Construction and Building Materials 411 (2024) 134648

Contents lists available at ScienceDirect

Construction and Building Materials

journal homepage: [Link]/locate/conbuildmat

An in-situ method to evaluate the interlayer bond between hot-mix asphalt

and portland cement concrete surface
Satyavati Komaragiri a ,∗, Thanos Drimalas a , Darren Hazlett a , Kevin Folliard b , Amit Bhasin b
a
Center for Transportation Research, The University of Texas at Austin, Austin, TX 78759, USA
b
Department of Civil, Architectural and Environmental Engineering, The University of Texas at Austin, Austin, TX 78712, USA

ARTICLE INFO ABSTRACT

Keywords: Proper bonding between different layers of a composite pavement is a major factor in ensuring that the entire
Tack coat structure acts as a monolithic layer under load. This paper studies the bonding between hot mix asphalt (HMA)
Asphalt overlay overlay and Portland cement concrete (PCC) pavement using tack coats. An in-situ pull-off test method was
Concrete pavement
developed for use on project sites on a routine basis to screen and test the quality of bond between asphalt layer
Direct shear test
and concrete layer. This test replaces the arduous process of coring through often times very thick concrete
Pull-off test
layer with steel rebars, transporting the core to a lab, trimming the concrete side of the core to fit the shear
jig, and finally conducting a shear test. Results show that the in-situ pull-off test method developed in this
study is strongly correlated to the laboratory based direct shear test. Limits were also suggested to indicate
an adequate bond strength in the PCC–HMA interlayer on the basis of laboratory and field testing. This study
also evaluates the influence of various factors on the performance of the interlayer bond in laboratory. These
factors include the type of tack coat, application rate, the texture of the concrete pavement surface, the surface
cleanliness and the surface moisture. Results indicate that sandblasting and hydro-demolition textures offer the
best interlayer bond strength, the influence of application rate is not significant on the interlayer bond strength
and the influence of surface moisture and cleanliness are dependent on the type PCC surface texture.

1. Introduction appropriate bonding between the PCC and HMA layers by utilizing the
most promising surface preparation techniques and materials. Based on
According to the American Society for Testing and Materials literature, poor bonding between the HMA and PCC layers appears to be
(ASTM), a tack coat can be defined as a light application of asphalt related to several factors, such as properties of the tack coat, presence
emulsion or liquid asphalt on an existing HMA or PCC layer to enhance of moisture in the concrete surface, cleanliness of the PCC surface and
the adhesion between the existing layer and the HMA overlay. The concrete surface preparation [6]. This study examines the influence of
primary function of a well-performing tack coat is to provide strong these factors on the interlayer bond strength in the laboratory setting.
adhesive bond between the existing pavement layer and the overlaid Many performance tests/apparatus have been developed in previous
surface so that the pavement does not experience rapid surface failures studies to assess tack coats. Overall, all these tests can be listed as
causing a reduction in service life. The traffic loads applied to the direct shear tests, pull-off tests, or torsional shear tests with variations
pavement source are transmitted through different layers within the
in loading and specimen size. Fig. 1 illustrates the different modes
pavement structure. Proper bonding between layers is crucial for the
of mechanical testing that are appropriate for the evaluation of tack
entire pavement structure to act as a monolithic layer under this
coats [7,8]. Out of these, direct shear tests performed in the laboratory
loading. It has been widely reported in the literature that the poor
has attracted most attention due to its relative simplicity, good repeata-
bond of a hot mix asphalt (HMA) with the underlying layers causes
bility, and quick results. For example, Florida direct shear test [9],
deformation or delamination of the HMA layer [1–3]. The failures
Louisiana interlayer shear strength tester [10] and NCAT shear test [11]
due to poor bonding include, but are not limited to slippage cracking,
top-down cracking, potholes, and fatigue cracking [4]. Even the mech- are direct shear tests that can test both laboratory fabricated samples as
anistic responses in the HMA layer such as tensile, compressive, and well as field cores. However, laboratory-prepared specimens are likely
shear stress are significantly influenced by the HMA–PCC interfacial to overestimate the actual interface shear strength in the field [4,12,
bonding conditions [5]. Therefore, it is of great significance to ensure 13]. Therefore, it is essential to conduct field measurements to evaluate

∗ Corresponding author.
E-mail address: [Link]@[Link] (S. Komaragiri).

[Link]
Received 23 June 2023; Received in revised form 22 November 2023; Accepted 15 December 2023
Available online 3 January 2024
0950-0618/© 2023 Elsevier Ltd. All rights reserved.
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

Fig. 1. Schematic showing test modes for evaluation of tack coats [7].

