Early-Age Structural Properties of Base
Material Treated with Asphalt Emulsion
Tyler Quick and W. Spencer Guthrie
The objectives of this research were to investigate the early-age structural
properties of base materials stabilized with asphalt emulsion and to
assess the rate at which emulsion-treated base (ETB) design properties
are achieved. Three experimental sections were established along a
pavement reconstruction project near Saratoga Springs, Utah. Field
tests were performed to assess the structural properties of the ETB
immediately after construction and at 2, 3, 7, and 14 days; 4 months; and
1 year. Measured values were plotted against time to determine trends
in ETB strength development and to determine pavement capacity.
Modulus values were consistently low in all three sections during the
first 2 weeks after construction, increased dramatically by 4 months,
and then decreased considerably by 1 year. In the first 2 weeks after
construction, the average ETB structural coefficient was 0.04. Only two
of the three sections reached the design structural coefficient of 0.25,
which occurred after approximately 3 months; however, the average
structural coefficient measured for all three sections after 1 year of
curing, which included a winter, was only 47% of the design strength.
The results of this research showed that, while pavement capacity was
sufficient at 4 months, it was severely reduced during the first 2 weeks
and at 1 year. Trafficking under these reduced capacities would be
expected to greatly compromise long-term pavement performance.
For this reason, trafficking of materials similar to those investigated
in this research is not recommended during at least the first 2 weeks
after construction.
Over the past few decades, full-depth reclamation (FDR) has become
increasingly prevalent in the transportation industry as a means of
rehabilitating and reconstructing flexible pavements (1, 2). The FDR
process involves recycling the existing pavement structure by pul-
verizing and mixing the in-place asphalt layer into the upper portion
of the existing base to form a new base layer. FDR provides a feasible
solution to problems such as pavement geometry restrictions, lack
of quality aggregate, and the cost of asphalt disposal (3); however,
the reclamation process can cause a reduction in the strength of the
base layer because it both disturbs the existing base material and mixes
reclaimed asphalt pavement (RAP) into it. The addition of RAP has
been shown to decrease the bearing capacity of some base materials
and, possibly, to impede proper compaction of the base material
(4, 5). For these reasons, stabilization is often needed to improve the
structural properties of reclaimed base materials (5). One product
that has been used in road stabilization since the early 1900s is
Department of Civil and Environmental Engineering, Brigham Young University,
Provo, UT 84602. Corresponding author: W. S. Guthrie, guthrie@byu.edu.
Transportation Research Record: Journal of the Transportation Research Board,
No. 2253, Transportation Research Board of the National Academies, Washington,
D.C., 2011, pp. 40–50.
DOI: 10.3141/2253-05
40
asphalt emulsion (6). The application of asphalt emulsion is an easy
method by which to add asphalt binder to the road base during the
reclamation process. The resulting product, emulsion-treated base
(ETB), may then be surfaced with a wearing course for trafficking.
Several studies have been performed to assess the long-term strength
of ETB in the field. In pavement tests performed 6 months or more
after pavement reconstruction, Illinois Department of Transportation
personnel measured resilient modulus values as high as 200 ksi, and
researchers at the Texas Transportation Institute measured resilient
modulus values as high as 275 ksi (1, 2). Although these studies deter-
mined that long-term ETB strengths are sufficient to support even
heavy traffic loads, little research has been performed to determine
the rate at which ETB develops strength in the period between
construction and the time that ETB design properties are achieved.
The ability of a pavement to withstand early trafficking depends on
the strength developed in the pavement system immediately after
construction. If the strength of the pavement system is not sufficiently
high before traffic is introduced, early trafficking will cause permanent
deformation in the treated layer, compromising long-term pavement
performance. Furthermore, if the ETB does not reach design strengths
for an extended period of time, continued trafficking may cause
premature failure of the pavement system. For these reasons, an
understanding of early strength development in ETB is vital for
pavement engineers interested in utilizing FDR in conjunction with
asphalt emulsion treatment.
The objectives of this research were, therefore, to investigate the
early-age structural properties of base materials stabilized with asphalt
emulsion and to assess the rate at which ETB design properties are
achieved. The following sections provide background information,
describe the procedures, present the results, and offer conclusions
and recommendations pertaining to this study.
BACKGROUND
The following sections provide background information on ETB
construction processes, ETB strength gain, construction and environ-
mental factors affecting ETB performance, and the effects of early
trafficking.
ETB Construction Processes
The construction process for FDR with emulsion stabilization begins
with the pulverization and blending of the existing asphalt layer and
a specified thickness of the underlying base material. Partial milling
of the existing asphalt layer may be needed before reclamation to
ensure uniform RAP content within the reclaimed base layer (7 ).
The reclaimed material is then graded and compacted to approximate
Quick and Guthrie
final elevations before the emulsion treatment is applied (8). Water
can be added during the initial reclamation process to facilitate a
more uniform distribution of the emulsion (9, 10). Emulsion is then
injected and mixed into the base material with a reclaimer. Additional
water can be added during the injection to attain the optimum mois-
ture content of the ETB material. The treated base is then com-
pacted with sheepsfoot rollers or vibratory breakdown rollers. After
compaction, the ETB is graded and finish rolled. Paving of the
ETB is often delayed by as much as 2 weeks after construction to
allow moisture to escape from the ETB during the early curing
process (9–11); the maximum allowable ETB moisture contents
