 4.43 YES 0.60 0.11 81.7 N/A N/A
A2 3.81 3.86 1.52 YES 2.14 0.18 108.4 N/A PG64-22S
A2 3.82 4.18 9.42 YES 4.84 0.01 99.8 N/A N/A
A2 3.79 3.93 3.72 YES 0.5 0.1 86.1 N/A PG64-22S
A2 4.03 4.09 1.64 YES 1.8 0.1 105.5 N/A N/A
A2 3.53 3.88 9.82 YES 5.0 0.1 97.1 N/A PG64-22S
A2 3.78 4.20 11.26 YES 6.3 0.0 100.0 N/A N/A
A2 3.89 4.14 6.35 YES 2.2 0.1 96.8 N/A N/A
A2 3.89 4.17 7.21 YES 2.9 0.1 98.2 N/A N/A
A2 3.63 3.89 7.28 YES 3.0 0.1 95.4 N/A PG64-22S
A2 3.83 3.98 3.81 YES 0.0 0.2 599.0 N/A PG64-22S
A2 2.28 2.66 16.75 YES 9.2 1.5 83.7 N/A PG64-22S
A3 2.41 2.59 7.64 YES 1.9 1.6 14.2 N/A PG64-22S
A4 2.10 2.39 13.59 YES 67.48 47.89 29.0 N/A PG64-22S
A4 1.90 2.31 21.64 YES 7.61 1.51 80.2 N/A PG64-22S
A4 2.12 2.52 18.66 YES 12.97 2.99 76.9 N/A PG64-22S
A4 2.59 2.96 14.20 YES 9.40 2.40 74.5 N/A PG64-22S
A10 2.136 2.227 4.2603 YES 0.3614 1.686 366.5 N/A PG64-22S
A11 3.832 4.178 9.0292 YES 4.272 0.1212 97.1 N/A N/A
— — — — — — — — — —
— — — — — — — — — —
A11 3.943 3.942 0.025 YES 3.296 0.1193 103.6 N/A PG64-22S
A11 3.993 3.963 0.751 YES 3.908 0.1316 103.3 N/A PG64-22S
A11 4.224 4.379 3.6695 YES 1.29 0.0622 104.8 N/A N/A
A11 4.135 4.289 3.7243 YES 0.141 0.0676 148.0 N/A N/A
A11 3.821 4.073 6.5951 YES 2.291 0.1146 94.9 N/A N/A
A11 3.67 3.949 7.6022 YES 3.001 0.1615 94.6 N/A PG64-22S
A12 2.197 2.522 14.793 YES 8.127 1.813 77.6 N/A PG64-22S

then the user is at risk (user risk). On the contrary, if the binder
meets the % ER but fails the MSCR % recovery, then the sup-
plier is at risk (supplier risk). If neither % ER nor MSCR %
recovery criterion is satisfied, both the supplier and the user are
at risk (both at risk). If both % ER and MSCR % recovery crite-
ria are met, neither the supplier nor the user is at risk (none at
risk). The ER and MSCR % recovery at 3.2 kPa (R3200) values of
PG 76-28 binders range from 92.5 to 100 %, and from 80.9 to
98.1 %, respectively. Thus, an MSCR % recovery value of 55 %,
based on the AI recommendations and ODOT’s current ER
limit, is applicable for PG 76-28 binders. However, a value of
55 % as the MSCR % recovery for PG 76-28 binders appears
to be a conservative approach for conditions prevailing in
Oklahoma. Thus, a value of 80 % for the MSCR % recovery is
recommended for PG 76-28 binders without putting any
supplier or user at risk.
In the case of PG 70-28 binders, only 2 out of 11 suppliers
are at risk for not meeting the MSCR % recovery value of 50 %
(Fig. 5(c)). However, the MSCR % recovery values of binders
from these suppliers are about 45 %, which is 5 % lower than
the AI recommended MSCR % recovery of 50 % for a PG 70-28
binder. Therefore, an MSCR % recovery value of 50 % can be
adopted for PG 70-28 binders without putting many suppliers
at risk. Even though a few suppliers are risk, it is suggested that
ODOT communicate the adoption plan of the MSCR test
method with all suppliers and establish some mutually agreed
516 Journal of Testing and Evaluation