the bond strength of the interfacial layer. In this regard, pull-off tests are not developed in isolation at the HMA–PCC interface causing the
and torsional shear tests have an advantage as they can performed both shear failure. In fact, normal stresses are also present due to the traffic
in field as well as in laboratory. A pull-off test developed by University load and hence it is the combination of shear and normal stresses that
of Texas at El Paso (UTEP) is one such test [14]. However, the drawback exists during the shear failure (Fig. 2). Furthermore, preliminary shear
of this test is the inability to control the rate of loading which influences testing of samples showed that the use of normal stresses during the
its repeatability [15]. Similarly, torque bond test [16] and Oregon field direct shear test reduced the variability of the test results. For both
torque tester [17] are field tests that are performed in torsion. Due to these reasons, the shear test in this study was conducted by applying a
manual application of torque some of these tests show high variability normal force perpendicular to the direction of the shear. Fig. 3 shows
in results. a commercially available testing jig that allows for application of a
It can be observed from the aforementioned studies that there have normal stress while conducting the shear strength test. A spring was
been very limited attempts to develop a simple field test for in-situ used to apply the confinement pressure, and a dial indicator was used
evaluation of the quality of the tack coat. This study addresses this gap to read the displacement of the spring. The dial gauge was calibrated to
by developing a simple pull-off in-situ test method to assess the bond the spring constant and the compression in the spring can be converted
of HMA–PCC interlayer. to the applied normal stress. Based on the aforementioned studies
and recommendations, this work conducted the direct shear test in
2. Scope displacement-controlled mode at a rate of 5.08 mm/min (0.2 inch/min)
with a confining pressure of 68.95 kPa (10 psi) both for field samples
The goals of this study were to: (i) develop a test method that can and laboratory samples to take the normal forces due to traffic into
be used on a routine basis to screen and/or field-test the quality of the consideration. The confinement pressure of 68.95 kPa (10 psi) was
bond between the asphalt and the concrete layer; (ii) use this method selected based on the capacity of the load frame.
in tandem with a conventional direct shear test to evaluate the impact HMA–PCC samples were placed in the direct shear device such that
of various factors on the performance of the bond, including, type of the interface was in the middle of the gap between the two frames. A
tack coat, application rate, concrete surface texture (including cost- shear load at a constant rate of displacement was then applied along
effective and innovative ways to prepare concrete surfaces), cleanliness, with a constant confining pressure on the specimens. The test was
and surface moisture; and (iii) propose guidelines for future selection terminated when the displacement reached the maximum allowable
of surface preparation techniques and materials that will meet the displacement of the testing frame. After testing, the load and displace-
requirements for adequate bonding at the interlayer surfaces. This study ment data was analyzed to obtain the shear strength. Additionally, the
compares the developed pull-off test method with the direct shear test interface of the HMA–PCC layers was inspected for each specimen.
which is generally used to evaluate interlayer bond in a laboratory Although researchers fully understand that the rate of loading, mode
setting. Both laboratory and field testing were performed to provide of loading, temperature and confining stress influence the results, the
guidelines and recommendations. However, it must be noted that very focus of this study was a simple index test for QM and screening
limited field testing was performed and hence the recommendations are purposes.
mostly based on testing the samples in a laboratory setting.
3.2. Pull-off test
3. Test methods to evaluate the interlayer bond
In this approach, the pull-off strength of a tack coat is measured by
Two test methods were used in this study to evaluate the interlayer using a commercially available motorized portable direct pull-off test
bond strength both in field and laboratory specimens (i) direct shear device. The procedure developed to conduct this test is summarized
test, and (ii) pull-off test. Please note that all materials used in this below.
study were sourced from producers in and around the state of Texas in Specimens were partially cored from the HMA–PCC pavement to a
the United States. depth that reached approximately 3.175 to 6.35 mm (0.125 to 0.25
inch) into the PCC layer to expose the PCC–HMA bonding interface but
3.1. Direct shear test without fully extracting or separating the core from the PCC base. The
top surface of the HMA was glued to the metal plate attachment of the
Several previous studies have conducted the shear test. pull-off tester using a high strength epoxy and allowed to cure while
Table 1 summarizes the parameters used in these studies. placing a load on the metal plate to ensure proper bonding between
It can be observed from the table that many of these studies includ- HMA top surface and metal plate attachment after curing. A temper-
ing AASHTO and TxDOT specifications do not include the confining ature control system was used to condition the sample before testing.
pressure in their test procedure. However, in the field, shear stresses This temperature control system comprised of temperature controller

2
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

Table 1
Direct shear parameters used in previous studies.
Mode Rate Confining pressure
Song et al. [18] Displacement-controlled 50 mm/min (0.04 gal./sq. yd.) 137.90 kPa (20 psi)
Mohammad et al. [19] Load-controlled 222.41 N/min (50 lbs/min) –
Mohammad et al. [12] Displacement-controlled 2.54 mm/min (0.1 inch/min) 138 kPa (20 psi)
Tashman et al. [20] Displacement-controlled 50.8 mm/min (2 inch/min) –
Bahia et al. [21] Displacement-controlled 2.54 mm/min (0.1 inch/min) 48.26 kPa (7 psi)
aas [22] Displacement-controlled 2.54 mm/min (0.1 inch/min) –
Texas department of Displacement-controlled 5.08 mm/min (0.2 inch/min) –
transportation specification,
Tex [23]

Fig. 2. Schematic of shear loads applied on an asphalt specimen, (a) shear load only, (b) a combination of shear and normal loads.

in rapid HMA failure or have continued to perform very well over

several years. The field sections along with the material information
are summarized in the section below.

4.1. Field sections

Table 2 presents a summary of the identified field sections for this

study. Researchers were able to perform the in-situ pull-off test on all
sections. Researchers were also able to obtain 6-inch cores from the
remaining field sections in order to perform direct shear strength test.
Detailed test protocols for these two testing methods can be found in
Section 3 of this article.
Fig. 3. Tack coat shear strength tester.
4.2. Specimen coring

Researchers aimed at obtaining a minimum of four 152.4 mm (6

which is fixed in a wooden box. The wooden size was enough to cover
inch) diameter cores from all identified sections. However, the coring
an area of 22.86 cm 𝑥 22.86 cm (9 inch 𝑥 9 inch) on the pavement.
failed in Section 3 due to the deterioration of the bond between the
This area was enough to drill four 50.8 mm cores (Fig. 4). The shallow
HMA and the PCC layer during the operation (Fig. 5). From this loca-
depth and smaller diameter make it much easier to create these cores
tion, the crew obtained one 152.4 mm (6 inch) core and one 101.6 mm
for in-situ testing. A commercially available portable motorized pull-off
(4 inch) core specimen. Also, due to time and traffic constraints, only
device was attached to the metal plate attachment to pull the core out
one 152.4 mm (6 inch) specimen was cored in Section 5.
at the rate of 6.89 kPa/sec (1 psi/sec) in the direction perpendicular to
the bond layer to measure the tensile strength of the bond between PCC
and HMA. The specimen was then visually inspected for the location of 4.3. Procedure
failure, whether it occurred at the interlayer, at the HMA layer, or at
the glued interface between the metal plate and the HMA layer. Fig. 4 Prior to drilling the cores, the direction of the traffic was marked
shows the step-by-step procedure for the field pull-off test. A laboratory on the designated coring spot. The core barrel was then driven all the
version of this pull-off test was also developed in this study and is way to the bottom of the concrete layer. The full depth cores were then
discussed in Section 6.3. removed with a core extractor and packaged for transportation (Fig. 6).
A cold mix patch was used to fill the holes made from the coring
4. Field evaluation of interlayer bond operation. After transporting these cores to the laboratory, the concrete
side of the core was trimmed using a saw to achieve a 76.2 mm (3 inch)
One of the goals of this study was also to evaluate the performance thickness. This was necessary to ensure that the finished specimen fit in
of field sections that have shown issues in the past or sections that have the direct shear jig for testing. All cores were allowed to air dry at 25◦ C
shown no signs of any deformation/delamination after several years for at least two days after coring and sawing before they were used for
of application. Researchers reached out to maintenance engineers at direct shear testing. Pull-off test was performed using 50.8 mm (2 inch)
different districts in the state of Texas with the intent of identifying cores in the field. The procedure described in Section 3.2 was used to
tack coat projects and materials combinations that may have resulted conduct this test.