before paving are typically 2% to 3% (10). On some projects, traffic
is reintroduced immediately after the final compaction of the ETB
layer, even before paving. If no visually apparent deflections are
observed under a heavy truck, the ETB layer is usually judged to be
ready for traffic (9, 10).
ETB Strength Gain
Asphalt emulsion is typically considered to be an oil-in-water
emulsion, meaning that it consists of asphalt particles that are sus-
pended in water through the use of an emulsifier (6, 12). Emulsifiers
create charges on the surfaces of the asphalt particles that cause
them to repel one another, thereby stabilizing the particles within the
emulsion. Asphalt emulsions typically contain 25% to 60% water,
40% to 75% bitumen, and 0.1% to 2.5% emulsifier (12). Solvents
are sometimes added to modify emulsion properties and behavior. The
specific composition of an asphalt emulsion determines the emulsion
characteristics, such as reactivity, viscosity, and stability.
The process of curing involves the gradual evaporation and
expulsion of water from the emulsion. The curing of ETB begins
when the emulsion starts to destabilize due to compaction and water
evaporation (11). During compaction, the asphalt particles are
forced together, causing them to overcome static repulsion and
begin to coalesce into larger asphalt droplets. The asphalt droplets
eventually become large enough to bind the aggregate particles
together. The rate of curing depends on several factors, including
the reactivity of both the emulsion and the aggregate, the emulsion
chemistry, and environmental factors, such as wind speed, humidity,
and temperature (11, 13). Compaction or trafficking of the ETB
can increase the rate of curing by forcing the asphalt particles closer
together (12).
Curing to the design strength may require a few weeks to a couple
of years, depending on the properties of the emulsion used (11–13).
ETB exhibits low strengths immediately after construction due to
the nature of uncured emulsion. One study found that the stiffness
of ETB after compaction was actually lower than in the reclaimed
material before the emulsion treatment (14); however, ETB layers
have been found to exhibit large increases in resilient moduli during
the first 28 days of curing (15, 16). Others have measured a 300%
increase in resilient moduli during the first 10 months (17 ). The
Asphalt Institute suggests that ETB remains relatively weak dur-
ing the first month after construction, stiffens dramatically during
the next few months, and then levels out after approximately 6 months.
The Asphalt Institute has also found that curing times longer than
6 months do not significantly increase ETB strength (18). Other
studies have found that ETB can take as long as 2 years to fully cure
(13). These results show that, although the final strength of ETB
can be very high, the ETB layer remains fairly weak immediately
after construction, while the emulsion is curing.
41
Construction Factors Affecting ETB Performance
Several construction factors can affect the performance of ETB
materials, including the gradation of the reclaimed base material, the
strength of the subgrade, the degree of ETB compaction, and the
total moisture content (TMC) of the ETB. Where gradation is
concerned, the fraction of material passing the No. 200 sieve should
be less than 25% to avoid weakening the ETB (6, 7). Therefore, the
inclusion of excessive portions of fine-grained subgrade soils into the
base material during the reclamation process should be avoided (1).
The subgrade strength is especially important immediately after
construction (2). The stabilized base layer is fairly weak after pul-
verization and reclamation, so the ability of the subgrade to withstand
construction and early traffic loads greatly affects the support offered
to the ETB. The strength of the subgrade also affects the degree of
compaction possible in the ETB (7 ).
The degree of ETB compaction affects strength development in
the ETB. Compaction can contribute to the initial destabilization
of the asphalt emulsion but also affects the rate of curing within the
ETB. The percentage of voids remaining after compaction should
be low enough to prevent water ingress but high enough to allow
water to evaporate from the emulsion during the curing process.
The TMC includes the in situ moisture that existed before the
emulsion injection and the water added during the injection process,
including the water present in the emulsion. If the TMC is not within
an acceptable percentage of the optimum moisture content for the
ETB material, compaction of the ETB layer to the specified density
may not be possible. If the TMC approaches saturation, compaction
of the ETB can be extremely difficult, if not impossible. If in situ
moisture contents are such that the addition of emulsion will increase
the TMC to unacceptable values, the reclaimed material must be
allowed to dry before the emulsion can be added (11).
Environmental Factors Affecting ETB Performance
Environmental factors such as temperature and moisture can also
affect the performance of ETB materials. Because of the viscoelastic
nature of asphalt, the structural properties of ETB are affected by
ETB temperature (6, 13, 19). As pavement temperature increases,
the strength of the ETB layer decreases due to the softening of the
asphalt binder (2, 20).
Several studies have found that, during the early stages of curing,
excessive moisture contents from rain or other water sources can
cause pavement weakness and even failure (2, 21). After the emulsion
has fully cured, the ability of moisture to affect the pavement system
is limited, but high moisture contents in the period immediately after
construction can slow the curing process and lower the early strength
of the ETB (11).
Ambient air temperature and relative humidity also affect the rate
at which ETB will cure because they affect water evaporation rates.
Low temperatures and high humidity reduce evaporation rates, thus
preventing water from being removed from the ETB and slowing the
curing rates (11).
Effect of Early Trafficking on ETB Performance
The degree to which the emulsion has cured has a large effect on
the stiffness and strength of the ETB in the first 2 years after con-
struction (13). During this time, permanent deformation is usually
42
the primary failure mechanism of ETB layers because the curing
process has not yet fully been completed (12). Because of the nature
of this failure mechanism, the stiffness of the ETB will, in large part,
