Construction and Building Materials 105 (2016) 391–399

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat

Measuring moisture susceptibility of Cold Mix Asphalt with a modified

boiling test based on digital imaging
Cheng Ling a,⇑, Andrew Hanz b, Hussain Bahia a
a
Department of Civil and Environmental Engineering, University of Wisconsin-Madison, Madison, WI 53706, United States
b
MTE Services Inc., Onalaska, WI 54650, United States

h i g h l i g h t s

A quick and simple modified boiling test utilizing digital imaging is introduced.

Results from the modified boiling test correlates well with BBS and TSR tests.

Use of modified emulsion can improve the moisture resistance of CMA significantly.

a r t i c l e i n f o a b s t r a c t

Article history: Cold Mix Asphalt (CMA) is a promising alternative material for a wide range of paving applications from
Received 6 August 2015 preventive maintenance and repair, to new pavement construction. However, the reported higher mois-
Received in revised form 11 December 2015 ture susceptibility of CMA relative to traditional Hot Mix Asphalt (HMA) presents a challenge that limits
Accepted 14 December 2015
the more widespread usage of CMA in the field up to date. To identify the potential of moisture suscep-
Available online 28 December 2015
tibility for CMA, a simple and effective test procedure based on modifications to the Boiling Test specified
in ASTM D3625 is proposed in this paper. The modified boiling test can provide a quick measurement of
Keywords:
moisture susceptibility for CMA in the material selection stage of the mix design. Digital imaging is used
Cold Mix Asphalt
Moisture susceptibility
to quantify the coating of CMA mixture samples before and after the boiling test instead of subjective
Modified boiling test visual assessment. The ‘‘Coating Ratio” parameter is proposed as an index to evaluate the coating quality
Digital imaging of CMA, and as a means to ensure adequate bonding between asphalt and aggregate. The results from the
CMA Mix Design modified boiling test were verified by AASHTO TP-91 Binder Bond Strength (BBS) test and the AASHTO
T-283 Tensile Strength Ratio (TSR) test of the compacted mixture. It is found that the results from mod-
ified boiling test correlated very well with those from BBS test and TSR test, indicating that the modified
boiling test can be used as an effective tool to evaluate the moisture damage potential of CMA.
Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction
Hot Mix Asphalt (HMA) is the most widely used construction
material for pavements in the United States as it provides good
quality and smooth ride. In recent years an emphasis on sustain-
ability has resulted in implementation of warm mix asphalt which
is similar to conventional HMA, but produced at lower tempera-
tures as a means to reduce energy consumption and emissions.
In another effort to reduce energy consumption of the production
phase of asphalt-based highway materials, research on improved
asphalt emulsion characterization and expanding possible
emulsified asphalt applications is being conducted. One of these
⇑ Corresponding author at: 3346 Engineering Hall, 1415 Engineering Drive,
Madison, WI 53706, United States.
E-mail address: cling4@wisc.edu (C. Ling).
applications is Cold Mix Asphalt (CMA) for use as a pavement layer.
CMA in this paper refers to the mixtures produced with virgin
aggregate (without RAP materials) mixed with emulsified asphalt
instead of hot asphalt binder, with possible environmental benefits
including reduced energy consumption and emissions due to stor-
ing the emulsified asphalt at ambient temperature and the ability
to use aggregates from the stock pile without drying [1]. Also, CMA
can be stockpiled and has a longer working life, meaning it can be
transported longer distances and placed in locations generally
inaccessible or impractical for more traditional methods.
Despite the potential environmental benefits of CMA, a number
of performance-related concerns need to be addressed before CMA
can be systematically applied in the field, one of which is the mois-
ture damage in the form of stripping due to the reduced bonding
between binder residue and aggregate caused by the low
temperature application of the asphalt emulsion in production,