3
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

Fig. 4. In-situ pull-off test procedure (from top-left clockwise: using coring equipment to get four 5 cm (2 inch) diameter cores; drying and cleaning of cores using an air jet;
application of high strength epoxy to glue pull-off plate; use of a simple thermoelectric temperature control box to cure and condition the test specimen; pull-off test being
conducted on one of the four replicates; image after completion of all four replicate pull-off tests).

Table 2
Identified field sections.
Section Tack coat type Application Asphalt mix type Pull-off test Direct shear
rate (L/sq. m) test
AR Type II ****
Section 1 2.5 3.8 cm (1.5 inch) PFC ****
(likely as an underseal) ****
AR Type III *******
Section 2 2.5 2.54 cm (1 inch) TOM ****
(likely as an underseal) ****
Non-tracking ****
Section 3 – 3.8 cm (1.5 inch) TY D *
tack coat emulsion ****
∼7.6 cm (3 inch) ****
Section 4 N/A N/A ****
multiple layers ****
Section 5 N/A N/A ∼5 cm (2 inch) **** *

Note: ∗ indicates number of replicates, ∗ in 2 rows represent two different locations on the same section.
Acronyms: PFC — Permeable Friction Course; TOM — Thin Overlay Mix; TY D — Type D mixture according to TxDOT specifications; AR
Type II — Asphalt rubber binder type II according to TxDOT specifications; AR Type III — Asphalt rubber binder type III according to TxDOT
specifications.

strength decline. This indicates that the reason for the bond failure
(as discussed in Section 4.2) during the drilling operation to obtain
the cores for performing direct shear test was due to poor adhesion
between tack coat and the PCC surface. Direct shear results also showed
that Section 3 performed better than the cores from Sections 1 and 2
on which AR binders were used as a tack coat. This shows that non-
tracking tack coats perform better as compared to AR binders. It must
be also be noted here that the direct shear test results for Section 3 are
based only on one replicate.
It can be observed that for Section 1, 2 and 3, the direct shear
strength follows the same trend as pull-off strength. Sections 4 and 5
showed lower pull-off strengths compared to the direct shear strength.
By looking at the sheared surfaces, it was found that the underlying PCC
surfaces were highly uneven contributing to a higher shear strength.
This is the reason high shear strengths for these field sections was not
Fig. 5. Cores from Section 3 with failure at the interface during coring. reflected with high pull-off strengths. Documents providing the reason
of having these highly uneven PCC textures could not be found as these
sections were very old.
4.4. HMA-PCC interfacial layer evaluation Authors acknowledge that the data obtained from these field sec-
tions is not sufficient to observe a clear correlation between direct
The direct shear strengths and pull-off strengths obtained from
shear test and pull-off test. However, it can be stated based on the five
testing the five field sections are summarized in Fig. 7. At least four
sections tested in this study that most of these field sections exhibited
replicates were tested at each location. It can be observed that Sec-
tion 3 showed the highest pull-off strength among all the samples. This a pull-off strength higher than 138 kPa (20 psi) which can be used as a
section was repaired with a non-tracking tack coat emulsion. It must frame of reference for acceptable performance for routine QM testing.
be noted here that the failure was observed to be happened at the It must be noted that this pull-off strength reference is based on the five
interface of tack coat and the PCC surface in the pull-off test which sections used in this study and testing on more field sections is required
is an adhesive failure. The failure was followed by a rapid post-peak to obtain a more accurate pull-off strength reference.

4
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

Fig. 6. Procedure for obtaining 15.24 cm (6 inch) cores for the laboratory direct shear test (from left to right: coring equipment to core full depth cores; core extractor to extract
the cores; extracted cores from the field section).

• Polished: Samples were polished to achieve a worst-case scenario.