govern its ability to support traffic loads (19). In current pavement
design methods, the stiffness of a pavement layer is directly tied to
pavement structural capacity. If the layer modulus is low, pavement
structural capacity is significantly reduced.
One of the stated benefits of ETB is that the pavement can be
opened to traffic within hours of construction (1, 2, 9–11, 22);
however, some ETB projects have experienced severe rutting prob-
lems during early pavement life due to the adverse effects of traffic
on the weak pavement (2, 22). One study found that the life of a
pavement comprised of an ETB layer increases by more than 400%
when traffic is withheld from the pavement for 48 hours after con-
struction, compared with when traffic is reintroduced within 2 hours
of construction (22).
PROCEDURES
The test site chosen for this research was a section of Redwood Road
(SR-68) located just north of Saratoga Springs, Utah. This section
of Redwood Road was part of a multiphase reconstruction project.
The pavement design applied to the test section included the use of
1 in. of open-graded surface course on 5 in. of hot-mix asphalt (HMA)
on 8 in. of ETB. Construction of the Redwood Road test area occurred
in June 2009. The test section was neither paved nor opened to
regular traffic for 2 weeks after construction so that it could dry after
a couple of rain storms.
The experimental area was divided into three 800- × 24-ft test
sections, labeled as Sections A, B, and C. Ten individual test stations
were established in each of the three sections. These stations were
randomly located across each section. The following sections of the
paper detail the field procedures, laboratory procedures, and data
analyses performed during the investigation.
Field Procedures
Field testing was conducted over the course of the project to char-
acterize the in situ structural properties of the ETB layer. The field
instruments utilized in this research included the portable falling
weight deflectometer (PFWD) and the dynamic cone penetrometer
(DCP). Testing with these instruments was performed several times
after the construction of the pavement sections. The first series of
field tests began within 30 min of the final compaction of the ETB
layer. Additional tests were performed at 2, 3, 7, and 14 days, 4 months,
and 1 year. The stations tested at 4 months and 1 year were limited
to those in the right lane due to constraints associated with traffic
control.
The PFWD test was performed in general accordance with
ASTM E2583 (Standard Test Method for Measuring Deflections
with a Light Weight Deflectometer). The PFWD system utilized a
44.1-lb weight dropped 30 in. onto a 7.87-in.-diameter load plate.
Three sensors were used to measure pavement deflection at radial
distances of 0, 12, and 24 in. from the point of impact. A seating load
was applied before the actual measurements were taken to ensure
that the load plate was properly situated on the pavement surface.
Three PFWD tests were performed at each station during each series
of testing, and, in each case, the average deflections were used to
backcalculate the modulus of each layer of the pavement system.
Transportation Research Record 2253
The DCP testing was performed in general accordance with
ASTM D6951 (Standard Test Method for Use of the Dynamic Cone
Penetrometer in Shallow Pavement Applications). In this research,
the DCP tests were performed to a depth of 31.5 in., which allowed
average penetration rates for both the ETB and the subgrade layers
to be determined. One test was performed at each station during
each series of testing.
On the day of construction, samples of the reclaimed base material
were removed from each test station before and after the emulsion
treatment. The samples of untreated material were bagged and
transported to the Brigham Young University Highway Materials
Laboratory for sieve, moisture content, and burn-off analyses. The
samples of the treated material were compacted on site with the
modified Proctor compaction protocol in general accordance with
ASTM D1557 (Standard Test Methods for Laboratory Compaction
Characteristics of Soil Using Modified Effort [56,000 ft-lbf/ft3
(2,700 kN-m/m3)]) Method B. This method involved compaction of
the ETB material in 4-in.-diameter molds in five lifts of 25 blows each.
Four or five samples were created for each of the test stations. At the
laboratory, these samples were subjected to unconfined compressive
strength (UCS), moisture content, and burn-off analyses.
Laboratory Procedures
Material characterization testing was conducted on the field-
sampled base material from each test station to determine the average
properties of each of the three test sections. A sieve analysis was
performed on a representative sample of the untreated material in
general accordance with ASTM D422 (Standard Test Method for
Particle-Size Analysis of Soils) to determine the soil classification
of the untreated material sampled from each test section. The material
was determined to be nonplastic, so Atterberg limits could not be
determined. A burn-off test was performed on the untreated and treated
materials in general accordance with ASTM D6307 (Standard Test
Method for Asphalt Content of Hot-Mix Asphalt by Ignition Method).
The results of these tests were used to determine the asphalt content
of the reclaimed base material at each test station before and after
the emulsion injection. The amount of emulsion injected at each sta-
tion was then calculated as the measured difference in the asphalt
content of the base material before and after the emulsion treatment,
divided by the design asphalt content of 64% by weight of the
emulsion.
UCS tests were performed in general accordance with ASTM
D1633 (Standard Test Methods for Compressive Strength of Molded
Soil–Cement Cylinders) on one emulsion-treated sample from each
test station at 7 days, 28 days, 3 months, and 1 year after construction.
Samples were allowed to cure at room temperature in an open-air
condition before testing; they were then capped with gypsum and
subjected to UCS testing at a strain rate of 0.05 in./min. UCS values
were plotted to develop strength-gain curves for the ETB under
laboratory curing conditions.
Data Analysis
The soil properties measured in the field and during laboratory testing
were analyzed to determine equivalent modulus values for the ETB
layer. Moduli were determined from the PFWD test with BAKFAA
backcalculation software (23). The original pavement design layer
thicknesses were used during backcalculation; these thicknesses
Quick and Guthrie 43