http://dx.doi.org/10.1016/j.conbuildmat.2015.12.093
0950-0618/Ó 2015 Elsevier Ltd. All rights reserved.
392 C. Ling et al. / Construction and Building Materials 105 (2016) 391–399
which represents one of the ‘‘weak aspects” of CMA with respect to
the traditional HMA. A primary challenge associated with measur-
ing the moisture susceptibility of CMA is the presence of water in
the mixture within the wet aggregate and in the asphalt emulsion.
In both cases, the water slowly evaporates over time as the mixture
cures, which in turn affects the bulk mixture physical property and
mechanical response. As a direct result of prolonged exposure to
moisture, the CMA has higher potential in exhibiting early-life
distress relative to HMA.
In order to address the moisture susceptibility of the CMA
materials, a simple and effective evaluation method is needed to
support the selection of quality materials. A number of methods
have been successfully used to evaluate the moisture susceptibility
of asphalt mixtures [2–5] either on loose mixture or compacted
mixture for HMA. The boiling test specified in ASTM D3625 is a
well-established test on loose mixture to evaluate coating effec-
tiveness by visually assessing the percentage of coated area
remaining after boiling. To promote the application of boiling test,
however, a quantitative analysis is needed to better characterize
the effect of moisture on the coating to remove the subjectivity
of visual observation from the test method. In addition, no exten-
sive application of boiling test to CMA has been conducted due
to the lack of a specific and uniform sample preparation procedure.
Other than boiling test, the Binder Bonder Strength (BBS) test
developed by previous researchers [6–9] and currently specified
in AASHTO TP-91 provides a means to directly measure the ability
of asphalt binder to maintain its bond on an aggregate surface
under specific moisture conditioning. The BBS test has been suc-
cessfully applied in the prediction of the moisture susceptibility
of HMA and chip seal samples [10,11], indicating the bond strength
as a good indicator of moisture susceptibility of the asphalt
materials.
For compacted mixtures, the Tensile Strength Ratio (TSR) test
specified in AASHTO T283 is the most commonly used procedure
to evaluate the moisture resistance of HMA. The procedure
involves testing of the indirect tensile strength on dry and mois-
ture conditioned samples to assess the moisture resistance of the
mixture. Based on the empirical correlations to observations in
the field as well as the need to include moisture damage resistance
in mix design, the TSR was included in the SuperPave mix design as
a pass/fail criterion [12].
2. Research objectives
This paper is focused on developing a modified boiling test pro-
cedure that includes use of digital imaging to evaluate the coating
quality of the mixtures and identify the potential moisture suscep-
tible materials for CMA. The specific objectives are as follows:
1. Propose a modified boiling test procedure by providing a speci-
fic material preparation guideline for CMA and using digital
imaging to quantify the coating quality of the mixtures.
2. Verify that the analysis results from the modified boiling test is
an effective indicator of moisture susceptibility for CMA by
comparing the Ratio to results of the bond strength measure-
ments from BBS test and the Tensile Strength Ratios from
compacted mixture TSR test.
3. Materials
Both coating quantity and quality are dependent on the physical and chemical
properties of the individual components of the emulsion–aggregate system [13]. In
this study three different emulsion types and aggregates with different mineralogy
were selected, and the experimental matrix for the modified boiling test is
introduced in details below.
Two unmodified cationic emulsion CSS-1h and CSS-2 with different viscosities
(CSS-2 is higher) and one latex modified cationic emulsion CSS-1hL were selected to
evaluate the effect of modification of asphalt emulsion on coating quality. The resi-
due asphalt contents for CSS-1h, CSS-2 and CSS-1hL are 57%, 62% and 61%, respec-
tively. Both granite and limestone were included to provide both charge
compatibility and charge incompatibility conditions [13,14]. The residual asphalt
content of 2.0% was determined on the basis of the calculated specific surface area
of the aggregates with the asphalt film thickness of 8–10 microns [13]. Only coarse
aggregates with size of 3/800 were used in this study. Aggregate with SSD condition
was used to simulate the aggregate condition in the field. In this study the lime-
stone and granite aggregates had values of absorption of 1.53% and 0.40%, respec-
tively. All the three emulsion types and two aggregate sources used in the
modified boiling test were maintained in the BBS test as well.
To prepare the compacted mixture samples for mix design and TSR test, a speci-
fic gradation is required. In this study a gradation which meets the SuperPave E-10