The intent was to replicate a rigid pavement surface that had
experienced polishing over time due to the action of traffic. After
seven days of wet curing the concrete disks, the samples were
removed from the moist cure room. The samples were ground
on a wet 100 grit polishing wheel. Each sample was ground for
5 min with a hand applied pressure. Samples that may have had
more surface irregularities from after casting took a few more
minutes to polish. After reaching the desired polish, the samples
were placed back into the moist cure room.
• Tined-polished: A set of samples were tined initially after casting
and then polished after seven days. The intent was to replicate
a rigid pavement surface that had tining immediately after con-
struction but had also experienced polishing over time due to
Fig. 7. Comparison of shear strength and pull-off strength for the five identified field the action of traffic. The tining took place shortly after the wood
sections.
trowel strike off. A quarter inch tine mark spaced every one inch
was created on the surface of the disk. The samples were then
Table 3
Test matrix to evaluate the influence of application rate.
wet cured for seven days which then followed the same procedure
Tack coat type Application rate
outlined in the polished section.
• Sandblasting: Sandblasting equipment was used to increase the
0.181 L/sq. m (0.04 gal./sq. yd.)
0.226 L/sq. m (0.05 gal./sq. yd.) surface texture of the concrete disks. After seven days of moist
Trackless tack #2
0.272 L/sq. m (0.06 gal./sq. yd.) curing, the samples were removed and placed on a table outside.
0.317 L/sq. m (0.07 gal./sq. yd.) A sandblasting gun with 70 grit blast sand was used to texture the
0.91 L/sq. m (0.2 gal./sq. yd.) surface of the disks. After sandblasting the surface, the samples
1.36 L/sq. m (0.3 gal./sq. yd.)
AR binder were placed back into the moist cure room.
1.81 L/sq. m (0.4 gal./sq. yd.)
2.26 L/sq. m (0.5 gal./sq. yd.) • Hydro-demolition: Hydro-demolition was used to remove the top
paste layer of each concrete disk. Hydro-demolition usually oc-
curs after the concrete has hardened with the use of high-pressure
water to remove layers of concrete. The concrete disks for this
5. Laboratory evaluation of various factors on the interlayer bond
project were subjected to hydro-demolition after casting. An hour
after casting the concrete disks, a water hose nozzle was used
A laboratory study was conducted as a part of this research to
at intermittent blasts to remove the top layer of concrete. This
examine the influence of various factors on the interlayer bond. The
procedure produced a hydro-demolition texture to the concrete
first step for this study was to create cylindrical PCC samples in the lab.
samples. The samples were demolded at 24 h and placed in the
These samples were intended to mimic the existing concrete pavement.
moist cure room for 28 days.
The next paragraph describes the PCC sample preparation process in
detail.
These PCC samples were used to evaluate interlayer bond in the
Four concrete surface textures were applied to concrete disks to
laboratory using two different test methods: (1) a direct shear test and
determine how well an asphalt overlay would adhere to the surface.
(2) a pull-off test. The direct shear test was intended to be used on
All of the samples began as 152.4 mm (6 inch) diameter concrete
laboratory-prepared specimens, as well as on field cores, to evaluate the
disks that were 50.8 mm (2 inches) thick. The concrete mixture de-
shear strength of the PCC–HMA bond. The pull-off test was developed
sign followed TxDOT Class P concrete specifications which included a
primarily as a tool to evaluate the quality of the PCC–HMA bond in
maximum 235.87 kg (520 lbs) of cementitious material with a water
to cement ratio of 0.45. This mix design was used for two different the field but without the need for taking cores and running the test
coarse aggregates (river gravel and limestone). The mixing procedure in the laboratory. However, in order to validate the efficacy of this
followed AST [24]. After mixing, concrete was cast into the disk molds. method, a version of this test was developed so it could be conducted on
The concrete was placed in two layers and rodded 25 times per layer. A laboratory samples and allow for comparison of results from the direct
wooden trowel was used to strike off the excess concrete. The samples shear and the pull-off tests.
were then covered with wet burlap and plastic to cure for 24 h. After These test methods were used to evaluate the four different factors
24 h, the samples were demolded and placed into the moist cure room. on the HMA–PCC interlayer bond strength: (i) tack coat type, (ii)
The following bullets summarize the process used to create different application rate, (iii) PCC surface texture, (iv) surface cleanliness, and
textures. (v) surface moisture.

5
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

Table 4
Four different PCC surface textures evaluated in this study.

5.1. Influence of tack coat type were evaluated in this task as shown in the Table 3. The application
rates were chosen based on field experience and previous studies [4,
Two types of tack coats are mainly used in the state of Texas: (i) 25,26]. The application procedure has been discussed in Section 6.1 in
trackless tack which is either in the form of an emulsion or it is hot detail.
applied on the existing pavement, and (ii) asphalt rubber binder (AR
binder) which is applied as a chip seal. These type of tack coats can also 5.3. Influence of PCC surface texture
be observed in the field sections discussed in Table 2. Hence, this study
included different trackless tacks (both were hot applied tack coats), In order to evaluate the influence of PCC surface structure on the
AR binder and a performance grade binder. In total, four different tack
quality of the bond, four different PCC surface textures were evaluated
coats were evaluated in this part of the laboratory study: (i) Trackless
using two different types of tack coats and one application rate: (1)
tack #1, (ii) AR Binder, (iii) Trackless tack #2, and (iv) PG 70–22.
tined-polished, (2) sandblasted, (3) hydro-demolition and (4) polished.
These materials were obtained from the producers in and around the
These textures are shown in Table 4 and a summary of the process used
state of Texas. Different tack coats were used while keeping all other
conditions constant to compare their performance. to produce these textures is presented in the previous paragraphs.

5.2. Influence of application rate 5.4. Influence of surface cleanliness

In order to evaluate the impact of application rate on the quality of The influence of cleanliness of PCC surface on PCC–HMA bond was
the bond, four different application rates using two different tack coats also examined as a part of the laboratory evaluation. Two different

6
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

Table 5
Clean and dusty conditions for different PCC surface textures.

cleanliness conditions were considered: (1) clean and (2) dusty. The procedures are discussed below. The main goal of the application was
clean surface was obtained by wiping the PCC surface clean with a to achieve a good bond between tack coat and PCC surface.
piece of dry paper towel prior to the application of the tack coat. In
fact, this was the procedure used for all specimens except when a dusty 6.1.1. Application of AR binder
surface was being evaluated by design. The dusty surface was obtained AR binder is applied as a chip seal on the PCC surface in this study.
by spreading 0.1 g of fines of igneous material (passing 75 μm sieve) This was done to emulate the same process that is used in the field
over clean PCC surface to achieve a surface coating of fines. when using an AR chip seal between concrete and HMA as a bonding
Table 5 shows clean and dusty surfaces for each PCC surface texture. layer. For the chip seal, precoated aggregates were created by mixing
Grade 4 igneous non-absorptive aggregates with 0.2% asphalt binder of
5.5. Influence of surface moisture grade PG 64–22. Fig. 8 shows the precoated aggregates obtained from
this procedure.
Three different surface moisture conditions were used to evaluate
On the PCC surface hot AR binder was applied. This was achieved
the impact of surface moisture on PCC–HMA bond: (1) wet, (2) moist,
by pouring the required amount of hot AR binder on the PCC surface
and (3) dry. Post fabrication all PCC samples were stored under water.
which was computed based on the surface area of the PCC surface. The
The three surface moisture conditions are based on the time that the
AR binder was spread on the surface using a spatula while the PCC
PCC sample was in air in the lab conditions prior to the application
cylindrical specimen is on the scale to make sure that the required
of the tack coat. Samples out in air for 2-4 h were considered as wet,
amount of AR binder is on the PCC surface after spreading with the
22-26 h as moist, and 7 days as dry.
spatula. The precoated aggregates were placed on the hot applied
6. Sample preparation and testing for laboratory evaluation binder so that approximately half the surface area was covered with
the precoated aggregate particles. Fig. 9 shows a PCC sample with chip
6.1. Application of tack coats seal applied on it.