were confirmed through DCP testing. A Poisson’s ratio of 0.35 and remained unpaved for 2 weeks after construction, the analysis of the
full interface bonding were assumed for all backcalculations. 2-week capacity was performed for two scenarios—one in which
DCP penetration rates were used with Equations 1 and 2 to the ETB was paved and one in which it was unpaved—to provide
determine ETB modulus values (24): an understanding of pavement strengths under both conditions. All
analyses of pavement capacity after 2 weeks were performed with
292 the assumption that the ETB was paved.
CBR = (1)
PR1.12

where CBR is the California bearing ratio (%) and PR is the RESULTS
penetration rate (mm/blow).
The following sections describe the results of the laboratory and
MR = 2 ,550 i CBR 0.64 (2) field testing, as well as the results of the data analyses.

where MR is the resilient modulus (psi).

Laboratory Results
AASHTO correlation charts were used to determine an equivalent
structural coefficient (a2) for the ETB modulus values determined The results of the sieve analysis indicated that all three sections could
during testing (18). The structural coefficients were plotted against be classified as A-1-a and GW, under the AASHTO and Unified Soil
time to determine trends in ETB strength development and to deter- Classification System methods, respectively. A minimal variation in
mine the time at which the ETB had reached the design structural aggregate gradation existed between the three test sections.
coefficient value of 0.25. The results of the burn-off tests are shown in Table 1. The asphalt
The structural capacity of the pavement in each testing period was content before treatment was fairly consistent across the three sections,
calculated with the AASHTO flexible pavement design method. The implying that they had comparable RAP contents. However, the
structural coefficient of the ETB layer measured in each time period variability in the emulsion content added to each section during
was used to calculate the total structural number of the pavement treatment was higher. The design emulsion content, specified as a
system, as shown in Equation 3 (18): percentage of asphalt, water, and emulsifier by dry weight of reclaimed
base material, was 4.0% for the ETB layer. Sections A and B both had
SN = a1 i D1 + a2 i D2 i m2 (3) emulsion contents higher than the design specification, and Section C
had an emulsion content lower than the design specification.
where Results of the UCS testing are shown in Figure 1. Sections B and C
had nearly identical values; Section A had consistently lower values.
SN = structural number, The reduced compressive strength of the samples from Section A
a1 = structural coefficient of the HMA layer, may have been due to the higher amount of emulsion, which could
D1 = thickness of the HMA layer, act as a lubricant. The average UCS values for all three sections were
a2 = structural coefficient of the ETB layer, 30, 96, 145, and 179 psi at 7 days, 28 days, 3 months, and 1 year,
D2 = thickness of the ETB layer, and respectively. Although no direct correlation between UCS and
m2 = drainage coefficient of the ETB layer. modulus values for ETB materials was identified in the literature
The allowable number of passes of equivalent single-axle loads review performed in this research, the resilient modulus determined
(ESALs) before failure was then computed with Equation 4 (18): under laboratory conditions would be expected to exhibit a similar
pattern of strength development. The percentage of 1-year ETB
ΔPSI strength developed after 7 days, 28 days, and 3 months was 17%,
log 54%, and 80%, respectively. The ETB compressive strength was
log Wt18 = Z R S0 + 9.36 log ( SN + 1) − 0.20 + 2.7
very low immediately after compaction and remained below 50% of
1, 094
0..4 + the 1-year strength during at least the first 2 weeks.
(SN + 1) 5.19

+ 2.32 log MR − 8.07 (4) Field Results

where Results from the PFWD and DCP tests are shown in Table 2. PFWD
Wt18 = allowable ESALs before pavement failure, results are given as modulus values backcalculated from measured
ZR = standard normal deviate corresponding to a specified
level of reliability,
S0 = standard deviation, TABLE 1 Burn-Off Test Data
ΔPSI = allowable change in present serviceability index over
Asphalt Content Emulsion Added
pavement life, and Before Treatment (%) During Treatment (%)
MR = effective roadbed soil resilient modulus (psi).
Section Average SD Average SD
To calculate the structural number, a structural coefficient of
0.40 was used for the HMA layer, and a drainage coefficient of 1.0 A 5.0 0.6 4.8 2.0
was used for the ETB layer. To calculate the allowable number of B 4.5 0.5 4.1 1.1
ESALs, a reliability level of 95%, a standard deviation of 0.45, and C 4.8 0.4 3.3 1.2
a change in present serviceability index of 2.0 were used, with a
design subgrade modulus of 11.4 ksi. Although all three test sections NOTE: SD = standard deviation.
 200
180

Compressive Strength (psi)

160
140 Section A
120
100 Section B
80
Section C
60
40 Average
20
0
0 100 200 300 400
Time (days)

FIGURE 1 Average unconfined compressive strength results for each section.

TABLE 2 PFWD and DCP Test Data

DCP

Penetration Rate
PFWD Modulus (ksi) (mm/blow) Modulus (ksi)
Curing Time
(days) Average SD Average SD Average SD

Section A
1 21.7 4.4 6.4 1.2 26.1 3.7
2 22.0 3.6 6.7 1.3 25.2 3.8
3 31.2 5.9 4.3 0.8 34.7 4.8
7 39.5 8.9 3.2 0.9 43.2 7.1
14 36.6 11.0 3.4 0.5 40.5 4.0
113 342.4 90.1 1.1 0.6 — —
365 85.4 11.9 1.4 0.1 — —
Section B
1 13.8 4.1 9.5 2.0 19.6 3.0
2 18.9 3.3 8.0 1.1 21.9 2.1
3 22.9 4.5 6.7 0.6 24.8 1.7
7 23.4 4.5 5.9 1.1 27.5 3.3
14 27.2 4.0 5.4 0.8 29.1 3.0
113 339.9 87.4 1.0 0.3 — —
365 94.3 11.1 1.8 0.3 — —
Section C
1 17.2 3.1 9.2 1.1 19.9 1.8
2 22.0 4.7 8.4 1.2 21.2 2.2
3 28.7 5.2 6.3 0.7 25.9 2.3
7 30.4 6.1 4.6 0.5 32.7 2.9
14 35.5 6.3 4.1 0.6 35.6 4.0
113 176.9 29.3 1.2 0.3 — —
365 89.2 8.2 2.0 0.1 — —
Average
1 17.6 3.9 8.4 1.4 21.9 2.8
2 21.0 3.9 7.7 1.2 22.8 2.7
3 27.6 5.2 5.8 0.7 28.5 2.9
7 31.1 6.5 4.6 0.8 34.5 4.4
14 33.1 7.1 4.3 0.6 35.1 3.7
113 286.4 68.9 1.1 0.4 — —
365 89.6 10.4 1.7 0.2 — —

NOTE: — = penetration rates below recommended range for the equations used.
Quick and Guthrie 45

50
45

ETB Resilient Modulus (ksi)

40
Section A
35
30 Section B
25
Section C
20
15 Average
10
5
0
0 5 10 15
Time (days)
(a)

50
45
ETB Resilient Modulus (ksi)

40
Section A
35
30 Section B
25
Section C
20
15 Average
10
5
0
0 5 10 15
Time (days)
(b)

FIGURE 2 ETB resilient modulus values during first 2 weeks: (a) PFWD test results and
(b) DCP test results.