requirements for a nominal maximum aggregate size (NMAS) of 12.5 mm was used,
the gradation curve on a 0.45 power chart is shown in Fig. 1. To maintain consis-
tency, the same limestone and granite sources as in the modified boiling test were
used and the SSD condition for aggregate before mixing was still applied. Only two
emulsion types CSS-1h and CSS-1hL representing unmodified and modified asphalt
emulsion respectively were included for the compacted mixture tests.
4. Test methods
4.1. Modified boiling test using digital imaging
In this study, the boiling test procedure specified in standard
ASTM D3625 was followed with modifications to the CMA specific
sample preparation and inclusion of digital imaging method [13] to
evaluate aggregate coating before and after the boiling instead of
visual assessment. The detailed procedure is described as follows.
A loose mixture sample of at least 500 g is cured at a specified
condition (as discussed in the next section) after mixing. Then
the sample is scanned using a standard office scanner to obtain
the image of sample coating before boiling. After being scanned,
the sample is placed in the 5000 ml-beaker filled with at least
3000 ml boiling water. A steel basket is used to prevent mixture
directly touching the bottom of beaker which has much higher
temperature than the boiling point due to the contact with the
heater. The boiling test apparatus is presented in Fig. 2. Based on
preliminary testing results, 10 min of boiling time (specified in
ASTM D3625) is insufficient to cause significant stripping after
boiling, therefore longer boiling time (20–60 min) was used to
expand the range of stripping. After boiling, the stripped asphalt
on the surface of water was skimmed and the basket with mixture
sample was removed from the beaker. The samples were spread on
silicone mats for drying at ambient temperature for at least 24 h
until sufficiently dry (no water on the sample surface observed)
for scanning. Then the dry samples are scanned again to obtain
the coating image after boiling.
Instead of using visual observation as specified in ASTM D3625,
the digital imaging method for evaluation of coating developed in
previous study [13] was followed to evaluate coating extent before
100
90
80
70
% Passing
60
50
40
30
20
10
0
0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00
Sieve Size (mm) Raised to the 0.45 Power
Max Denisty Control Points Gradation Used in TSR Test
Fig. 1. Gradations used in compacted mixture TSR test.
 C. Ling et al. / Construction and Building Materials 105 (2016) 391–399
Fig. 2. Boiling test apparatus.
and after boiling. Image processing software ‘iPas2’ developed in
the University of Wisconsin-Madison [15] was utilized in which
images of the mixture are converted to binary (black and white)
images and individual pixels are identified as coated (black) or
uncoated (white) based on their intensities, from which the coating
extent can be calculated. An example of using the digital imaging
tool to evaluate coating extent before and after boiling is presented
in Fig. 3.
The Coating Ratio (CR) was proposed as an indicator for evalu-
ating the stripping from the boiling test, and defined as follows:
Coating AfterBoiling
CR ¼  100%
Coating BeforeBoiling
4.2. Binder Bond Strength test
AASHTO TP-91 designates the sample preparation procedure
and testing method for the hot asphalt binder. In this study the
(a) Original image before boiling
(c) Original image after boiling
Fig. 3. An example of using digital imaging in evaluating coating extent before and after boiling.
393
same testing procedure was followed with minor modification in
the sample preparation. The asphalt emulsion was first recovered
using the procedure specified in ASTM D7497 Method B and the
emulsion residue was collected and molded, and then considered
as the regular hot asphalt in the following procedure of AASHTO
TP-91. Aggregate substrates were cleaned using ultrasonic cleaner,
dried and placed at 145 °C oven together with pull-out stubs for
conditioning for at least 45 min to reach the application tempera-
ture. Then the stubs were removed from the oven and the binder
sample was placed immediately on the stub for approximately
10 s, and pressed on the aggregate substrate firmly, as shown in
Fig. 4. Dry-conditioned samples were left dry at ambient tempera-
ture for 24 h before testing while wet-conditioned samples were
kept at ambient condition for 1 h and then placed in water bath
of 40 °C for 96 h. After conditioning, wet-conditioned samples
were kept at ambient temperature for 1 h to reach the testing tem-
perature prior to testing.
The BBS test was conducted at ambient temperature with load-
ing rate of 689.5 kPa/s using the test apparatus shown in Fig. 5. The
output of the test is the pull-off tensile strength (POTS), calculated
as:
ðBP  AgÞ  C
POTS ¼ ð2Þ
Aps
where Ag is the contact area of gasket with reaction plate (mm2), BP
is the burst pressure (kPa), Aps is the area of pull stub (mm2), and C
