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

net/publication/327271250

Performance Evaluation of Whitetopping with Improved Design Practices in

Texas

Article in Journal of Testing and Evaluation · May 2019

DOI: 10.1520/JTE20170665

CITATIONS READS

0 161

3 authors:

Wujun Zhou Moon Won

Texas Tech University Texas Tech University
11 PUBLICATIONS 21 CITATIONS 63 PUBLICATIONS 442 CITATIONS

SEE PROFILE SEE PROFILE

Pangil Choi
Texas Tech University
31 PUBLICATIONS 93 CITATIONS

SEE PROFILE

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

Bond Strength View project

All content following this page was uploaded by Pangil Choi on 07 September 2018.

The user has requested enhancement of the downloaded file.

 Journal of Testing
and Evaluation
Wujun Zhou,1 Moon Won,2 and Pangil Choi2

DOI: 10.1520/JTE20170665

Performance Evaluation of
Whitetopping with Improved
Design Practices in Texas

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6 19:49:54 EDT 2018
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 Journal of Testing and Evaluation

doi:10.1520/JTE20170665 available online at www.astm.org

Wujun Zhou,1 Moon Won,2 and Pangil Choi2

Performance Evaluation of Whitetopping

with Improved Design Practices in Texas

Reference
Zhou, W., Won, M., and Choi, P., “Performance Evaluation of Whitetopping with Improved
Design Practices in Texas,” Journal of Testing and Evaluation
https://doi.org/10.1520/JTE20170665. ISSN 0090-3973

ABSTRACT
Manuscript received November 15, The design method of whitetopping in Texas is a method suggested by
2017; accepted for publication
the American Concrete Pavement Association (ACPA). However, one of the
February 20, 2018; published
online August 28, 2018. limitations on this particular design method is that an existing asphalt pavement
1
condition is not considered for slab thickness. Recently, the Texas Department of
School of Civil Engineering and
Mechanics, Huazhong University Transportation developed an improved whitetopping design procedure
of Science and Technology, 1037 considering mechanistic-empirical (ME) behavior. In the present study, performance
Luoyu Rd., Wuhan, Hubei 430074,
P. R. China (Corresponding
of whitetopping projects constructed by the ACPA design method in Texas were
author), e-mail: wujun.zhou@ investigated, and improved design details were suggested in order to minimize
hotmail.com, https://orcid.org/
distresses, such as corner break and slab movement, in transition areas. To
0000-0002-2747-9395
evaluate the structural performance of whitetopping, various tests, including the
2
Department of Civil and
falling weight deflectometer, ground penetration radar, and investigation of
Environmental Engineering, Texas
Tech University, Box 41023, concrete coring were conducted on US 90. According to the field survey and tests
Lubbock, TX 79409, USA results, most of the distress was caused by insufficient slab support conditions
related to asphalt concrete pavement materials and inadequate slab thickness
design. In addition, early-age behavior of whitetopping constructed by the
improved ME design method was evaluated and analyzed, and various gage
installations, such as vibrating wire strain gage, electric resistance-type strain
gages, porous nonstress cylinder, and impervious nonstress cylinder, are discussed.

Keywords
performance and structural evaluation, whitetopping, various pavement testing, support condition
evaluation

Copyright by ASTM
Copyright © 2018Int'l (all rights
by ASTM reserved); Thu
International, 100 Sep
Barr 6Harbor
19:49:54 EDTPO
Drive, 2018
Box C700, West Conshohocken, PA 19428-2959
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