deflections, and DCP results are given as penetration rates and cor-
responding modulus values. Modulus values were not determined
from the DCP tests at 4 months and 1 year because the penetration
rates measured in those tests were below the recommended range
for the given equations (25). The average modulus values obtained
for each section from PFWD and DCP testing in the first 2 weeks of
testing are shown in Figure 2. With a few exceptions, the PFWD
and DCP tests produced similar curves within each section, although
the DCP modulus values were, in general, slightly higher than the
PFWD modulus values. Overall average modulus values for this
2-week period were 26.0 and 28.6 ksi, determined from the PFWD
and DCP data, respectively.
The average ETB modulus values obtained for each section from
PFWD testing during the entire testing period are shown in Figure 3.
The PFWD modulus values were consistently low in all three sections
during the first 2 weeks, increased dramatically by 4 months, and
then decreased considerably by 1 year. The average ETB moduli were
286.4 and 89.6 ksi at 4 months and 1 year, respectively. Therefore, the
average ETB moduli during the first 2 weeks and at 1 year were only
9% and 31%, respectively, of the average ETB modulus at 4 months.
These trends were confirmed by the DCP penetration rates: the av-
erage penetration rates were 4.3, 1.1, and 1.7 mm/blow at 2 weeks,
4 months, and 1 year, respectively.
Much of the increase in modulus values observed at 4 months
was probably due to the curing of the emulsion in the base, but the
lower temperatures experienced during late fall may also have
caused an apparent increase in stiffness. Although the air temper-
ature of 57°F at 4 months was only slightly different than the air
temperature of 64°F at 2 weeks, the difference in subsurface tem-
peratures was most likely greater due to the onset of colder nights
by 4 months. Section C had a lower emulsion content, so it might
not have been as affected by colder temperatures, possibly causing
it to exhibit lower strengths at 4 months; however, at 1 year, all three
sections had approximately the same modulus value. The modulus
decreased significantly between 4 months and 1 year despite the
additional curing of the emulsion that may have occurred during
this time. Although a comparatively high air temperature of 80°F
may have contributed to the apparent reduction in ETB strength at
46 Transportation Research Record 2253

400
October 2009
350

300

ETB Modulus (ksi)

Section A
250
Section B
200
Section C
150 June 2010
Average
100

50
June 2009
0
0 100 200 300 400
Time (days)
(a)

0.35
October 2009
ETB Structural Coefficient (a2)

0.30

0.25 Section A

0.20 Section B

June 2010
0.15 Section C

0.10 Average

0.05
June 2009
0.00
0 100 200 300 400
Time (days)
(b)

FIGURE 3 ETB properties during first year: (a) resilient modulus values and
(b) layer coefficients.