is the piston constant.
The Bonding Ratio (BR), defined as the indicator for evaluating
moisture susceptibility of emulsion–aggregate bond, is calculated
similarly to the Coating Ratio (CR) previously defined for the mod-
ified boiling test:
POTSWet
BR ¼  100% ð3Þ
POTSDry
ð1Þ
4.3. CMA Mix Design
To verify the modified boiling test results with the compacted
mixture TSR test, a mix design procedure needs to be established
to maintain consistent air voids within the samples and determine
the design emulsion content. Two emulsion types (CSS-1h and
CSS-1hL) and two aggregate sources (limestone and granite) were
(b) Binary image before boiling
(d) Binary image after boiling
394 C. Ling et al. / Construction and Building Materials 105 (2016) 391–399
Fig. 4. Application of asphalt binder on stub and aggregate substrate [10].
Fig. 5. Binder Bond Strength test apparatus [10].
included and four trial emulsion contents (7.25%, 8.00%, 8.75% and
9.50%) were used for this mix design. Note that only two emulsion
contents of 7.25% and 9.50% were used for CSS-1hL mixes. The
detailed mixture preparation and design process are described as
follows.
Aggregates were poured into mixing bucket and mixed with
pre-mixing water followed by emulsion of specified amount for
1.5–2 min by mechanical mixer. Mixtures were poured into the
mold and compacted to 100 gyrations using a constant vertical
pressure of 600 ± 18 kPa by SuperPave Gyratory Compactor (SGC)
which is shown in Fig. 6. A perforated compaction mold was used
in this study to prevent the build-up of hydrostatic pressure during
compaction by providing a path for water to escape.
After compaction, samples were cured with perforated
confining molds at 60 °C oven for 72 h and then left at ambient
temperature for another 24 h. Aluminum perforated confining
molds were used to prevent samples from collapsing immediately
after compaction and during the early stages of curing. After curing
the same procedure used for determination of Gmm and Gmb for
HMA as specified in ASTM D2041 and D6752 are used in this study.
The Gmm was run on the loose samples after 24 h curing at ambient
temperature as previous study [13] indicates that very little
moisture will be lost after curing at ambient temperature for
24 h. The Gmb was run on samples after 72 h curing using the same
procedure as HMA.
The standard Indirect Tension (IDT) test as specified in ASTM
D6931 was used to evaluate the effect of trial emulsion contents
on mixture quality after the volumetric properties of the cured
bulk samples were measured. Results were analyzed to provide
the basis for selecting the design emulsion content and the target
density for TSR test.
4.4. Tensile Strength Ratio test
As mentioned previously, the Tensile Strength Ratio test is one
of the most commonly used test for evaluating the moisture sus-
ceptibility of the asphalt mixtures. The standard AASHTO T283
was followed in this study with modifications to the target air
voids content of the mixture and the moisture conditions to
account for the use of CMA. For HMA the standard requires testing
of mixes at 7% air voids, which could not be achieved for CMA.
Therefore the target air voids content was selected on the basis
of the results from the mix design. In addition, a different wet con-
ditioning regime (exposure at 25 °C for 48 h) was included due to
the potential that the standard conditioning (exposure at 60 °C
for 24 h) could be too severe for CMA because of the high air voids
and potential incomplete curing as indicated in previous research
[13]. Two replicates were run for both dry and moisture condi-
tioned samples. No freeze–thaw cycle was included. The Tensile
Strength Ratio is calculated as follows:
Stm
TSR ¼  100% ð4Þ
Std
where:
TSR = Tensile Strength Ratio,
Stm = average tensile strength of the moisture conditioned sam-
ples, kPa,
Std = average tensile strength of the dry samples, kPa.
5. Results and discussion
5.1. Modified boiling test
The sensitivity of CMA to boiling time and curing condition was
assessed to determine optimum experimental conditions for the
modified boiling test. As part of the development of the procedure,
the proper curing condition for CMA loose mix prior to using the
boiling test was evaluated. A small study that included one emul-
sion (CSS-1h) and two aggregate types (limestone and granite) was
designed. The mixtures were subjected to the following curing
conditions:
1. Curing for 6 h at 60 °C;
2. Curing for 24 h at 60 °C;
3. Curing for 72 h at 60 °C.
The overall goal of further development of the boiling test is to
assess the potential for moisture susceptibility of fully cured in-
service CMA mixtures. Therefore, the most appropriate curing con-
dition that meets these practical requirements needs to be deter-
mined. To further develop the test procedure, boiling times of
20 min and 60 min were performed for the samples cured under
 C. Ling et al. / Construction and Building Materials 105 (2016) 391–399 395