Introduction
Deep rutting and shoving are typical distresses at asphalt concrete pavement (ACP)
intersection areas. Distresses at ACP intersections are more severe than those in nonin-
tersection areas. The main reason for this difference is the plastic deformations in asphalt
concrete under high stress concentrations that is caused by stop-and-go traffic loading
conditions at intersections. Repeated remedial actions, such as asphalt concrete overlay,
quite often do not mitigate the rutting and shoving problems, and portland cement
concrete (PCC) overlay, which is called whitetopping, provides an excellent rehabilitation
option for ACP intersections with rutting and shoving distresses. There are three types of
whitetopping in terms of PCC slab thickness: (1) conventional whitetopping (thickness
greater than 178 mm [7.0 in.]), (2) thin whitetopping (thickness between 102 to
178 mm [4.0 to 7.0 in.]), and (3) ultra-thin whitetopping (thickness of 51 to 102 mm
[2.0 to 4.0 in.]) [1]. In this study, the behavior and performance of all three types of
whitetopping in Texas were evaluated.
If designed and constructed properly, whitetopping could provide the following advan-
tages: (1) improved safety, (2) lower life cycle cost, and (3) optimum utilization of existing
pavement structures. Rutting and shoving at intersections pose serious safety hazards, espe-
cially under wet weather conditions, with increased potential for hydroplaning and reduced
stopping distance. Whitetopping with proper macrotexture, such as transverse tining, could
facilitate drainage and improve skid resistance. At the Texas Department of Transportation
(TxDOT), the design life of whitetopping is 10 years; however, evaluations of whitetopping
performance in Texas indicate a much longer performance period. In Texas, the average life
of asphalt pavement varies from 7 years to about 13 years before major rehabilitation is re-
quired. Any rehabilitation work at intersections causes traffic delays and increased potential
for accidents. Since whitetopping provides a longer performance period with a reasonable
initial construction cost, the life cycle cost of whitetopping would be lower than that of asphalt
pavement, especially when considering user delay costs that are caused by frequent repairs
and rehabilitations. Even though rutting and shoving distresses take place in ACP intersec-
tions, they do not necessarily represent inferior structural capacity of the pavement system,
and the existing ACP pavement with those distresses could provide an excellent support
system for concrete overlay. From a sustainability and environmental standpoint, the use
of existing asphalt pavement at intersection areas as a support system for PCC pavement,
instead of removing and replacing them, provides an optimum rehabilitation strategy.
Whitetopping in Texas: the first whitetopping project in Texas was constructed in 1999
in Abilene at the intersection of BI 20 and BU 83. Existing pavement structure consisted of
76-mm (3-in.) ACP on 230-mm (9-in.) asphalt stabilized base, which was placed on a
flexible base. This intersection experienced continued rutting and shoving problems, even
with regular maintenance activities. To address the rutting and shoving problems at this
intersection, the District decided to try whitetopping and developed design standards
and special specifications. Details of this whitetopping project are as follows:

(1) Project size: 3.65 m (12 ft) wide and 53.3 m (175 ft) long
(2) Slab thickness and joint spacing: 76 mm (3 in.) and 0.91 m (3 ft) by 0.91 m (3 ft),
respectively
(3) Saw-cut depth: 19 mm (3/4 in.)
(4) Concrete strength: 22.8 MPa (3,300 psi) at 18 hours and 43.8 MPa (6,350 psi) at
28 days

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

(5) Opening to traffic: 20 hours after concrete placement

(6) Total intersection closure time: 26 hours

The initial performance of this first whitetopping project was excellent because it
addressed rutting and shoving problems experienced in ACP at this intersection.
Based on the initial excellent performance of this project, additional whitetopping projects
were built at ACP intersections with rutting and shoving distresses. Meanwhile, there was a
need to improve the structural design of whitetopping as well as design details, especially
for transition areas since traffic levels and geometric conditions vary from intersection to
intersection. The TxDOT initiated a research project to develop whitetopping structural
design procedure [2]. Several whitetopping sections, designed in accordance with the
newly developed design procedure, were built, and their structural behavior and perfor-
mance have been closely monitored. This article presents general performance informa-
tion on whitetopping projects in Texas and in-depth discussions on structural responses
and behaviors of a whitetopping project that was monitored to validate the structural
models used for the development of new whitetopping design procedure.

Performance of Whitetopping Projects in Texas

WHITETOPPING PERFORMANCE EVALUATIONS
In the early 2000s, several whitetopping projects with different slab sizes and thicknesses
were constructed in Texas, as shown in Table 1. In the early 2000s, whitetopping projects
were designed with relatively small slab thickness and small sizes based on the concept of
“short slab sizes reduce bending and curling stresses” [3–5]. The American Concrete
Pavement Association (ACPA) recommended that the maximum joint spacing be about
12 to 15 times the slab thickness [3]. In Table 1, it is noted that all the projects followed this
recommendation.
Performance of the whitetopping projects was evaluated by periodic site visits under
the TxDOT rigid pavement database research project [6]. Even though overall perfor-
mance was quite excellent, several distinctive distress types were observed in isolated areas.
They include corner breaks and cracking, and slab slippage at transition areas.
Fig. 1 illustrates corner breaks and cracking observed in several projects. From a
design standpoint promoted by the ACPA, small slab sizes should minimize the poten-
tial for slab cracking, as curling and flexural stresses should be at a minimum. Distresses
shown in Fig. 1 indicate other distress mechanisms at work. Since dowels are not used in
whitetopping, if saw-cut joints are placed near the wheel path, portions of concrete slabs
near saw-cut joints could be in edge- or corner-loading conditions, which might result

TABLE 1
Whitetopping projects in Texas.

