Effectiveness of Loaded Wheel Tracking Test to Ascertain Moisture Susceptibility of Asphalt Mixtures

Abstract

Moisture damage of asphalt mixtures is a major distress affecting the durability of asphalt pavements. The loaded wheel tracking (LWT) test is gaining popularity in determining moisture damage because of its ability to relate laboratory performance to field performance. However, the accuracy of LWT’s “pass/fail” criteria for screening mixtures is limited. The objective of this study was to evaluate the capability of the LWT test to identify moisture susceptibility of asphalt mixtures with different moisture conditioning protocols. Seven 12.5 mm asphalt mixtures with two asphalt binder types (unmodified PG 67-22 and modified PG 70-22), and three aggregate types (limestone, crushed gravel, and a semi-crushed gravel) were utilized. Asphalt binder and mixture samples were subjected to five conditioning levels, namely, a control; single freeze–thaw-; triple freeze–thaw-; MiST 3500 cycles; and MiST 7000 cycles. Frequency sweep at multiple temperatures and frequencies, and multiple stress creep recovery tests were performed to evaluate asphalt binders. LWT test was used to evaluate the asphalt mixture samples. Freeze–thaw and MiST conditioning resulted in an increase in stiffness in the asphalt binders as compared with the control. Further, freeze–thaw and MiST conditioning resulted in an increase in rut depth compared with the control asphalt mixture. The conditioning protocols evaluated were effective in exposing moisture-sensitive mixtures, which initially showed compliance with Louisiana asphalt mixture design specifications.

Keywords

infrastructure materials asphalt mixture evaluation and performance AKM40 moisture damage testing moisture damage/stripping

Moisture-induced damage of asphalt mixtures is a significant distress that affects the durability and structural integrity of asphalt pavements. Moisture damage in asphalt pavements can be characterized by loss of cohesion in asphalt binder mastic or adhesive failure (stripping) between asphalt binder and aggregates ( 1 ). Moisture-induced cohesive failure in asphalt mixtures results in reduced stiffness and loss of strength, thereby affecting the ability of the pavement to support traffic. Further, adhesive failure results in reduced stiffness and strength, loss of material, and therefore total disintegration of the pavement (1 –3). Moisture damage in asphalt pavement may be exacerbated by freeze–thaw cycles or traffic-induced pore pressure ( 3 ). Factors influencing moisture-induced distresses in pavements include mixture design, mixture production, construction, climatic, and other related factors ( 3 ). Asphalt mixture design factors that affect moisture damage include asphalt binder and aggregate chemistry, aggregate absorption and texture, air void content, addition of mixture additives, aggregate particle distribution, and others ( 3 ).

Since the 1930s, researchers in the asphalt industry have been developing tools for evaluating moisture-induced damage in asphalt pavements ( 4 ). Test methods that have been used over the years for characterization of moisture-induced distress include Immersion Compression Test, Asphalt Film Retention Test, Retained Stability Test, Modified Lottman Test, Loaded Wheel Tracking (LWT) test (LWT) (Hamburg type), and many others (4 –6). Among test methods used for evaluating moisture damage of asphalt mixtures, the modified Lottman test is the most widely used (72% of States in the U.S. use modified Lottman) ( 7 ). The modified Lottman test uses tensile strength ratio (TSR) of moisture-conditioned specimen to dry specimen to evaluate moisture damage. Despite the widespread use of the modified Lottman test, researchers ( 8 , 9 ) have demonstrated that the TSR is not a consistent and reliable indicator of moisture sensitivity of asphalt mixtures. Further, it has been reported that the freeze–thaw conditioning utilized in the modified Lottman test is not practical or capable of simulating moisture damage observed in the field (8 –12).