FIG. 5
(a) MSCR % recovery vs. Jnr at 3.2 kPa, (b)
quadrant plot of MSCR % recovery at 3.2 kPa
vs. % elastic recovery for PG 76-28, and (c)
quadrant plot of MSCR % recovery at 3.2 kPa
vs % elastic recovery for PG 70-28.

contingency plan in the event their binders do not meet the
MSCR % recovery. Also, it can be noted that ER data is not
available for PG 64-22 binders as the ER test is not required for
un-modified binders. Thus, the quadrant plots do not include
any PG 64-22 binders.
• The MSCR % recovery increases with increased aging.
Furthermore, results of MSCR tests performed on unaged,
RTFO-aged, and PAV-aged binders at temperatures (67 and
70 C) other than 64 C, are shown in Table 5 and Fig. 6. From
data presented in Table 5 and Fig. 6, the following observations
are made:
• A general trend is that the higher the testing temperature,
the lower the MSCR % recovery. For instance, in the case
of the RTFO-aged PG 70-28 binders from A3, the MSCR
% recovery values at 67 C and 70 C are 41.5 % and
30.8 %, respectively. On the other hand, at 64 C, the
average MSCR % recovery of the same binder from A3 is
about 56 % (Table 3). Mehta et al. [18] also mentioned a
chance of temperature dependency for binders that
exhibit high MSCR % recovery values, which can result in
decreasing Jnr.
For instance, in the case of 70-28 binders from A3, at
64 C, the MSCR % recovery values under unaged and
PAV-aged conditions are 32 and 88 %, respectively. As
mentioned earlier, the average MSCR % recovery of the
same binder under RTFO-aged condition is about 56 %.
It can be noted that the MSCR % recovery is the ratio of
the recoverable strain and the peak strain, and the recov-
erable strain is the difference between the peak strain and
the non-recoverable strain. The increase of the MSCR %
recovery for the PAV-aged binder is mostly likely due to
 HOSSAIN ET AL. ON MSCR ASSESSMENT 517

TABLE 5 Analyses of results of MSCR tests conducted at different temperatures and aging conditions.

1 2 3 4 5 6 7 8 9 10
Source ID, Aging Condition, Jnr@0.1 kPa Jnr@3.2 kPa Jnrdiff Stress Sensitivity R100 R3200 Rdiff % Recovery (Meets MSCR
Testing temperature ( C) (kPa1) (kPa1) (%) (Meets AASHTO MP 19) (%) (%) (%) AASHTO TP 70) GRADE
PG 76-28
A3 (unaged,64) 0.07 0.07 5.90 YES 90.18 89.11 1.2 YES PG64-28E
A3 (PAVaged,64) 0.00 0.01 221.43 NO 98.31 94.55 3.8 NO PG64-28E
A7 (Unaged,64) 0.10 0.09 10.00 YES 92.17 91.90 0.3 YES PG64-28E
A7 (PAVaged,64) 0.01 0.00 160 YES 106.74 96.44 9.6 NO PG64-28E
A1 (RTFOaged,67) 0.02 0.02 3.23 YES 97.14 97.01 0.1 YES Undetermined
A7 (RTFOaged,67) 0.03 0.02 8.23 YES 95.70 95.43 0.3 YES Undetermined
A3 (RTFOaged,67) 0.10 0.10 2.56 YES 88.58 87.83 0.8 YES Undetermined
A7 (RTFOaged,70) 0.03 0.03 1.20 YES 95.8 94.9 0.9 YES Undetermined
A1 (RTFOaged,70) 0.03 0.03 3.25 YES 96.6 96.5 0.1 YES Undetermined
A3 (RTFOaged,70) 0.14 0.13 3.97 YES 87.2 86.2 1.2 YES Undetermined
PG 70-28
A3 (unaged,64) 0.99 1.63 64.84 YES 55.9 32.1 42.5 YES PG64-28H
A3 (PAVaged,64) 0.01 0.02 27.67 YES 90.5 87.9 2.9 YES PG64-28E
A1 (RTFOaged,67) 0.09 0.07 22.87 YES 93.3 94.1 0.8 YES Undetermined
A3 (RTFOaged,67) 0.54 0.84 55.70 YES 62.1 41.5 33.2 YES Undetermined
A1 (RTFOaged,70) 0.13 0.10 21.95 YES 92.5 93.1 0.6 YES Undetermined
A3 (RTFOaged,70) 0.86 1.40 63.19 YES 56.2 30.8 45.1 YES Undetermined