District Location Construction Year Slab Size, m (ft) Thickness, mm (in.)

Odessa Loop 250 (Midkiff Rd. to SH 349) 2001 0.91 × 0.91 (3 × 3) 76 (3)
Abilene BU 83D & N. 2nd St. Intersection 2003 0.91 × 0.91 (3 × 3) 102 (4)
Childress US 287 & FM 91 Intersection 2004 1.52 × 1.52 (5 × 5) 127 (5)
Odessa Loop 250 (SH 158 to Midkiff Rd.) 2005 0.91 × 0.91 (3 × 3) 76 (3)
Paris Loy Lake Rd. & US 75 2011 1.83 × 1.83 (6 × 6) 152 (6)
Paris US 69 & SH 19 Intersection 2012 1.83 × 1.83 (6 × 6) 178 (7)

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 1 Whitetopping condition survey: (a) BU 83D & N. 2nd St. Intersection, (b) US 287 & FM 91 Intersection, and (c) distresses on Loop
250 (SH 158 to Midkiff Rd.).

in corner breaks and cracking. Quite similar distress types were observed in Minnesota
and Illinois where 0.91 (3 ft) by 0.91-m (3-ft) joint spacing was used [7,8].
Another distress type observed was slab sliding at transition areas, as shown in
Fig. 2. In whitetopping design, it is assumed that full bond exists between the asphalt

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 2 Slab movement that was due to slab sliding at transition areas: (a) sliding on SH 226 whitetopping at transition area, and
(b) sliding on SH 36 whitetopping at transition area.

and concrete. It is a reasonable assumption since cement paste from fresh concrete
penetrates into voids in the asphalt surface. Coring of concrete slabs shows that the
asphalt base is attached to the concrete, which validates this assumption. The sliding
of whitetopping slabs against asphalt appears to take place within the asphalt layer.
Also, this sliding distress at transition areas indicates the need for improved designs
for transition areas.
In Table 1, the project in the Laredo District represents a unique pavement system,
where the asphalt layer was omitted and the concrete was placed directly on an aggregate
base, which is called a flexible base in Texas. In a strict sense, this pavement structure does
not meet the definition of whitetopping; however, the performance evaluations of this
pavement type could provide valuable information on the behavior and performance
of whitetopping without an asphalt layer.

IMPROVED SAWCUT DESIGN

Performance evaluations of whitetopping in Texas indicated the relevance of saw-cut joint
locations relative to wheel paths. Since dowels are not used in whitetopping, it would be
best practice to place saw-cut joints away from wheel paths. Accordingly, the ideal location
of longitudinal saw-cut joints should be at the middle of a lane width (1.83-m [6-ft] joint
spacing), which will minimize the corner breaks. As for the transverse saw-cut joints, from
a geometric standpoint, the same joint spacing as that used for longitudinal joints (1.83 m
[6 ft]) will minimize the potential for cracking. At the same time, wheel loading will subject
transverse saw-cut joints to edge-loading conditions, increasing the potential for slab
cracking. To minimize this cracking potential, slab thickness needs to be increased as long
as geometric conditions allow. Since the project size of whitetopping is usually not exten-
sive, as it is usually built at intersection areas, increasing slab thickness adds slightly more
to a project cost. Another important aspect of saw-cut design is the angle of the cuts. The
ideal angle of saw cuts is 90°; however, it is not always possible to provide a 90° saw-cut
angle, especially near the intersection curves. In that case, efforts should be made to pro-
vide obtuse angles instead of acute angles. Fig. 3a presents an ideal saw-cut layout as no
acute angle saw cuts were made. On the other hand, saw-cut design shown in Fig. 3b

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 3 Improved saw-cut design: (a) adequate saw-cut design, and (b) inadequate sawcut practice.

(a) (b)

presents an unideal practice as saw-cuts were made at an acute angle. If saw cuts at acute
angles cannot be avoided, it is best practice to place them away from wheel paths.