Different tack coats were applied to the PCC samples as discussed in 6.1.2. Application of other tack coats
Section 5.1. Note that the PCC samples were stored under water after The three remaining tack coats: Trackless tack #1, Trackless tack
the samples were fabricated and cured. The time period for which the #2 and PG 70–22, were directly applied on the PCC surface using a
samples were kept in air after being removed from water was based spatula and a heat gun. The target amount of binder was weighed and
on the condition of the surface moisture, as discussed in Section 5.5. poured over the surface of the concrete. The heat gun and a spatula
The tack coat was applied after this time period. Tack coat application were used to spread the tack coat uniformly over the surface of the

7
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

Fig. 8. Precoated aggregates for creating chip seal.

Fig. 9. A PCC sample with AR binder as chip seal.

Fig. 10. A PCC sample with Trackless tack #2 applied on it.

concrete. Fig. 10 shows a PCC sample with Trackless tack #2 applied gun. This film was peeled off the parchment paper after it was allowed
on it. to cool. The film was applied on the hot surface of the PCC sample by
For hydro-demolition and sandblasted surfaces, a thin film of tack setting the film down carefully on the sample. After applying the tack
coat was applied on hot PCC surface, as these surfaces were very rough coat, the PCC sample was left overnight to cure.
to evenly spread the tack using a spatula. The PCC surface was heated
to approximately 160◦ C using a temperature-controlled heat gun for a 6.2. Compaction of asphalt mix on PCC sample
very short duration. Although the PCC surface is not heated in practice,
this was done to compensate for the cooling of the small amount of A thin overlay mix (TOM) was used as the surface mix. A large
binder used in a laboratory environment and to ensure a good bond sample of this mix was obtained from a hot mix plant in several small
between tack coat and PCC surface. The tack coat film was created by boxes and placed in cold storage. A sample of the loose asphalt mixture
pouring the desired quantity of hot tack coat on a parchment paper and (as needed) was heated in the oven for about two hours at 165◦ C.
spreading it in a circle of 150 mm diameter using a spatula and heat The concrete sample with the tack coat on its surface is placed in

8
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

Fig. 11. Procedure for preparation of a composite PCC–HMA sample for shear testing.

the Superpave gyratory compactor (SGC) mold. Note that the concrete and average values along with the difference in the two replicates are
specimen and the tack coat are at room temperature at this stage to reported in this study. The sample was then cleaned with water to
emulate field conditions. An amount of 2 kgs (4.4 lbs) of the HMA remove any debris from the drilling process followed by compressed
(preheated to compaction temperature) was placed on top of the tack air. The sample was dried overnight before the pull-off tester was glued
coat surface inside the mold. The sample was then compacted to 110 to the steel disk on top of the drilled asphalt surface. The glued sample
gyrations using the SGC using conditions as described in Tex [27]. was cured overnight before testing at a load rate of 6.89 kPa/sec (1
Fig. 11 shows the step-by-step procedure that was used to prepare psi/sec) with the pull-off tester to obtain the pull-off strength of the
laboratory specimens for testing. The prepared composite samples were PCC–HMA bond. Note that the field process uses an environmental
left overnight prior to testing. chamber to accelerate the drying and curing process to within one hour.
Fig. 14 shows the step-by-step procedure of performing the pull-off test.
6.3. Testing
7. Results from laboratory testing
Two test methods were used to evaluate the PCC–HMA interlayer
bonding for the laboratory specimens: (i) direct shear test (ii) pull-off This section presents the results from the direct shear test in terms
test. Three replicate specimens were created for testing each of the of shear strength and brittleness factor. For brevity, pull-off strength
aforementioned factors. Two of these specimens were used with the test results are compared to shear strength results for all scenarios at
direct shear test and one specimen was used for the pull-off test. the end of this section.
Direct shear test:
The PCC–HMA samples were tested for direct shear by placing the 7.1. Tack coat type
sample in the jig and applying a confining pressure of 68.95 kPa (10
psi) at the rate of 5.08 mm/min (0.2 inch/min) which was also used in Figs. 15 and 16 show the shear strength and the brittleness factor of
the testing of field cores. Fig. 12 shows the procedure for performing the four different tack coats, respectively. It should be noted here that
the direct shear test in the laboratory. dry, clean, and polished PCC surface condition was used to evaluate
From the direct shear test stress–strain curve, two parameters were these tack coats. Three observations can be made from these figures.
used to evaluate the bond between PCC and HMA: (1) shear strength, First, Trackless tack #1 has the highest pull-off strength as well as shear
and (2) brittleness factor. Shear strength was defined as the maximum strength followed by Trackless tack #2 whereas the AR binder has the
stress from the stress–strain response captured during the test. Brittle- least. Second, although Trackless tack #1 and Trackless tack #2 have
ness factor was defined as a ratio of shear strength and stress at the high strength they also have high brittleness factor showing less ductile
strain of 1.2 times the strain at the maximum stress. This factor ranges nature of failure whereas AR binder is more ductile in nature. Third, the
from 1 to infinity. A higher value of this factor implies low ductility variability in shear strength between replicates for all four tack coats
and a value close to 1 implies very high ductility. Fig. 13 shows these is very similar. These results show that trackless tack type of tack coats
two parameters in the stress–strain curve. Two replicates were used for have more strength and less ductility as compared to AR binder tack
the direct shear test and the average values along with the difference coats.
in the two replicates are reported in this study.
Pull-off test: 7.2. Application rate
A pull-off test was developed and used to evaluate the in-situ quality
of tack coat. A more detailed description of this test along with field test Two different tack coats (1) Trackless tack #2 and (2) AR binder,
results is provided in previous sections of this paper. One of the goals were evaluated for four different application rates as described in Sec-
of this task was to evaluate the accuracy of this method as a field test as tion 5.2. Dry, clean, and polished PCC surfaces were used to evaluate
compared to measurement of the bond strength using the direct shear the application rate parameter. Figs. 17 and 18 show shear strength
test. Hence a laboratory version of the field pull-off test was developed. and brittleness factor for the two tack coats, respectively. It can be
To perform the pull-off test on PCC–HMA laboratory samples, a 2- seen from these figures that the influence of the application rate is
inch diameter coring bit was used to core through the asphalt layer. The not significant for both types of tack coats in terms of shear strength.
coring continued into the concrete layer to a depth of approximately However, in the case of the Trackless tack #2, ductility improved with
3.175 to 6.35 mm (0.125 to 0.25 inches) to ensure that the asphalt an increase in application rate. In the case of AR binder, the brittleness
core was completely detached from its surroundings. The compacted factor is close to 1 for all four application rates showing a good ductility
sample was cored at two locations to obtain two replicate for testing for these four application rates.