a very low non-recoverable strain of the binder, which is
expected as the PAV-aging process makes the binder
significantly stiffer than the RTFO-aged binder.
• It is expected that the additional test data (under different
aging conditions and MSCR test temperatures) will fur-
ther enhance the MSCR database, which will be useful
in formulating the MSCR test guidelines and grading
system.
REPEATABILITY AND REPRODUCIBILITY OF MSCR
TEST DATA
The repeatability of test results within the Materials Testing
Laboratory at the University of Oklahoma was ascertained by
testing replicate specimens. At least three specimens were tested
FIG. 6 MSCR % recovery vs Jnr @ 3.2 kPa at different testing temperatures.
for each test condition and the average of test results was
reported. The d2s% (difference two-sigma limit in %) values of
Jnr and MSCR % recovery of replicate specimens at 3.2 kPa
were found to be 5 and 9 %, respectively. A round-robin
study conducted by the NEAUPG reported d2s% values for
Jnr and MSCR % recovery at 3.2 kPa as 15.6 and 16.1 %,
respectively [5].
The reproducibility of test results between two selected lab-
oratories in the MSCR database was examined for PG 70-28
binders from A2. The results are presented in Table 6. It can be
noted that the maximum acceptable difference of a test parame-
ter between two laboratories recommended by ASTM D7405-
10a [33] is 18.1 %, whereas the differences in MSCR % recovery
values between two laboratories did not exceed 6 %. Therefore,
MSCR % recovery at 3.2 kPa is highly reproducible when the
same binder is tested by two independent operators. However,
high d2s% values of the Jnr at 3.2 kPa were observed for a major-
ity of the binders (9 out of 10), possibly due to a low Jnr value
(less than 0.1 kPa1). The ASTM D7405 method also predicts a
high variability of test results due to a significantly low value of
Jnr at 3.2 kPa. For PG 70-28 binders at 3.2 kPa, the NEAUPG
round-robin study also reported a high variation in Jnr
(d2s% ¼ 31.8 %) compared to the MSCR % recovery
(d2s% ¼ 23.9 %) [34]. Thus, due to the high repeatability and
reproducibility of MSCR % recovery data, tested PG 70-28
and PG 76-28 binders would still fall within the same quadrant
(PASS: meets both MSCR % recovery and ER criteria), indicat-
ing that the established MSCR % recovery limits for these
binders would still be applicable.
518 Journal of Testing and Evaluation

TABLE 6 Reproducibility of MSCR test data between two laboratories for PG 70-28 binders from A2.

Jnr,3.2 kPa MSCR %Recovery at 3.2 kPa

Multiple Lab Data OU Lab Obtained Acceptable Range of Two Multiple Lab Data OU Obtained Acceptable Range of Two
(ODOT Database) data d2s%a Test Results (ASTM D7405) (ODOT Database) Lab data d2s%a Test Results (ASTM D7405)

Lab 1 0.054 0.02 92 n/ab Lab 1 94.99 95.39 0 18.1

Lab 2 0.051 0.02 87 n/ab Lab 2 95.15 95.39 0
Lab 3 0.066 0.02 107 n/ab Lab 3 93.27 95.39 2
Lab 4 0.141 0.02 150 42.6 Lab 4 90.18 95.39 6
Lab 5 0.090 0.02 127 n/ab Lab 5 91.10 95.39 5
Lab 6 0.079 0.02 119 n/ab Lab 6 92.45 95.39 3
Lab 7 0.085 0.02 124 n/ab Lab 7 90.65 95.39 5
Lab 8 0.056 0.02 94 n/ab Lab 8 94.33 95.39 1
Lab 9 0.057 0.02 97 n/ab Lab 9 93.59 95.39 2
Lab 10 0.070 0.02 111 n/ab Lab 10 92.70 95.39 3
a
These limits represent the d2s% limits prescribed in Practice C670.
b
For Jnr values below 0.1 kPa1 high variability is likely due to the very low strain values that are measured.