IMPROVED TRANSITION AREA DESIGN

It has been observed that pavement distress in the form of bumps occurred at juncture
areas of PCC and asphalt pavements; to minimize distresses, various designs were tried
with not much success because substantial changes exist in the stiffness of the pavement
structures at a juncture area, resulting in different behaviors under environmental and
traffic loading. It is expected that one of the main reasons for whitetop slab sliding at
transition areas is low friction between whitetopping slab and ACP. The second reason
for sliding could also be impact loading caused by repetitive traffic loading applications.
The best way to provide high friction between whitetopping slab and ACP would be to
apply different slab thickness at near transition areas and provide rough surfacing by cold
milling. Impact loading that is due to traffic could also be minimized by providing a thicker
slab at the transition area. Fig. 4a shows the pavement condition at the transition area
before placing concrete on ACP. As shown in this figure, cold milling was performed at the
transition area to construct a thicker slab. Fig. 4b illustrates a design detail of the tran-
sition area in whitetopping project on SH 69 and SH 19 intersection. This project was built
in April 2012. A pavement condition survey was conducted to validate an improved
transition area design, as shown in Fig. 4c and d. No distress or slab sliding were observed
at the time of the pavement condition survey. This project was annually monitored for the
long-term performance evaluation of improved design.

Early-Age Behavior of Whitetopping

LOCATION OF TEST SECTION AND GAGE INSTALLATION
Field testing was conducted to monitor early-age behavior at the intersection between US
69 and SH 19 in the Paris district. There were a number of rutting and shoving distresses in

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 4 Improved design detail at transition area and its performance: (a) existing ACP condition, (b) improved design detail
at transition area, (c) condition at transition area (SB), and (d) condition at transition area (EB).

(a)

(b)

No movement No movement

(c) (d)

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 5 Gages installation: (a) crack inducer installation, (b) VWSGs installation, (c) concrete strain gages installation, and (d) PNC and
INC installation.

Various Gages Installation VWSGs Installation

Joint 3
VWSGs

Joint 2
Electric
Gages

Joint 1
PNC and INC
Joint 3 Joint 2

(a) (b)
Electric Gages Installation PNC and INC Installation

Joint 1

Joint 2 Joint 1

(c) (d)

the existing ACP, as shown in Fig. 4a. Various gages were installed to evaluate the
early-age behavior of whitetopping that was due to environmental loading.
Crack inducers (thin steel plates) were installed at Joints 1–3, as illustrated in Fig. 5a.
Vibrating wire strain gages (VWSGs) and electric resistance-type concrete strain gages
were embedded at different depths in both longitudinal and transverse directions, as
shown in Fig. 5b and c. A porous nonstress cylinder (PNC) and impervious nonstress
cylinder (INC) were installed 25 mm (1.0 in.) below the surface of the whitetopping slab,
as shown in Fig. 5d. Two different types of NCs were employed. One was an INC, which
does not allow moisture exchange to or from the surrounding concrete. Thus, the strain
measured from the concrete inside the INC presents solely thermal strain at the depth of
installation. The other was PNC, which was designed to allow moisture exchange with the
surrounding concrete through the holes in the tube surface. Accordingly, the strain moni-
tored from the concrete inside the PNC is the sum of drying shrinkage and thermal strain.
By subtracting the strain measured by the INC from the strain measured by the PNC, the
drying shrinkage strain over time can be estimated [9]. Fig. 6 illustrates a bird’s-eye view
and cross section view of installed gages in the field.

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 6 Installed gages in field.

COEFFICIENT OF THERMAL EXPANSION AND IN SITU DRYING SHRINKAGE

OF CONCRETE
The two main factors affecting joint pop or crack are early-age drying shrinkage and
the thermal deformation of concrete. Fig. 7 illustrates the relationship between the con-
crete thermal strain and corresponding temperature measured by VWSG inside the INC.
The result indicates that the coefficient of the thermal expansion of concrete, evaluated by
the slope of temperature-strain curve, was about 11.2 με/°C (6.2 με/°F).
The drying shrinkage of concrete evaluated from PNC and INC is presented in Fig. 8.
The drying shrinkage history can be estimated by subtracting the strain measured by INC
from that measured by PNC. It is noted that most drying shrinkage is developed within 7
days after concrete placement. Thereafter, the development of drying shrinkage tended to
stabilize and follow daily temperature variation. The drying shrinkage developed within 50
days after concrete placement was measured at about 50 με, which is relatively small.