9
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

Fig. 12. Procedure for direct shear testing in laboratory.

Fig. 13. A typical stress–strain curve obtained from direct shear test showing the two parameters (1) shear strength and (2) brittleness factor.

10
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

Fig. 14. Pull-off test procedure in laboratory setting.

Fig. 15. Shear strength for different tack coat types.

Fig. 17. Shear strength at different application rates for (top) Trackless tack #2,
(bottom) AR binder.

Fig. 16. Brittleness factor in shear for different tack coat types.

11
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

Fig. 18. Brittleness factor at different application rates for (top) Trackless tack #2,
(bottom) AR binder.
Fig. 19. Shear strength for different PCC surface textures for two different type of tack
coats: (top) Trackless tack #2, (bottom) AR binder.

7.3. PCC surface texture

The influence of PCC surface texture on PCC–HMA bond was eval-

uated for two different tack coats and two different application rates
for each tack coat. Clean and dry PCC surface conditions were used.
Figs. 19 and 20 show shear strength and brittleness factor, respectively.
Following observations can be made from these figures: (1) shear
strength is higher for Trackless tack #2 for all four textures as compared
to the AR binder, (2) higher average shear strength can be observed for
sandblasted and hydro-demolition surfaces, (3) hydro-demolition and
tined-polished surfaces were more ductile in shear as compared to other
surfaces, and (4) no significant difference can be observed between the
two application rates used across different PCC surface textures. Recall
that the same observation was also made in Section 7.2. In most cases,
sandblasting and hydro-demolition appear to be better performing PCC
surface textures in terms of both average strength and ductility.

7.4. Surface cleanliness

Dry PCC surfaces with four different textures were used to evaluate
the effect of surface cleanliness (1) hydro-demolition, (2) polished, (3)
polished for limestone coarse aggregate concrete mix, and (4) sand-
blasted. Two different tack coats were also considered at an application
rate of 0.27 L/sq. m (0.06 gal./sq. yd.) for Trackless tack #2 and
1.81 L/sq. m (0.4 gal./sq. yd.) for AR binder. Two surface cleanliness
conditions were considered as described in Section 5.4. Figs. 21 and
22 show shear strength and brittleness factor, respectively. It can be
observed from these figures that the influence of surface cleanliness
is different on different PCC surface texture. It can also be observed
that variability in replicates is different for different scenarios. Hence
showing no clear pattern of the PCC–HMA bond strength solely on the
basis of the cleanliness of the PCC surface and that its influence is also Fig. 20. Brittleness factor for different PCC surface textures for two different types of
dependent on the PCC surface texture. It is speculated that a slight tack coats: (top) Trackless tack #2, (bottom) AR binder.
increase in strength due to the presence of dust in some cases could

12
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

Fig. 21. Shear strength based on surface cleanliness for different PCC surface textures
for two different types of tack coats: (top) Trackless tack #2, (bottom) AR binder. Fig. 22. Brittleness factor based on surface cleanliness for different PCC surface
textures for two different types of tack coats: (top) Trackless tack #2, (bottom) AR
binder.

be due to the increased friction resulting from the dust at the interface.
However, it is still recommended that cleanliness should be as good as
can be obtained in the field, possibly by air blowing the surface before
tack coat application.

7.5. Surface moisture

Three surface moisture conditions were evaluated as described in

Section 5.5. Clean PCC surfaces with four different textures were used
to evaluate the effect of surface moisture (1) hydro-demolition, (2)
polished, (3) polished for limestone coarse aggregate concrete mix,
and (4) sandblasted. Two different tack coats were considered with
an application rate of 0.27 L/sq. m (0.06 gal./sq. yd.) for Trackless
tack #2 and 1.81 L/sq. m (0.4 gal./sq. yd.) for AR binder. Figs. 23
and 24 show shear strength and brittleness factor, respectively. It can
be seen from these figures the influence of moisture of interlayer
bond strength is different for different PCC surface texture. Also, the
variability between the replicates is different for different scenarios.
Hence it can be suggested that no clear pattern was observed solely
on the basis surface moisture and that its influence also varies with the
type surface PCC surface texture.