Concluding Remarks • Furthermore, a decrease in MSCR % recovery with an

Three different types of commonly used binders (PG 64-22, PG
70-28, and PG 76-28) in Oklahoma were tested at 64 C and at
two stress levels (0.1 and 3.2 kPa). These asphalt binders were
obtained from 12 different sources throughout Oklahoma, and
the MSCR % recovery limits for the tested polymer-modified
binders were established. Based on the test results and discus-
sions of the current study, the following conclusions are drawn:
• Overall, the MSCR method can be used to characterize
high-temperature performance of PG binders for condi-
tions prevailing in Oklahoma.
• The MSCR % recovery is the largest for PG 76-28 from
all the sources followed by PG 70-28. A very low recovery
was found for PG 64-22 since it is not a polymer-
modified binder and its position is below the MSCR
curve.
• All tested PG 76-28 binder samples and a majority (78 %)
of PG 70-28 binder samples have been graded as PG
64-XXE, in accordance with AASHTO TP 70 [2]. On the
other hand, 91 % of PG 64-22 binder samples have been
graded by PG 64-XXS.
• The quadrant method can be used to formulate MSCR %
recovery guidelines. In the case of PG 70-28 binders, only
2 out of 11 suppliers are at risk for not meeting the target
MSCR % recovery of 50 %. The MSCR % recovery values
of binders from these two suppliers are about 45 %. There-
fore, an MSCR % recovery of 50 % can be adopted for PG
70-28 binders without putting many suppliers at risk.
•
In the case of PG 76-28 binders, none of the sources are
at risk for meeting the target MSCR % recovery of 55 %.
However, a value of 55 % for the MSCR % recovery
appears to be a conservative approach for conditions pre-
vailing in Oklahoma. Thus, a value of 80 % as the MSCR
% recovery is recommended without putting any supplier
or user at risk.
increase in temperature and an increase in MSCR %
recovery with an increased aging of the binder is
observed.
IMPLEMENTATION PLAN
• ODOT can incorporate the MSCR test method in
its quality assurance process to characterize the high-
temperature performance of RTFO-aged binders and
follow the MSCR grading system based on the aforemen-
tioned Jnr and MSCR % recovery criteria.
• Even though only a few suppliers are risk, it is suggested
that ODOT communicate the adoption plan of the
MSCR test method with all suppliers and establish some
mutually-agreed contingency plan in the event that their
binders do not meet the MSCR % recovery.
• It is also recommended that ODOT enhance the current
MSCR database with MSCR data of binders in use
for future construction projects and re-evaluate the
established guidelines periodically.
ACKNOWLEDGMENTS
The writers would like to express their sincere appreciation to
the Oklahoma Department of Transportation and Oklahoma
Transportation Centre for providing funding for this project.
The writers also want to be grateful to the refineries that are the
sources of the testing material. Also, the authors acknowledge
Mr. Danny Gierhart from the Asphalt Institute for his technical
assistance and Mr. Jackson Autrey, from the University of Okla-
homa, for his assistance in laboratory testing.
References
[1] AASHTO T315: Dynamic Shear Rheometer, AASHTO,
Washington, D.C., 2011.
[2] AASHTO TP 70: Standard Method of Test for Multiple
Stress Creep Recovery (MSCR) Test of Asphalt Binder