JOINT BEHAVIOR RELATED TO CRACK DEVELOPMENT AT JOINT

The variations of concrete strain at the top, middle, and bottom are presented in Fig. 9a–c,
respectively. Fig. 9a and b demonstrate that the first crack occurred in the test section
during the night time on the sixth day when concrete temperature decreased. After the first
cracking on May 3, the daily variations of concrete strain significantly increased at the top
and middle. The daily concrete strain variations at the bottom, however, were not com-
parable to those at the middle and top. Also, when cracked, the concrete slab curled up,
exhibiting the fact that the concrete strain at the top went into compression whereas that at
the bottom went into tension. The concrete strain fluctuation at the middepth was quite

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 7
Concrete coefficient of thermal
expansion.

FIG. 8
Concrete drying shrinkage
evaluated from PNC and INC.

small. This is probably because the position of the gage was close to the neutral axis. The
concrete strain at the middle moved to the compression side when the crack occurred,
which indicates that the neutral axis was located slightly below the middepth because
of its positive bond with the base layer. Based on the information obtained from the gages,
it appears that the crack occurred at the location of Joint 1, which is close to the electric
strain gages and slightly further from the VWSGs. When a crack occurred on May 3, only
two VWSGs close to Joint 1 moved to the tension side, as shown in Fig. 9c. The concrete
gages placed between Joints 1 and 2 did not detect the crack occurrence. Moreover, the rest

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 9
Concrete strain variations
(before and after crack):
(a) concrete strain variation at
top, middle, and bottom in
longitudinal direction near Joint
1 (measured by electric strain
gages), (b) concrete strain
changes near Joint 1 when the
first crack was detected
(measured by electric strain
gages), and (c) concrete strain
changes near Joint 2 when the
first crack was detected
(measured by VWSGs).

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

of the VWSGs that were 2.7 m (9 ft) away from Joint 1 did not capture any concrete strain
changes, most likely because of the good bond condition between the whitetopping slab
and existing asphalt pavement. It appears that the good bond condition made the concrete
strain that was due to cracking diminish rapidly, as it moves away from the crack location.
About 40 days after construction, both electric stain gages and VWSGs detected addi-
tional crack occurrences. As shown in Fig. 10a, the behavior of VWSGs near Joint 3 was
quite similar to that of the electric stain gages near Joint 1 at the time of first cracking; the
top concrete element went into compression while the bottom concrete element went into
tension. The electric stain gages near Joint 1 also captured the concrete strain changes that
were due to additional cracking, as can be seen in Fig. 10b. However, the concrete strain
variations were not as significant as those at the first cracking at Joint 1. Fig. 10c illustrates
the variations of concrete strain at the bottom along the longitudinal position when the
second crack occurred. When the second crack occurred, most of the bottom concrete
elements between Joints 1 and 3 had moved to the tension side, whereas some bottom
concrete elements experienced compression mainly because of the temperature drop.
Based on the behavior observed in bottom concrete elements at the time of the second
cracking, it was found that the crack inducer at Joint 2 malfunctioned, and the center of
two slab segments between Joints 1 and 3 was located at the midslab between the adjacent
Joints 2 and 3, as clearly indicated in Fig. 10c. During the 34 days between occurrence of
the first and second cracks, the concrete elements close to Joint 1 experienced 34 contrac-
tion and expansion cycles that were due to daily temperature variations, which may reduce
the friction between the concrete slab and existing asphalt pavement near Joint 1. The
concrete elements with less friction between asphalt pavement move to the tension side
much more easily when cracking took place. This behavior mechanism resulted in the
second crack at the location of Joint 3 on June 6th, 2012. The condition of cracks at
Joints 1, 2, and 3 shown in Fig. 11 provides information regarding the joint cracking se-
quence. No crack was found at Joint 2. The crack at Joint 1 was wider than that at Joint 3.
This finding demonstrates that the crack at Joint 1 took place earlier than that at Joint 3.
This pavement has performed well without transverse crack within whitetopping slab.
Faulting at joints had not been observed until the time of the pavement condition survey.
This implies that 178-mm (7-in.) slab thickness and relatively large saw-cut spacing are
suitable to support traffic without any distresses. Pavement conditions will also be annually
investigated in this test section.