7.6. Pull-off strength vs direct shear strength

One of the major goals of this research was to compare the pull-
off test results with the direct shear test results. Although the pull-off
test is not intended for use in the laboratory, the intent of including
this test on laboratory samples was to allow for a direct comparison
between the direct shear test results and the pull-off test results so that
the latter could be used in the field.
Fig. 23. Shear strength based on surface moisture for different PCC surface textures
Fig. 25 shows a plot of pull-off strength obtained from the pull-off for two different types of tack coats: (top) Trackless tack #2, (bottom) AR binder.
test against the shear strength obtained from the direct shear test for

13
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

Fig. 25. Pull-off strength versus direct shear strength for all the combinations of
different parameters in this study.

are also present in the concrete layer making the coring process for
direct shear test more cumbersome. The new in-situ pull-off test method
mitigates these limitations of the direct shear test. A laboratory version
of direct shear test and pull-off test was also developed and used in the
study to evaluate the influence of various factors on the interlayer bond
strength. Following conclusions can be made from this study:

• A good correlation is observed between shear strength and pull-

off strength obtained from direct shear test and pull-off test,
respectively based on the laboratory version of these test meth-
ods. This means that a field pull-off test could be used to ensure
Fig. 24. Brittleness factor based on surface moisture for different PCC surface textures adequate bond of the HMA to the PCC. The field pull-off test is
for two different types of tack coats: (top) Trackless tack #2, (bottom) AR binder.
relatively much easier to carry out compared to a direct shear
and also the pull-off tests can be conducted in locations where it
is not feasible to obtain a full depth core such as bridge decks. The
all the combinations analyzed in this study. A good correlation with field pull-off tests can be conducted at locations that are suspected
some scatter can be observed between the two test methods and results to have a poor bond strength for forensic purposes and can also
from these test methods. The correlation is expected, as typically a
be used as a quality assurance measure on newly constructed
tack coat with a higher pull-off strength is likely to also have a higher
sections.
shear strength. Therefore, this correlation qualitatively validates the
• Data from both laboratory and field components using the shear
data reported in this study and shows the potential of using the pull-off
and pull-off tests suggests the minimum shear strength between
tester in the field to test the PCC–HMA bond. It is recognized that shear
the HMA and concrete pavement should be 200 kPa and corre-
strength also accounts for mechanical interlock and friction between
sponding pull-off strength of 138 kPa. Field testing also indicates
the two layers, which is not explicitly accounted for in the pull-off test
138 kPa as adequate, but more field sections need to be tested
which explains the slope value of 0.55 in the linear correlation between
both with the laboratory direct shear test and the newly devel-
the two tests. Also, this is possibly one of the reasons for the observed
oped in-situ pull-off test to get a correlation between the tests to
scatter in the plot resulting in an r-squared value of 0.69.
Nevertheless, the correlation from Fig. 25 shows that the in-situ test obtain a more accurate reference pull-off strength for acceptable
can be used for screening or QM purposes. Note that the in-situ pull-off performance.
test requires about 1.5 h to test four cores at one location whereas the • Amongst all the tack coats used in this study, Trackless tack #1
direct shear test requires full depth coring (sometimes unsuccessfully and Trackless tack #2 have higher pull-off strength and shear
due to steel), transportation to a lab, trimming and testing. It can also strength as compared to AR binder. However, AR binder is more
be observed from the figure that most of the samples with good bond ductile as compared to other tack coats.
were above 200 kPa shear strength which corresponds to 138 kPa pull- • The influence of application rate on strength is not significant.
off strength. This limit could be used as minimum bond strength for However, ductility improved for Trackless tack #2 with an in-
acceptable performance. Authors recognize that this limit is based on crease in application rate.
the test scenarios included in this study. More testing is required to • Sandblasted and hydro-demolition PCC surface textures are the
obtain a more accurate acceptable limit. best performing surface textures both in terms of strength and
ductility. In practice, these methods can be used to get better
8. Summary, conclusions and recommendations interlayer bond.
• The influence of PCC surface cleanliness and PCC surface moisture
In this study, a new pull-off test method was developed to test the on PCC–HMA bond depends on the type PCC surface texture of the
PCC–HMA interlayer bond strength. This test is performed on the site existing pavement. However, good construction practices should
and has the potential of replacing the existing direct shear test for be used to ensure surfaces of PCC are as clean and dry as possible.
routine testing of bond between PCC and HMA layer. In the existing
direct shear test, full depth cores need to be extracted then these cores This study recommends that for an overlay project over concrete
need to be transported to a lab and tested after trimming the concrete pavement, that one of two courses of action be chosen: (i) lab shear
layer to fit the direct shear jig. Moreover, in some locations steel rebars and pull-off analysis and (ii) Field pull-off test only.