Using a Dynamic Shear Rheometer (DSR), AASHTO,
Washington, D.C., 2009.
[3] Anderson, M., “Using the Multiple-Stress Creep-Recovery
(MSCR) Test,” presented at the Association of Modified
Asphalt Producers Annual Meeting, Savannah, GA, Febru-
ary 2–3, 2010, AMAP, Avon, OH, -unpublished.
[4] Horan, B., “Multiple Stress Creep Recovery (MSCR) Task
Force,” presented at the Southeastern Asphalt User/Pro-
ducer Group (SEAUPG) Meeting, Savannah, GA, Novem-
ber 11–17, 2011, SEAUPG, Ridgeland, MS, -unpublished.
[5] Asphalt Institute Technical Advisory Committee, Guid-
ance on the Use of the MSCR Test With the AASHTO M
320 Specification, Asphalt Institute, Lexington, KY, 2010.
[6] D’Angelo, J. A. and Dongre, R., “Practical Use of Multiple
Stress Creep and Recovery: Characterization of Styrene-
Butadiene-Styrene Dispersion and Other Additives in
Polymer-Modified Asphalt Binders,” J. Transp. Res. Board,
Vol. 2126, 2009, pp. 73–82.
[7] Li, X., Clyne, T., Reinke, G., Johnson, E. N., Gibson, N.,
and Kutay, E., “Laboratory Evaluation of Asphalt Binders
and Mixtures Containing Polyphosphoric Acid,” J. Transp.
Res. Board, Vol. 2210, 2011, pp. 47–56.
[8] Zelelew, H., Paugh, C., Corrigan, M., and Belagutti, S.,
“Comparative Evaluation of MSCR Tests on Warm-Mix
Technologies,” presented at the 2nd International Warm-
Mix Conference, St. Louis, MO, October 11–13, 2011,
National Asphalt Pavement Association, Lanham, MD, -
unpublished.
[9] Gibson, N., Qi, X., Shenoy, A., Al-Khateeb, G., Kutay, M.
E., Andriescu, A., Stuart, K., Youtcheff, J., and Harman, T.,
“Performance Testing for Superpave and Structural Vali-
dation,” Report No. FHWA-HRT-11-045, Federal Highway
Administration, Washington, D.C., 2012.
[10] D’Angelo, J. A., “The Relationship of the MSCR Test to
Rutting,” J. Road Mater. Pave. Des., Vol. 10, 2009,
pp. 61–80.
[11] AASHTO M320: Standard Specification for Performance-
Graded Asphalt Binder, AASHTO, Washington, D.C.,
2010.
[12] AASHTO MP 19: Standard Specification for Performance-
Graded Asphalt Binder Using Multiple Stress Creep
Recovery (MSCR) Test, AASHTO, Washington, D.C.,
2010.
[13] Kabir, M. S., “Louisiana’s Experience with Multiple Stress
Creep Recovery (MSCR) Test,” presented at the Southeast-
ern Asphalt User/Producer Group (SEAUPG) Annual Meet-
ing, Baton Rouge, LA, November 11–14, 2013, SEAUPG,
Ridgeland, MS, -unpublished.
[14] Mehta, Y., Nolan, A., DuBois, E., Zorn, S., Batten, E.,
and Shirodkar, P., “Correlation Between Multiple Stress
Creep Recovery (MSCR) Results and Polymer Modifica-
tion of Binder,” Final Report No. FHWA-NJ-2014-002,
New Jersey Department of Transportation, Trenton, NJ,
2013.
[15] D’Angelo, J., Kluttz, R., Dongré, R., Stephens, K., and Zan-
zotto, L., “Revision of the Superpave High Temperature
Binder Specification: The Multiple Stress Creep Recovery
Test,” J. Assoc. Asphalt Pav. Technol., Vol. 76, 2007, pp.
123–162.
[16] Shirodkar, P., Mehta, Y., Nolan, A., Dahm, K., Dusseau, R.,
and McCarthy, L., “Characterization of Creep and
HOSSAIN ET AL. ON MSCR ASSESSMENT 519
Recovery Curve of Polymer Modified Binder,” J. Constr.
Build. Mater., Vol. 34, 2012, pp. 504–511.
[17] Asphalt Institute, 2014, “State MSCR Implementation
Status Database,” http://www.asphaltinstitute.org/public/
engineering/mscr_info_page_pdfs/MSCR%20Implementa-
tion%20Info%20Map.dot (Last accessed 6 Aug 2014).
[18] Santagata, E., Baglieri, O., Dalmazzo, D., and Tsantilis, L.,
“Evaluation of the Anti-Rutting Potential of Polymer-
Modified Binders by Means of Creep-Recovery Shear
Tests,” J. Mater. Struct., Vol. 46, No. 10, 2013,
pp. 1673–1682.
[19] Kadrmas, A., “Report on Comparison of Residue Recov-
ery Methods and Rheological Testing of Latex and Poly-
mer Modified Asphalt Emulsions,” presented at the
International Symposium of Asphalt Emulsion Technol-
ogy, Arlington, VA, 2008, Sponsor, Sponsor Location,
-unpublished.
[20] Zoorob, S. E., Castro-Gomes, J. P., Oliveira, L. A. P., and
O’Connell, J., “Investigating the Multiple Stress Creep Re-
covery Bitumen Characterization Test,” J. Constr. Build.
Mater., Vol. 30, 2012, pp. 734–745.
[21] Bahia, H. U., Hanson, D. I., Zeng, M., Zhai, H., Khatri,
M. A., and Anderson, R. M., “Characterization of Modified
Asphalt Binders in Superpave Mix Design,” Project No. 9-
10 FY’96, Transportation Research Board, Washington,
D.C., 2001.
[22] Wasage, T. L. J., Stastna, J., and Zanzotto, L., “Rheological
Analysis of Multi-Stress Creep Recovery (MSCR) Test,”
Int. J. Pave. Eng., Vol. 12, No. 6, 2011, pp. 561–568.
[23] Zelelew, H., Paugh, C., and Corrigan, M., “Warm-Mix
Asphalt Laboratory Permanent Deformation Performance
in the State of Pennsylvania: A Case Study,” presented at
the 90th Transportation Research Board Annual Meeting,
Washington, D.C., January 23–27, 2011, Washington,
D.C., TRB, unpublished.
[24] Hesp, S. A., Johnson, K. A. N., McEwan, R., Kumar, P.,
Samy, S., Ritchie, S., and Thomas, M., “Effect of Ten
Commercial Warm Mix Additives on the Quality and
Durability of Cold Lake Asphalt Cement,” presented at the
93rd Transportation Research Board Annual Meeting,
Washington, D.C., January 12–16, 2014, TRB, Washing-
ton, D.C., -unpublished.
[25] Morea, F., Marcozzi, R., and Castaño, G., “Rheological
Properties of Asphalt Binders With Chemical Tensoactive
Additives Used in Warm Mix Asphalts (WMAs),”
J. Constr. Build. Mater., Vol. 29, 2012, pp. 135–141.
[26] Livia, W. G. and McBroom, D. G., 2009, “Warm Mix
Asphalt,” ftp://ftp.mdt.mt.gov/contract/bid-packages/_PAST_
LETTINGS/2010_LETTINGS/02_FEB_11_LETTING/102_JCT_
MT_7-SOUTH/_UPDATED-020510_WMA-FINALV3.PDF
(Last accessed June 24, 2015).
[27] Anderson, R. M., “Southeast Asphalt User-Producer
Group Interlaboratory Study to Determine the Precision of
AASHTO TP70—The Multiple-Stress Creep-Recovery
(MSCR) Test,” Report Prepared for the Southeast Asphalt
User-Producer Group (SEAUPG), Asphalt Institute, Lexing-
ton, KY, 2011.
[28] Federal Highway Administration, “The Multiple Stress
Creep Recovery (MSCR) Procedure,” FHWA-HIF-11-038,
TechBrief, Office of the Pavement Technology, Federal
Highway Administration, Washington, D.C., 2011.
520 Journal of Testing and Evaluation