Whitetopping Performance Evaluations

PAVEMENT CONDITION SURVEY ON US 90 WHITETOPPING PROJECT
Previous studies have stated that support conditions are critical to rigid pavement behavior
and performance [10–12]. Quite satisfactory performance of whitetopping projects in
Texas led to the design and construction of another whitetopping project on US 90 in
December 2011. This project was different from other whitetopping projects in that it
was a new pavement construction project, where a new ACP layer was designed as a sup-
port for the PCC layer. However, during the construction of this project, field changes were
made and the ACP layer was taken out. The actual pavement structure built consisted
of 102-mm (4-in.) concrete placed on a flexible base with varying thicknesses, without
asphalt layers in between. In Texas, the flexible base consists of aggregates and low PI
materials without any stabilizing agents. Accordingly, from a technical standpoint, this

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 10
Concrete strain changes related
to joint pop: (a) concrete strain
changes near Joint 3 when the
second crack was detected
(measured from VWSGs),
(b) concrete strain changes
near Joint 1 when the second
crack was detected (measured
from electrical gages), and
(c) concrete strain changes at
bottom when the second crack
was detected (1 in. = 25.4 mm).

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 11 Joint condition survey (April 2013): (a) overall joint condition, (b) Joint 1, (c) Joint 2, (d) Joint 3, and (e) Joint 4.

(a)

(b) (c)

(d) (e)

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

pavement is not a whitetopping project since there is no ACP under the concrete slab. On
the other hand, there has not been a universally accepted minimum asphalt slab thickness
for whitetopping. The range of 51 to 102 mm (2 to 4 in.) of asphalt slab is the minimum
thicknesses suggested. Accordingly, this project constitutes one extreme of whitetopping as
far as slab support is concerned. The details of this project are as follows:

(1) Highway and location: US 90, 69 km (43 miles) west of Del Rio over the Pecos River
(2) Pavement structure: 102-mm-thick (4-in.-thick) PCC slab on a flexible base
(3) Project length: 122 m (400 ft) (61 m [200 ft] in the east and west side of a bridge)
(4) Joint layout: 1.83 by 1.83 m (6 by 6 ft) in main lanes and 1.37 by 1.83 m (4.5 by 6 ft)
in shoulders

Distresses in the form of slab cracking developed within a few years after the
pavement was opened to traffic. Fig. 12a and b shows the pavement condition of
the east and west side of the bridge, respectively. It is observed that most of the distresses
are in the westbound lanes (35 distressed slabs out of a total 132 slabs), with the east-
bound lanes in a better condition (10 slabs showing minor distresses out of a total of 132
slabs). It was also observed that four areas in the eastbound lanes were repaired, as
indicated by the green lines in Fig. 12b. It is interesting to note that most of the distresses
in the westbound lanes were in the inside lane, where truck traffic must be lower than in
the outside lane.

STRUCTURAL EVALUATIONS (FALLING WEIGHT DEFLECTOMETER, GROUND

PENETRATING RADAR, AND CORING)
To identify the causes of distress and unusual behavior of this pavement (inside lane
more distressed), deflection testing with Falling Weight Deflectometer (FWD), coring,
and nondestructive testing with Ground Penetrating Radar (GPR) was conducted.
Fig. 13a illustrates slab deflections in three lanes (two lanes in the westbound and
one inside lane in the eastbound) in the west approach section. It is noted that (1)
deflections in the eastbound lane are much smaller than those in the westbound lanes,
(2) there is no appreciable difference in deflections between inside and outside lanes in
the eastbound lanes, and (3) deflections at distressed areas in the eastbound lane are
larger than in other areas. Average deflections in the westbound and eastbound lanes
in the west approach section were 152 (6 mils) and 76 μm (3 mils), respectively, which
are quite small considering the design slab thickness of 102 mm (4 in.). According to the
Westergaard equation for interior loading condition, deflections of about 254 μm
(10 mils) are expected with a modulus of subgrade reaction of 81.4 MPa/m (300 psi/in.),
which is the value used for the new PCC pavement design. The small deflections in non-
distressed areas indicate adequate slab support even without the asphalt concrete layer
under the concrete. On the other hand, even considering the effect of smaller slab sizes at
distressed areas, quite large deflections in those areas indicate deteriorated slab support.
Also, a rather large difference in deflections between eastbound and westbound lanes
indicates significant differences in either slab thickness or slab support condition. As
will be discussed later, coring shows a large difference in slab thicknesses between
two different directions.
Fig. 13b illustrates slab deflections in three lanes (two westbound lanes and an
inside eastbound lane) in the east approach section. As in the west approach section,
it is noted that (1) deflections in the eastbound lane are smaller than those in the west-
bound lanes, (2) unlike in the west approach section, rather large differences in

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 12 Visual pavement condition survey: (a) east approach and (b) west approach.