Fig. 6. SuperPave Gyratory Compactor with perforated mold.

the second condition (24 h at 60 °C) to identify the effect of boiling

time on the extent of stripping. The test results are shown in Fig. 7.
By comparing the results under three curing conditions using
the same boiling time (60 min) it can be seen that there’s very little
change of CR after curing for 6 h at 60 °C. In addition, CR after boil-
ing for 60 min shows significant decrease compared with CR after
boiling for 20 min. Therefore in order to expand the range of strip-
ping for observation and analysis, curing time of 24 h and boiling
time of 60 min were selected for further study.
The complete modified boiling test results based on the pro-
posed testing conditions are presented in Fig. 8, in which ‘‘L” refers
to limestone while ‘‘G” refers to granite. The modified emulsion
CSS-1hL shows the best moisture resistance for both limestone
and granite, while CSS-2 with higher viscosity performs better rel-
ative to CSS-1h. Regarding to aggregate mineralogy, limestone
shows better moisture resistance than granite for both CSS-1h Fig. 8. Modified boiling test results.
and CSS-2; however the ranking reverses for CSS-1hL due to the
effect of modifier.
are presented in Fig. 9. The results demonstrate that dry POTS for
5.2. Binder Bond Strength test limestone has no significant difference with granite for all three
emulsions. However limestone is more moisture resistant than
The Binder Bond Strength test results under dry and moisture granite for CSS-1h and CSS-2 with higher BR values, but the rank-
conditions for all the three emulsions and two aggregate types ing reverses for CSS-1hL. The same trend was also observed in the

Fig. 7. Determination of curing condition and boiling time.

396 C. Ling et al. / Construction and Building Materials 105 (2016) 391–399

Fig. 9. BBS test results on emulsion residues.

Fig. 10. Air voids and ITS for CSS-1h and CSS-1hL mixes.
 C. Ling et al. / Construction and Building Materials 105 (2016) 391–399 397

Table 1 preparation for this study. The outcome of the procedure is selec-
Design emulsion content for emulsion–aggregate systems. tion of a design emulsion content (DEC) based on volumetric anal-
System DEC (%) ysis and performance tests. Samples were then prepared at the
CSS-1h limestone 7.00 design emulsion content to evaluate the effects of emulsion type
CSS-1h granite 7.00 and aggregate mineralogy on moisture susceptibility of CMA. The
CSS-1hL limestone 7.25 air voids and indirect tensile strength (ITS) for all mixes from the
CSS-1hL granite 9.00 mix design are shown in Fig. 10.
Results show that the air voids for CSS-1hL mixes are higher
than CSS-1h mixes, but keep decreasing with the increased emul-
sion content similarly as CSS-1h mixes. It can be clearly seen that
ITS also decreases with the increased emulsion content for all com-
binations of emulsion and aggregate. The emulsion content vs. ITS
presented in this study was inconsistent with the typical relation-
ship between asphalt content and ITS which was a curve and the
asphalt content that corresponds to maximum ITS strength was
defined as optimum. Based on the results it appears that the levels
selected only captured behavior on the ‘‘wet side” of the ITS curve.
The results and analysis indicate that even though samples
prepared at higher emulsion contents result in lower air voids, it
does not necessarily lead to improved performance. Based on these
results and the literatures which state that the typical air voids
observed in the field after 1–2 years of service ranges from 10%
Fig. 11. Conditioning of CSS-1h limestone samples following AASHTO T283. to 15% [16,17], 12% is selected as the target air voids in this
study. The corresponding emulsion contents for all emulsion–
Table 2 aggregate systems were therefore estimated as DEC as shown in
Results of TSR with modified conditioning. Table 1.
Emulsion Aggregate Conditioning Air Saturation ITS TSR
voids (%) (kPa)
5.4. Tensile Strength Ratio results
(%)
CSS-1h Limestone Wet 11.4 76.8 252.5 0.37
In this study a modified moisture conditioning with longer time
Dry 12.2 – 686.3
Granite Wet 11.9 72.6 148.3 0.28 and lower temperature was used to prevent the damage of samples
Dry 11.8 – 538.7 due to extreme conditioning as shown in Fig. 11. The testing results
CSS-1hL Limestone Wet 11.9 72.0 306.0 0.74 for both CSS-1h and CSS-1hL emulsion are shown in Table 2 and
Dry 11.8 – 415.1 Fig. 12.
Granite Wet 11.4 72.0 379.3 0.97 After conditioning, no crack or damage was observed for all
Dry 11.2 – 391.6 samples, and IDT test was performed on all samples. From
Fig. 12 it can be seen that CSS-1h mixes are not sound in moisture
modified boiling test. By comparing different emulsions, it can be conditioning with relatively low TSR values. Modified emulsion
seen CSS-1hL performs much better than CSS-1h and CSS-2 no CSS-1hL mixes significantly improves the TSR values for both
matter for limestone or granite; while CSS-1h has higher dry POTS aggregate types. Therefore, modification of asphalt emulsion is
but also increased moisture sensitivity relative to CSS-2. capable of improving moisture susceptibility of CMA.
Direct comparison of CMA performance to limits defined for
5.3. CMA Mix Design HMA is not possible due to the significantly higher air voids and
modified conditioning time/temperature used for CMA in this
Prior to evaluating performance it was first necessary to study. As a result the effects of air voids and conditioning time/
establish a mix design procedure to ensure consistent sample temperature are confounded with possible differences in material