[29] AASHTO T316: Standard Method of Test for Viscosity Bending Beam Rheometer (BBR), AASHTO, Washington,
Determination of Asphalt Binder Using Rotational Vis- D.C., 2008.
cometer, AASHTO, Washington, D.C., 2010. [33] ASTM D7405-10a: Standard Test Method for
[30] AASHTO T240: Standard Method of Test for Effect of Multiple Stress Creep and Recovery (MSCR) of
Heat and Air on a Moving Film of Asphalt (Rolling Asphalt Binder Using a Dynamic Shear Rheometer,
Thin-Film Oven Test) (ASTM Designation: D 2872-04), ASTM International, West Conshohocken, PA, 2010,
AASHTO, Washington, D.C., 2009. www.astm.org
[31] AASHTO R28: Standard Practice for Accelerated Aging of [34] Anderson, M., “Northeast Asphalt User-Producer
Asphalt Binder Using a Pressurized Aging Vessel (PAV), Group Interlaboratory Study to Determine the Precision of
AASHTO, Washington, D.C., 2009. AASHTO TP70—The Multiple-Stress Creep-Recovery
[32] AASHTO T313: Standard Method of Test for Determining (MSCR) Test,” Project No. DTFH61-08-H-00030, Federal
the Flexural Creep Stiffness of Asphalt Binder Using the Highway Administration, Washington, D.C., 2011.

Copyright by ASTM Int’l (all rights reserved); Fri Jan 22 13:57:49 EDT 2016
Downloaded/printed by
Zahid Hossain (Arkansas State University, Civil Engineering, PO Box 1740, Assistant Professor, LSW#239, State University, Oklahoma, United States, 72467)
Pursuant to License Agreement. No further reproduction authorized.
View publication stats