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 13 FWD testing and coring: (a) west approach and (b) east approach.

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 13 Continued

deflections are observed between inside and outside lanes in westbound lanes, and (3)
deflections at distressed areas in westbound lanes are larger than in other areas. Average
deflections in the westbound and eastbound lanes in the east approach section were 203
(8 mils) and 102 μm (4 mils), respectively. These deflections are much lower than what
would be expected of PCC pavement with small slab thicknesses, which indicates overall
adequate slab support. Smaller deflections in the eastbound lane than in westbound

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

TABLE 2
Deflection and slab thickness at coring locations.

Core No. Distress Deflection, μm (mils) Thickness, mm (in.)

C1 Shattered slab 533 (21) 95 (3.75)

C2 Transverse crack 229 (9) 102 (4.0)
C3 Shattered slab 281 (15) 95 (3.75)
C4 None 76 (3) 165 (6.5)

lanes, which was also observed in the west approach section, might be due to larger slab
thicknesses in the eastbound lanes.
A total of four cores were taken, two each in east and west approach sections. Cores
taken at locations C1 and C2 in the westbound inside lane at east approach section, both at
distressed slabs, show slab thickness of less than the design thickness of 102 mm (4 in.), as
shown in Fig. 13a. Cores taken at the other two locations in the east approach section
(C3 in the distressed slab in the westbound inside lane and C4 in the nondistressed slab
in the eastbound inside lane), as indicated in Fig. 13b, indicate deficient slab thickness in
the westbound lane, while more than 165-mm (6.5-in.) slab thickness was observed in the
eastbound lane.
Deflections and slab thickness information at the four coring locations are summa-
rized in Table 2. This summary clearly illustrates the effect of the distressed condition of
the slab (slab dimension and potentially slab support condition) and slab thickness on
deflections.
Fig. 14 shows the base condition at the C2 location where the deflection was a little bit
over 203 μm (8 mils). Small voids were observed at the top of the base layer, which could
have been caused by washing out of fines that were due to moisture or coring action. It
indicates vulnerability of this type of base under thin concrete slabs because of moisture.
Subgrade condition was evaluated by GPR testing in both east and west approaches.
As shown in Fig. 15a and b at severe and minor distressed areas, respectively, voids at a

FIG. 14
Support condition of
whitetopping slabs.

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 15 East approach GPR results: (a) section with severe distresses, and (b) section with minor distresses.

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

FIG. 16 West approach GPR results (sections with severe distress and minor distress).

depth of about 0.91 m (3 ft) were detected, and the possibility of underground water was
identified in the east approach. In the west approach, GPR testing was focused on the
distressed area along a 24-m (80-ft) section. As shown in Fig. 16, voids were also detected
at 1.5-m (5-ft) depth with possible underground water.

SUMMARY OF PAVEMENT PERFORMANCE

Findings from the FWD, coring, and GPR testing to identify the causes of distresses in this
project where thin concrete slabs were placed without an asphalt concrete layer, can be
summarized as follows:

(1) Overall deflections on thin concrete slabs placed directly on a flexible base (aggregate
base) were small, indicating adequate slab support. However, in distressed areas, de-
flections were rather large and small voids were observed on top of the flexible base
potentially because of the washout of fines by moisture. Since joints are not sealed in
whitetopping, which could allow water to get into the base layer, the use of a more
durable asphalt concrete layer is needed under thin concrete slabs.
(2) The increase of slab thickness from 95 mm (3.75 in.) to a little over 165 mm (6.5
in.) on an unstabilized flexible base resulted in reduced slab deflections and better
performance, which indicates the sensitivity of whitetopping performance to slab
thickness.

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

Conclusions and Recommendations

In this article, the performance and improved design practices of whitetopping in Texas
are presented. Behavior and performance of whitetopping projects, which were con-
structed by the ACPA design method and the Texas mechanistic-empirical method, were
investigated. The key findings were obtained as follows:

• In Texas, corner breaks near wheel paths and slab sliding in transition areas are two
main distress types in whitetopping constructed according to the ACPA design
method.
• Improved Texas whitetopping design details include increased saw-cut joint spacing
and thicker slab thickness at the transition area.
• Whitetopping behavior under environmental loading and field coring indicate a
good bond between the concrete slab and existing ACP.
• Long-term performance of whitetopping is sensitive to slab thickness and support
conditions.