1 year, freeze–thaw damage may also have occurred in the layer
during the winter.
Although the trend shown in Figure 3 may not account for the
fluctuations in ETB strength that occurred between 2 weeks and
4 months or between 4 months and 1 year, the results still show that
the sections exhibited strengths at least as high as those measured at
4 months and that those strengths were significantly reduced by 1 year.
The laboratory UCS data shown in Figure 1 depict the trend that would
have been expected in the field without environmental effects.
Data Analysis
The equivalent AASHTO structural coefficients of the ETB are shown
in Figure 3 for each section. Because modulus values were not deter-
mined from the DCP penetration rates measured at 4 months and
1 year, structural coefficients were determined only by the PFWD
modulus values. The design structural coefficient for the ETB layer
was 0.25, as indicated by the bold line. In the first 2 weeks after
construction, the average ETB structural coefficients for Sections A,
B, and C were 0.04, 0.03, and 0.04, respectively. Sections A and B,
which behaved similarly, reached a structural coefficient of 0.25
after approximately 3 months and actually exceeded design values
by about 0.05 after 4 months; however, after 1 year, Sections A and B
had much lower structural coefficients: 0.11 and 0.12, respectively.
Section C did not ever reach a structural coefficient of 0.25. At
4 months, the ETB structural coefficient in Section C was only 0.21,
and after 1 year it had decreased to 0.12. For all three sections, the
average structural coefficient measured after 1 year of curing was
only 47% of the design value. These results show that ETB materials
similar to those investigated in this research should not be expected
to reach design values in the first few months after construction; this
conclusion is supported by the laboratory UCS results. Furthermore,
ETB materials may exhibit substantial decreases in stiffness after
the first winter in regions characterized by freeze–thaw cycling.
The design ESALs for this pavement were 10.9 million. However,
the number of allowable ESALs during the first 2 weeks for the
scenario in which the ETB was unpaved was calculated to be less than
50,000, which is 0.5% of the design requirement. For the scenario
in which the ETB was paved immediately after construction, the
allowable ESALs were calculated to be 0.4 million, 14.5 million,
and 1.7 million at 2 weeks, 4 months, and 1 year, respectively. These
Quick and Guthrie
values were equivalent to 4%, 133%, and 16% of the design capacity.
Although pavement capacity was sufficient at 4 months, it was severely
reduced during the first 2 weeks and at 1 year. Trafficking under
these reduced capacities would be expected to greatly compromise
long-term performance and might cause premature failure of the
pavement system.
CONCLUSION
The objectives of this research were to investigate the early-age
structural properties of base materials stabilized with asphalt emulsion
and to assess the rate at which ETB design properties are achieved.
Three experimental sections were established along a pavement
reconstruction project near Saratoga Springs. Field tests were per-
formed to assess the structural properties of the ETB immediately
after construction and then at 2, 3, 7, and 14 days, 4 months, and
1 year. Measured values were plotted against time to determine
trends in ETB stiffness development.
The average UCS measured in the laboratory for all three sections
was 30, 96, 145, and 179 psi at 7 days, 28 days, 3 months, and 1 year,
respectively. The percentage of 1-year ETB strength developed after
7 days, 28 days, and 3 months was 17%, 54%, and 80%, respectively.
Field-measured modulus values were consistently low in all three
sections in the first 2 weeks after construction, increased dramatically
by 4 months, and then decreased considerably by 1 year. The average
ETB moduli measured with the PFWD were 26.0, 286.4, and 89.6 ksi
during the first 2 weeks, at 4 months, and at 1 year, respectively, and
the DCP penetration rates followed the same trend; the corresponding
average penetration rates were 4.3, 1.1, and 1.7 mm/blow. Much of
the increase in modulus values observed at 4 months was probably due
to the curing of the emulsion in the base, but the lower temperatures
experienced during late fall may also have caused an apparent increase
in stiffness. The modulus decreased significantly between 4 months
and 1 year despite the additional curing of the emulsion that may
have occurred during this time. Although a comparatively high air
temperature of 80°F may have contributed to the apparent reduction
in ETB strength at 1 year, freeze–thaw damage may also have
occurred in the layer during the winter.
In the first 2 weeks after construction, the average ETB structural
coefficients for Sections A, B, and C were 0.04, 0.03, and 0.04,
respectively. Sections A and B reached the design structural coeffi-
cient of 0.25 after approximately 3 months; however, after 1 year,
Sections A and B had structural coefficients of 0.11 and 0.12,
respectively. Section C did not ever reach a structural coefficient
of 0.25. At 4 months, the ETB structural coefficient in Section C
was only 0.21, and after 1 year it had decreased to 0.12. For all three
sections, the average structural coefficient measured after 1 year of
curing was only 47% of the design strength.
The design pavement capacity was 10.9 million ESALs. The
pavement capacity during the first 2 weeks for the scenario in which
the ETB was unpaved was calculated to be less than 50,000 ESALs,
which is 0.5% of the design capacity. For the scenario in which
the ETB was paved immediately after construction, the pavement
capacity, in terms of millions of ESALs, was 0.4, 14.5, and 1.7 at
2 weeks, 4 months, and 1 year, respectively. These values are
equivalent to 4%, 133%, and 16% of the design capacity.
The results of this research show that, although pavement capacity
is sufficient at 4 months, it is severely reduced during the first 2 weeks
and at 1 year. Trafficking under these reduced capacities would be
expected to greatly compromise long-term performance and may
cause premature failure of the pavement system. For this reason,
47
trafficking of materials similar to those investigated in this research
is not recommended during at least the first 2 weeks after construction.
Furthermore, these research results suggest that ETB stiffness can be
affected by temperature and moisture changes and that ETB materials
may sustain freeze–thaw damage during the winter in cold regions.
Further research on these topics is needed.
ACKNOWLEDGMENT
The authors acknowledge the Utah Department of Transportation for
funding this research.
REFERENCES
1. Hilbrich, S. L., and T. Scullion. Evaluation of Laboratory Mix Design
and Field Performance of Asphalt Emulsion and Cement Stabilized Full-
Depth Reclamation Project in Texas. Presented at 87th Annual Meeting
of the Transportation Research Board, Washington, D.C., 2008.
2. Thompson, M. R., L. Garcia, and S. H. Carpenter. Cold In-Place Recycling
and Full-Depth Recycling with Asphalt Products. Illinois Center for
Transportation, University of Illinois at Urbana–Champaign, March 2009.
3. MacGregor, J. A. C., W. H. Highter, and D. J. DeGroot. Structural
Numbers for Reclaimed Asphalt Pavement Base and Subbase Course
Mixes. In Transportation Research Record: Journal of the Transportation
Research Board, No. 1687, TRB, National Research Council, Washington,
D.C., 1999, pp. 22–28.
4. Guthrie, W. S., A. V. Brown, and D. L. Eggett. Cement Stabilization of
Aggregate Base Material Blended with Reclaimed Asphalt Pavement.
In Transportation Research Record: Journal of the Transportation
Research Board, No. 2026, Transportation Research Board of the National
Academies, Washington, D.C., 2007, pp. 47–53.
5. Guthrie, W. S., D. A. Cooley, and D. L. Eggett. Effects of Reclaimed
Asphalt Pavement on Mechanical Properties of Base Materials. In Trans-
portation Research Record: Journal of the Transportation Research
Board, No. 2005, Transportation Research Board of the National Acad-
emies, Washington, D.C., 2007, pp. 44–52.
6. Terrel, R. L., J. A. Epps, E. J. Barenberg, J. K. Mitchell, and M. R.
Thompson. Soil Stabilization in Pavement Structures—A User’s Manual,
Volume 2. Terrel, Epps, and Associates, Seattle, Wash., 1979.
7. Collings, D., and H. Thompson. A Critical Appraisal of the Performance
of Foamed Bitumen and Bitumen Emulsion Treated Materials. Proc.,
9th Conference on Asphalt Pavements for Southern Africa, Gaborone,
Botswana, Sept. 2007, pp. 233–255.
8. Guthrie, W. S., and M. A. Rogers. Variability in Construction of
Cement-Treated Base Layers: Material Properties and Contractor
Performance. In Transportation Research Record: Journal of the Trans-
portation Research Board, No. 2186, Transportation Research Board of
the National Academies, Washington, D.C., 2010, pp. 78–89.
9. Finberg, C., D. Quire, and T. W. Thomas. Granular Base Stabilization
with Emulsion in Las Vegas, Nevada. Presented at 87th Annual Meet-
ing of the Transportation Research Board, Washington, D.C., 2008.
10. Kroge, M., K. McGlumphy, and T. Besseche. Full Depth Reclamation
with Engineered Emulsion in Fairburn, Georgia. In Transportation
Research Record: Journal of the Transportation Research Board, No. 2095,
Transportation Research Board of the National Academies, Washington,
D.C., 2009, pp. 136–143.
11. Mulusa, W. K. Development of a Simple Triaxial Test for Characterising
Bitumen Stabilised Materials. MS thesis. Department of Civil Engineer-
ing, University of Stellenbosch, Matieland, South Africa, Jan. 2009.
12. James, A. Overview of Asphalt Emulsions. In Transportation Research
Circular E-C102: Asphalt Emulsion Technology. Transportation Research
Board of the National Academies, Washington, D.C., Aug. 2006, pp. 1–15.
13. GEMS: The Design and Use of Granular Emulsion Mixes, Manual 14.
Southern African Bitumen Association, Cape Town, South Africa, 1993.
14. Budge, A. S., and W. J. Wilde. Monitoring Curing of Emulsion-Stabilized
Roadways Using the Dynamic Cone Penetrometer. In Soil and Material
Inputs for Mechanistic–Empirical Pavement Design (GSP 169). Proc.,
Geo-Denver 2007: New Peaks in Geotechnics, Denver, Colo. ASCE,
Reston, Va., 2007, pp. 1–8.
48 Transportation Research Record 2253