14
S. Komaragiri et al. Construction and Building Materials 411 (2024) 134648

In the first course of action, a lab study should be conducted to Chuck Fuller for his help in accessing a TOM mix for overlay testing.
determine the best combination of materials to ensure an adequate Authors would like to acknowledge Dr. Zahra Sotoodeh-Nia for her help
shear test and correlate this with the pull off test. This would involve: throughout the project. Finally, the authors would like to acknowledge
(i) making concrete cylinders with a texture treatment similar to that of Mr. Si-Yoon Woo for his help with the laboratory testing, and Mr. Tyler
the overlay project concrete or obtain cores from the field that present Seay and Dr. Angelo Filonzi for helping develop and carry out the field
the real polished surface, (ii) applying the selected tack coat materials test procedure.
at the selected rate, (iii) using the Superpave gyratory compactor to
compact the project-type HMA on top of the samples, measuring the References
direct shear strength on a minimum of two samples using the test pro-
cedure discussed in this study and (iv) measuring the pull-off strength [1] Y. Hachiya, S. Umeno, K. Sato, Effect of tack coat on bonding characteristics
at interface between asphalt concrete layers, Doboku Gakkai Ronbunshu 1997
on a minimum of two samples using the test procedure developed in
(571) (1997) 199–209.
this project. The shear strength should be greater than 200 kPa and the [2] H. Ozer, I.L. Al-Qadi, Z. Leng, Fracture-based friction model for pavement
correlation of shear and pull-off can be used to set a minimum pull-off interface characterization, Transp. Res. Rec. 2057 (1) (2008) 54–63.
strength in the field. Field pull-off testing can be accomplished to verify [3] A. Institute, The Asphalt Handbook, in: Asphalt Institute manual series, Asphalt
tack coat efficacy. Institute, 2007, URL [Link]
[4] L.N. Mohammad, Optimization of Tack Coat for HMA Placement, vol. 712,
In the second course of action, a minimum of 138 kPa pull-off Transportation Research Board, 2012.
strength should be verified in the field using the test procedure de- [5] X. Hu, L.F. Walubita, Effects of layer interfacial bonding conditions on the
veloped in this study. The first alternative allows an optimization of mechanistic responses in asphalt pavements, J. Transp. Eng. 137 (1) (2011)
the tack coat, while the second should ensure that a minimum tack 28–36.
[6] Z. Leng, H. Ozer, I.L. Al-Qadi, S.H. Carpenter, Interface bonding between hot-mix
coat bond is achieved. It is recommended to follow good construction
asphalt and various portland cement concrete surfaces: laboratory assessment,
practices to obtain a good interlayer bond. The receiving concrete Transp. Res. Rec. 2057 (1) (2008) 46–53.
surface must be cleaned and free of debris before placing the tack coat [7] J. Wang, F. Xiao, Z. Chen, X. Li, S. Amirkhanian, Application of tack coat in
and hot mix asphalt. If the concrete surface is new (as in the case pavement engineering, Constr. Build. Mater. 152 (2017) 856–871.
when HMA on new concrete is required to match existing profile during [8] I.L. Al-Qadi, S.H. Carpenter, Z. Leng, H. Ozer, J. Trepanier, Tack coat optimiza-
tion for HMA overlays: Laboratory testing, Technical report, Illinois Center for
expansion), it is vital to ensure that any curing compounds from the
Transportation, 2008.
concrete surface are removed before placing the tack coat. [9] G.A. Sholar, G.C. Page, J.A. Musselman, P.B. Upshaw, H.L. Moseley, Preliminary
investigation of a test method to evaluate bond strength of bituminous tack coats
CRediT authorship contribution statement (with discussion), J. Assoc. Asph. Paving Technol. 73 (2004).
[10] L.N. Mohammad, A. Bae, M.A. Elseifi, J. Button, J.A. Scherocman, Evaluation of
bond strength of tack coat materials in field: Development of pull-off test device
Satyavati Komaragiri: Writing – review & editing, Writing – orig- and methodology, Transp. Res. Rec. 2126 (1) (2009) 1–11.
inal draft, Project administration, Methodology, Investigation, Data [11] Y. Hachiya, S. Umeno, K. Sato, Effect of tack coat on bonding characteristics
curation. Thanos Drimalas: Investigation, Data curation. Darren Ha- at interface between asphalt concrete layers, Doboku Gakkai Ronbunshu 1997
zlett: Writing – review & editing, Methodology, Investigation, Concep- (571) (1997) 199–209.
[12] L.N. Mohammad, A. Bae, M.A. Elseifi, J. Button, N. Patel, Effects of pavement
tualization. Kevin Folliard: Writing – review & editing, Supervision, In-
surface type and sample preparation method on tack coat interface shear
vestigation, Conceptualization. Amit Bhasin: Writing – review & edit- strength, Transp. Res. Rec. 2180 (1) (2010) 93–101.
ing, Visualization, Supervision, Project administration, Investigation, [13] W. Zhang, Effect of tack coat application on interlayer shear strength of asphalt
Conceptualization. pavement: A state-of-the-art review based on application in the United States,
Int. J. Pavement Res. Technol. 10 (5) (2017) 434–445.
[14] I. Deysarkar, Test Set-Up To Determine Quality of Tack Coat, The University of
Declaration of competing interest Texas at El Paso, 2004.
[15] D. Gierhart, D.R. Johnson, NCHRP Report: Tack Coat Specifications, Materials,
The authors declare that they have no known competing finan- and Construction Practices, (20-05/Topic 48-02) 2018.
cial interests or personal relationships that could have appeared to [16] I. Walsh, J. Williams, HAPAS certificates for procurement of thin surfacing,
Highw. Transp. (2001).
influence the work reported in this paper.
[17] A. Mahmoud, E. Coleri, J. Batti, D. Covey, Development of a field torque test
to evaluate in-situ tack coat performance, Constr. Build. Mater. 135 (2017)
Data availability 377–385.
[18] W. Song, X. Shu, B. Huang, M. Woods, Effects of asphalt mixture type on asphalt
pavement interlayer shear properties, J. Transp. Eng. B: Pavements 144 (2)
Data will be made available on request.
(2018) 04018021.
[19] L.N. Mohammad, M. Raqib, B. Huang, Influence of asphalt tack coat materials
Acknowledgments on interface shear strength, Transp. Res. Rec. 1789 (1) (2002) 56–65.
[20] L. Tashman, K. Nam, A. Papagiannakis, Evaluation of the influence of tack coat
The authors acknowledge the financial support of Texas Department construction factors on the bond strength between pavement layers, Technical
report, Washington State Department of Transportation Olympia, 2006.
of Transportation, United States, RTI project 7070. The authors ac-
[21] H.U. Bahia, A. Sufian, D. Swiertz, L.N. Mohammad, M. Akentuna, L. Bitu-
knowledge support from the project monitoring committee members, mix Solutions, et al., Investigation of tack coat materials tracking performance,
Aido Madrid, Arash Motamed, Enad Mahmoud, Marta Stephens, Travis Technical report, Wisconsin. Dept. of Transportation. Research and Library Unit,
Patton, and RTI project manager Tom Schwerdt. The authors thank 2019.
[22] AASHTO TP-114, Standard method of test for determining the interlayer shear
Thomas Blackmore, Brenda Guerra, Steven Schmidt, Jacob Wells from
strength (ISS) of asphalt pavement layers, 2018,
Austin District; Soojun Han and Ben Mebarkia from Houston District; [23] Tex-249-F, Shear bond strength test, 2019.
Andres Gonzalez, Diana Rogerio from San Antonio District; Kevin Gris- [24] ASTM C192, Standard Practice for Making and Curing Concrete Test Specimens
som, Kenneth Wiemers, Ashley Dubose, Mark Kelley, Joe Seago, John in the Laboratory, 2015,
Sudela, Keith Horn from Beaumont District; Bill Compton, Jeff Jackson [25] Techbrief, Tack coat best practices, 2016.
[26] D.S. Decker, Best Practices for Emulsion Tack Coats, National Asphalt Pavement
from Waco District; Lacy Peters from Atlanta District for their help
Association, 2013.
in coordinating with field testing. Authors would also like to thank [27] Tex-241-F, Compacting bituminous specimens using the superpave gyratory
compactor (SGC), 2019.

15