ACKNOWLEDGMENTS
The support and assistance from Mr. Ruben Carrasco, Professional Engineer of the Texas
Department of Transportation, are much appreciated. This study was conducted under
research project (Development of Eco-Friendly Pavements to Minimize Greenhouse
Gas Emissions) funded by the Ministry of Land, Infrastructure and Transport (MOLIT)
and the Korea Agency for Infrastructure Technology Advancement (KAIA). The authors
would like to thank the members of the research team, MOLIT, and KAIA for their guid-
ance and support throughout the project.

References
[1] Texas Department of Transportation, “Rigid Pavement Design – Section 8: Thin
Concrete Pavement Overlay (Thin Whitetopping),” Pavement Manual, Texas
Department of Transportation, Austin, TX, 2017, pp. 8–28~30.
[2] Suh, C., Kim, D., and Won, M., Development of the Thickness Design for Concrete
Pavement Overlays over Existing Asphalt Pavement Structures, Technical Report
FHWA/TX-09/0-5482-2, Texas Department of Transportation, Austin, TX, 2008,
122p.
[3] American Concrete Pavement Association, “Whitetopping–State of the Practice,”
Engineering Bulletin EB210.02P, American Concrete Pavement Association,
Skokie, IL, 1998.
[4] Mack, J., Hawbaker, L., and Cole, L., “Ultrathin Whitetopping: State-of-the-Practice
for Thin Concrete Overlays of Asphalt,” Transp. Res. Rec., Vol. 1610, 1998, pp. 39–43,
https://doi.org/10.3141/1610-07
[5] Roesler, J. R., Bordelon, A. C., Ioannides, A., Beyer, M., and Wang, D., Design and
Concrete Material Requirements for Ultra-Thin Whitetopping, Technical Report
FWHA-ICT-08-016, Illinois Center for Transportation, Rantoul, IL, 2008, 175p.
[6] Ryu, S., Saraf, S., and Won, M. C., Project Level Performance Database for Rigid
Pavements in Texas, Part II, Report No. FHWA/TX-11-0-6274-1, Texas
Department of Transportation, Austin, TX, 2013, 92p.
[7] Burnham, T. and Rettner, D., “Whitetopping and Hot-Mix Asphalt Overlay
Treatments for Flexible Pavement: Minnesota Case History,” Transp. Res. Rec.,
Vol. 1823, 2003, pp. 3–10, https://doi.org/10.3141/1823-01

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
 ZHOU ET AL. ON PERFORMANCE OF WHITETOPPING

[8] King, D. and Roesler, J., Structural Performance of Ultra-Thin Whitetopping on

Illinois Roadways and Parking Lots, Technical Report FHWA-ICT-14-018, Illinois
Department of Transportation, Springfield, IL, 2014, 100p.
[9] Choi, S. and Won, M. C., “Thermal Strain and Drying Shrinkage of Concrete
Structures in the Field,” ACI Mater. J., Vol. 107, No. 5, 2010, pp. 498–507.
[10] Chen, D. H., Won, M., Chen, X., and Zhou, W., “Design Improvements to Enhance
the Performance of Thin and Ultra-Thin Concrete Overlays in Texas,” Constr. Build.
Mater., Vol. 116, 2016, pp. 1–14, https://doi.org/10.1016/j.conbuildmat.2016.04.093
[11] Chen, D. H. and Won, M., “Field Performance Monitoring of Repair Treatments on
Joint Concrete Pavements,” J. Test. Eval., Vol. 36, No. 2, 2008, pp. 119–127, https://
doi.org/10.1520/JTE101362
[12] Chen, D. and Lin, H., “Effects of Base Support and Load Transfer Efficiency (LTE) on
Portland Concrete Pavement Performance,” J. Test. Eval., Vol. 38, No. 1, 2010,
pp. 47–56, https://doi.org/10.1520/JTE102010

Copyright by ASTM Int'l (all rights reserved); Thu Sep 6of

Journal 19:49:54 EDT
Testing 2018
and Evaluation
Downloaded/printed by
Texas Tech Univ (Texas Tech Univ) pursuant to License Agreement. No further reproductions authorized.
View publication stats