15. Finn, F. N., R. G. Hicks, W. J. Kari, and L. D. Coyne. Design of Emul-

sified Asphalt Treated Bases. In Highway Research Record 239, HRB,
National Research Council, Washington, D.C., 1968, pp. 54–75.
16. Bondietti, M., D. Murphy, K. Jenkins, and R. Burger. Research on the
Stabilisation of Two Different Materials Using Bitumen Emulsion and
Cement. Proc., 8th Conference on Asphalt Pavements for Southern
Africa, Sun City, South Africa, Sept. 2004. http://www.capsa11.co.za/
capsa04/Documents/026.pdf. Accessed June 22, 2010.
17. DeBeer, M., and J. E. Grobler. Towards Improved Structural Design
Criteria for Granular Emulsion Mixes. Proc., 6th Conference on Asphalt
Pavements for Southern Africa, Cape Town, South Africa, 1994,
pp. III44–III68.
18. Huang, Y. H. Pavement Analysis and Design, 2nd ed. Prentice Hall,
Upper Saddle River, N.J., 2004.
19. Liebenberg, J. J. E., and A. T. Visser. Stabilization and Structural Design
of Marginal Materials for Use in Low-Volume Roads. In Transporta-
tion Research Record: Journal of the Transportation Research Board,
No. 1819, Vol. 2, Transportation Research Board of the National Acad-
emies, Washington, D.C., 2003, pp. 166–172.
20. Budge, A. S., and W. J. Wilde. A Modified Method for MR Testing to
Evaluate Temperature Effects in Emulsion-Stabilized Gravel. Proc.,
GeoCongress 2008: Characterization, Monitoring, and Modeling of FIGURE 4 Redwood Road typical pavement condition, April 2011.
GeoSystems (GSP 179), New Orleans, La., ASCE, Reston, Va., 2008,
pp. 12–19.
21. O’Flaherty, C. A. Highways: The Location, Design, Construction &
Maintenance of Pavements, 4th ed. Butterworth-Heinemann, Woburn, road was a fairly thick structure (6 in. of HMA, 8 in. of ETB, and
Mass., 2002. 4 to 5 in. of remaining aggregate base).
22. Botha, P. B., C. J. Semmelink, J. Raubenheimer, B. Perry, and The current pavement performance is good. Although the authors
A. Hodgkinson. Investigation into the Early Trafficking of Emulsion discuss pavement performance in terms of the ETB modulus, visual
Treated (ETB), Foamed Bitumen (FB) Bases Treated in Combination assessments or similar pavement rating parameters are not discussed.
with Cement and Cement (OPC) Only. In TREMTI 2005: Treatment and
Recycling of Materials for Transport Infrastructure, 2nd International
Figure 4 gives an indication of the performance of Redwood Road
Symposium, Paris, Oct. 2005, pp. 1–10. in April 2011; it was constructed in June 2009. This is one of the
23. Airport Technology, Research and Development Branch, FAA. Documents light FWD test sections and is representative of the overall road.
and Downloads Page. http://www.airporttech.tc.faa.gov/naptf/download/ The right wheelpath is still wet from rain from 2 days before. No
index1.asp. Accessed July 21, 2009. visible rutting or cracking has been observed; rutting and cracking
24. Dai, S., and C. K. Kremer. Improvement and Validation of Mn/DOT
DCP Specifications for Aggregate Base Materials and Select Granular
might be expected if the ETB modulus was significantly lower than
Test. Report MN/RC-2005-32. Office of Materials and Road Research, the design modulus.
Minnesota Department of Transportation, Maplewood, Jan. 2006. In conclusion, the authors are asked to report the data for all layers
25. Heavy-Duty Dual-Mass Dynamic Cone Penetrometer (H-4219) Instruction from the backcalculation analysis and answer the other questions
Manual. Humboldt Manufacturing Corporation, Norridge, Ill., 2003. presented in this discussion. It is also recommended that deflections be
measured with full-scale FWD apparatus.