Fig. 12. TSR results using modified conditioning.

398 C. Ling et al. / Construction and Building Materials 105 (2016) 391–399

Table 3
Summary of test results of modified boiling test, BBS test and TSR test.

System CR BR TSR
Emulsion Aggregate Value (%) Ranking Value (%) Ranking Value (%) Ranking
CSS-1h Limestone 74.00 5 53.60 5 37 3
CSS-1h Granite 70.50 6 45.80 6 28 4
CSS-2 Limestone 80.70 3 65.10 3 –
CSS-2 Granite 78.50 4 55.80 4 –
CSS-1hL Limestone 89.90 2 69.80 2 74 2
CSS-1hL Granite 96.70 1 97.50 1 97 1

compacted mixture TSR test. The main findings are summarized

as follows:

1. The proposed modified boiling test utilizing digital imaging is a

quick and simple test to evaluate coating quality of the mixture
and could be introduced in the lab or field in the future. The
results are comparable with those from the visual inspection
method.
2. Results of Coating Ratios from the modified boiling test corre-
late very well with BBS test results indicating the effectiveness
of the modified boiling test in evaluating effect of moisture on
bonding between emulsion and aggregates. With limited mix-
ture test data, the Coating Ratio correlated very well with the
TSR results of CMA testing, indicating its potential as a pass/fail
indicator that can be used in the materials selection phase of
mix design.
3. The low TSR values obtained for mixtures with unmodified
emulsion (CSS-1h) do not necessarily depends on the high air
voids content, but could be simply due to the low adhesion
properties between emulsion the residue and aggregate. By
using the modified emulsion (CSS-1hL), the moisture resistance
of CMA can be significantly improved.