DISCUSSION
Todd W. Thomas
Road Science, LLC, 6502 South Yale Avenue, Tulsa, OK 74136,
tthomas@roadsciencellc.com.
This paper considers the rate at which asphalt ETB material design
properties are reached. Modulus values peaked at 4 months but
dropped off significantly at 1 year, as calculated from backcalculation
analysis from PFWD data. Specifically, the paper concludes that the
modulus values obtained at 1 year were below design values and that
this may be due to moisture changes and freeze–thaw damage.
An understanding of the accuracy of the light falling weight
deflectometer (FWD) data is needed. ASTM D5858 provides a method
for evaluating the match between the actual and measured FWD
deflection basins through an equation for root mean square error.
The authors do not report the match of the deflection basins. It would
be useful to have several other questions answered. How many iter-
ations were performed in the backcalculation process? Was a hard
layer assumed, and what was its depth? What were the modulus
values of the other layers, and were they assumed? Was calibration
of the equipment up-to-date? What were the testing locations, as
referenced from the pavement centerline? This information is
important to the reliability of the data, given that the typical loads of
a PFWD are light, only three deflection sensors were used, and the
AUTHORS’ CLOSURE
The discussant requests clarification of the methods used during PFWD
testing and backcalculation analysis with respect to ASTM D5858
(Standard Guide for Calculating In Situ Equivalent Elastic Moduli of
Pavement Materials Using Layered Elastic Theory). He also proposes
that the pavement performance as of April 2011 is good, based on a
visual inspection of the road.
In this research, PFWD testing was performed to evaluate the in
situ modulus values at 10 different test stations randomly distributed
across each of three different construction sections, for a total of
30 stations. Figure 5 shows the station layout used for each section,
which generally encompassed the full width of both lanes of north-
bound traffic, bordered with a wide shoulder on one side and a median
on the other. Collected PFWD data were analyzed with BAKFAA,
as explained in the paper; details about this software may be obtained
from the website referenced in the paper (23). In the backcalculation
process, the modulus values of the specified layers were all computed;
none were assumed, and no apparent stiff layer was included. The
DCP data collected in this research did not consistently indicate the
presence of the remaining aggregate base cited by the discussant;
therefore, all materials below the ETB layer were treated as sub-
grade in the PFWD data analyses. PFWD results for all layers and
testing times are displayed in Table 3, in which dashes indicate that
Quick and Guthrie 49

800 ft
X
X X
24 ft

X X
X X X
X X

FIGURE 5 Typical station layout.

TABLE 3 PFWD Test Data

Modulus (ksi)

Asphalt ETB Subgrade

Curing Time
(days) Average SD Average SD Average SD

Section A
1 — — 21.7 4.4 19.7 5.1
2 — — 22.0 3.6 19.7 4.6
3 — — 31.2 5.9 18.7 3.8
7 — — 39.5 8.9 19.3 3.6
14 — — 36.6 11.0 21.1 5.1
113 1,065.1 307.5 342.4 90.1 26.7 3.8
365 288.3 63.9 85.4 11.9 24.6 3.3
699 251.0 49.5 52.2 14.2 25.6 11.7
Section B
1 — — 13.8 4.1 17.6 5.0
2 — — 18.9 3.3 18.6 4.1
3 — — 22.9 4.5 19.1 3.8
7 — — 23.4 4.5 18.1 3.1
14 — — 27.2 4.0 21.9 4.3
113 1,304.0 123.4 339.9 87.4 25.8 2.2
365 299.0 48.0 94.3 11.1 23.6 2.1
699 187.7 16.0 45.7 3.8 29.0 2.7
Section C
1 — — 17.2 3.1 23.0 5.8
2 — — 22.0 4.7 23.7 5.6
3 — — 28.7 5.2 23.1 4.9
7 — — 30.4 6.1 23.1 5.7
14 — — 35.5 6.3 26.4 5.0
113 770.1 122.3 176.9 29.3 28.0 1.7
365 289.9 32.3 89.2 8.2 26.5 3.0
699 227.3 48.5 58.0 29.1 30.1 6.9
Average
1 — — 17.6 3.9 20.1 5.3
2 — — 21.0 3.9 20.7 4.8
3 — — 27.6 5.2 20.3 4.2
7 — — 31.1 6.5 20.1 4.2
14 — — 33.1 7.1 23.1 4.8
113 1,046.4 184.4 286.4 68.9 26.8 2.6
365 292.4 48.0 89.6 10.4 24.9 2.8
699 222.0 38.0 51.9 15.7 28.2 7.1

NOTE: — = preceded placement of the asphalt layer.

50
the curing times preceded the placement of the asphalt layer. The
average root mean square percent errors were 2.4%, 2.0%, and
1.8% for Sections A, B, and C, respectively, for all data presented
in the table.
Concerning pavement performance, the paper focuses on the
early-age structural properties of the ETB layer at the test site. The
PFWD and DCP data both indicate that the ETB modulus values
were consistently low in all three sections during the first 2 weeks
after construction, increased dramatically by 4 months, and then
decreased considerably by 1 year. As stated in the paper, one possible
explanation for these results is the occurrence of freeze–thaw damage
to the ETB layer during the winter. Since the original submission
of the paper, the pavement has experienced a second winter, and
PFWD data collected in May 2011 indicated that the ETB layer had
experienced further reductions in modulus since the 1-year testing
was completed in June 2010. These data are labeled with a curing
time of 699 days in Table 3 and reflect a 42% reduction, on average,
of the ETB modulus in the as-tested condition since June 2010.
(For reference, the average ambient air and asphalt surface temper-
atures for the tests performed at 113, 365, and 699 days were 57°F and
68°F, 80°F and 94°F, and 80°F and 102°F, respectively.) The results
of a distress survey performed in May 2011 indicated that, although
cracking of the asphalt layer had not been observed, rut depths of
2 to 3 mm had been consistently measured in the right wheelpath of
Transportation Research Record 2253
the right lane in all three test sections. The affected wheelpath is the
darker one, pictured on the left-hand side of Figure 5.
Variable early-age stiffness of ETB has also been documented
in research performed by the Virginia Transportation Research
Council (26). In that study, both lanes of the pavement section
constructed with ETB exhibited a significant reduction in effective
structural number between the first winter and spring, a partial
recovery during the summer and fall, and then a second reduction in
stiffness between the second winter and spring. The authors of that
study also attributed this behavior to environmental conditions, the
effects of which warranted further study. To this end, work is under
way at Brigham Young University to relate in situ ETB modulus
measurements to both ETB temperature and moisture content and
also to evaluate the freeze–thaw durability of ETB materials.
REFERENCE
26. Diefenderfer, B. K., and A. K. Apeagyei. Time-Dependent Structural
Response of Full-Depth Reclamation. In Transportation Research
Record: Journal of the Transportation Research Board, No. 2253, Trans-
portation Research Board of the National Academies, Washington, D.C.,
2011, pp. 3–9.
The Chemical and Mechanical Stabilization Committee peer-reviewed this paper.