The mix design procedure proposed in this study was based on

Fig. 13. Correlations between CR, BR and TSR.
performance. The typical air voids for moisture resistance test of
HMA is about 7% and the TSR requirements are between 70% and
80% depending on the state. Future research can be performed to
provide a direct comparison between HMA and CMA by maintain-
ing consistent air voids and testing conditions.
5.5. Comparison of CR, BR and TSR
The test results from the modified boiling test, BBS test and TSR
test are summarized and ranked in Table 3 for comparison. All
three indices from different tests shows consistent ranking of
moisture resistance for different materials. By correlating CR, BR
and TSR as shown in Fig. 13, it is seen that very good correlations
exist between CR, BR and TSR, indicating the effectiveness of the
modified boiling test as a pass/fail indicator for TSR of CMA.
6. Summary and conclusions
This paper include a proposal for a modified boiling test
procedure and using digital imaging to evaluate the coating quality
and moisture susceptibility of CMA. The results from the modified
boiling test were compared and verified using BBS test and
limited emulsion and aggregate types. To further validate the
potential of the CMA mixture performances, it is recommended
that further testing including use of a broader range of emulsion
contents be conducted to verify an optimum range of density
and moisture resistance performance. Also an expanded study
including more emulsion types, aggregate sources and gradations
are needed to verify the mix design procedure. Meanwhile, the
high design air void content could bring a challenge in application
of CMA which can lead to mixes that are more permeable, thus
more prone to moisture and aging. Future research is needed to
evaluate and address this issue.
Acknowledgments
This study is part of the Asphalt Research Consortium project
and support provided by the FHWA and WRI is gratefully acknowl-
edged by authors.
References
[1] P.M. Jarrett, A.N.S. Beaty, A.S. Wojcik, Cold-mix asphalt technology at
temperatures below 10 °C, J. Assoc. Asph. Paving Technol. (AAPT) (1984) 50–
97.
[2] D.-W. Cho, Study of Asphalt–Aggregate Bond in Simple Simulation Using the
Dynamic Shear Rheometer (Ph.D. thesis), The University of Wisconsin-
Madison, Madison, WI, 2008.
[3] H.U. Bahia, A. Hanz, K. Kanitpong, and H. Wen, Testing methods to determine
aggregate/asphalt adhesion properties and potential moisture damage, WHRP
07-02, Wisconsin Highway Research Program, Madison, WI, 2007.
[4] A. Bhasin, Development of Method to Quantify Bitumen–Aggregate Adhesion
and Loss of Adhesion Due to Water (Ph.D. thesis), Texas A&M University,
College Station, TX, 2006.
 C. Ling et al. / Construction and Building Materials 105 (2016) 391–399
[5] Y.R. Kim, D.N. Little, R.L. Lytton, Effect of moisture damage on material
properties and fatigue resistance of asphalt mixtures, Transp. Res. Rec. 1891
(2004) 48–54.
[6] K. Kanitpong, H.U. Bahia, Role of adhesion and thin film tackiness of asphalt
binders in moisture damage of HMA, J. Assoc. Asph. Paving Technol. (AAPT) 72
(2003) 502–528.
[7] F. Canestrari, F. Cardone, A. Graziani, F.A. Santagata, H.U. Bahia, Adhesive and
cohesive properties of asphalt–aggregate systems subjected to moisture
damage, Road Mater. Pavement Des. 11 (Suppl. 1) (2010) 11–32.
[8] R. Moraes, R. Velasquez, H.U. Bahia, Measuring effect of moisture on asphalt–
aggregate bond with the bitumen bond strength test, Transp. Res. Rec. V 2209
(2011) 70–81.
[9] F. Canestrari, G. Ferrotti, F. Cardone, A. Stimilli, Innovative testing protocol for
evaluation of binder-reclaimed aggregate bond strength, Transp. Res. Rec. V
2444 (2015) 63–70.
[10] R. Moraes, Validation of the Bitumen Bond Strength Test Using Surface Energy
Measurements (M.S. thesis), The University of Wisconsin-Madison, Madison,
WI, 2011.
399
[11] T.D. Miller, Development of Bond Strength Test for Improved Characterization
of Asphalt Emulsions (M.S. thesis), The University of Wisconsin-Madison,
2010.
[12] TRB Committee on Bituminous–Aggregate Combinations to Meet Surface
Requirements, Moisture sensitivity of asphalt pavements, in: A National
Seminar, February 4–6, 2003, San Diego, CA.
[13] C. Ling, R. Moraes, D. Swiertz, H. Bahia, Measuring the influence of aggregate
coating on the workability and moisture susceptibility of cold-mix asphalt,
Transp. Res. Rec. V 2372 (1) (2013) 46–52.
[14] M.I. Darter, S.R. Ahfield, P.L. Wilkey, and R.G. Wasill, Development of
emulsified asphalt–aggregate cold mix design procedure. Project IHR-505,
Illinois Cooperative Highway Research Program, February 1978.
[15] Available at: <http://uwmarc.wisc.edu/ipas-software-package>.
[16] B. Eckmann, J.J. Potti, M. Bourrel, P. Verlhac, C. Such, F.L. Calderon, and J.S. Polo,
Cold mix technology: the contribution, Report of OPTEL Project, 2001, pp.
6–15.
[17] Bitumen Emulsions, USIRF, France, 2008.