Background

The Hamburg-type LWT is the second most widely used test method for evaluating moisture sensitivity in asphalt mixtures (16% of States in the U.S. use LWT for moisture sensitivity evaluation) ( 7 ). LWT is a laboratory-controlled rut-depth test that uses loaded wheel(s) to apply a moving load on asphalt mixture specimens to simulate traffic load applied on asphalt pavements ( 5 , 13 ). The LWT device (Hamburg type) was developed by Helmut-Wind Incorporated of Hamburg in the 1970s to evaluate rutting performance of asphalt mixtures by rolling a steel wheel across asphalt mixture specimens submerged in hot water ( 14 ). LWT was introduced into the U.S. in the 1990s, and has gained popularity in recent times because of its ability to correlate laboratory moisture sensitivity results to field moisture damage performance ( 6 , 15 ). Currently states that use the LWT for moisture sensitivity evaluation include Louisiana, Iowa, Maine, Massachusetts, Texas, Utah, Washington, and California. The use of the LWT for moisture sensitivity evaluation is based on pass/fail criteria using different parameters. Figure 1 shows a typical LWT test result. Parameters obtained from LWT results for the evaluation of moisture sensitivity resistance include post-compaction consolidation, creep slope, stripping slope, maximum rut depth (i.e., 12.5 mm), passes at maximum rut depth, and stripping inflection point (SIP) ( 13 , 14 ) (Figure 1). Post-compaction consolidation is the rut depth at 1000 passes. Creep slope characterizes the inverse of deformation rate in the creep phase of rut depth versus number of wheel passes plot (Figure 1). The creep phase starts after the post-compaction consolidation phase and ends before stripping occurs. There is a steady increase in deformation in the creep phase as a result of viscous flow ( 13 ). Stripping slope is the inverse of the deformation rate at points where rut depth increases rapidly as moisture damage occurs. A mixture with a larger stripping slope value is more susceptible to moisture damage. The ratio of the creep slope to the stripping slope has been used to characterize moisture sensitivity of asphalt mixtures in some states ( 13 , 14 ). The SIP is the number of passes at the intersection of creep slope and stripping slope. The SIP is an indication of moisture damage and represents the point where stripping initiates in a mixture ( 14 ). Researchers have established an excellent relationship between laboratory-measured SIP and field moisture damage performance ( 16 ). Table 1 presents a summary of LWT moisture sensitivity specifications ( 7 ).

Table 1.

Moisture Sensitivity Specification for Loaded Wheel Tracking Test

State	Test standard	% AV	Test temperature and condition	Moisture sensitivity criteria
				Max. rut depth (mm)	No. of passes for max. rut depth	Stripping inflection point, min.
Cal.	AASHTO T 324	7±0.5%	PG 58: 45°C, W PG 64: 50°C, W PG 70/>: 55°C, W	12.5	PG 58/<: 10000 PG 64: 15000 PG 70: 20000 PG 76/>: 25000	PG 58/64: 10000 PG 70: 12500 PG 76/>: 15000
Iowa	AASHTO T324	7±1.0%	PG 52/58: 40°C, W PG 64/>: 50°C, W	20.0	20000	TDS: 10000 TDV/H: 14000^c
La.	AASHTO T 324	7±0.5%	50°C, W	TL1: 10 TL2: 6	20000	na
Maine	AASHTO T 324	7±0.5%	45°C, W	12.5	20000	15000
Mass.	AASHTO T 324	7±0.5%	50°C, W	12.5	20000	TL1: 10000 TL2/3: 15000
Texas	Tex-242-F	7±1.0%	50°C, W	12.5	PG 64/<: 10000 PG 70: 15000 PG 76/>: 20000	na
Utah	AASHTO T 324	7±1.0%	PG 58/<: 46°C, W PG 64: 50°C, W PG 76/>: 54°C, W	10.0	75/< Gyr.: 10000 75/> Gyr.: 20000	na
Wash.	AASHTO T 324	7±0.5%	50°C, W	10.0	<0.3 m ESALs: 10000 0.3 to > 3 m ESALs: 12500 >3 m ESALs: 15000	20000

Note: Max. = maximum; Min. = minimum; %AV = percent air void content; Cal. = California; La. = Louisiana; Mass. = Massachusetts, Wash = Washington; W = wet; TL1 = traffic level 1; TL2 = traffic level 2; PG = performance grade; / = or; /< = or lower; /< = or higher; Gyr. = gyrations; m = million; ESALs = equivalent standard axel loads; TDS = traffic designation S; TDV/H = traffic designation V or H; na = not applicable.

Figure 1.

Typical loaded wheel tracking (LWT) results.

Despite the ability of the LWT to relate laboratory result to field moisture sensitivity performance, the accuracy of the “pass/fail” criteria for screening mixtures is limited ( 6 , 16 –18). Further, researchers have questioned the capability of the LWT to simulate field exposure conditions and reliably predict moisture sensitivity of a wide range of asphalt mixtures ( 7 ). Among State Departments of Transportation that currently specify the LWT test for mix design, quality assurance testing, or both, none of them considers moisture conditioning protocols aside from testing under submerged conditions as specified in AASHTO T 324. Therefore, there is a need to perform evaluation of the LWT test by considering moisture conditioning protocols to screen a wide range of moisture-sensitive asphalt mixtures.

Objectives and Scope

The objective of this study was to evaluate the capability of the LWT test (Hamburg type) to evaluate the moisture susceptibility of asphalt mixtures with different moisture conditioning protocols. Specific objectives include:

evaluate the effect of asphalt binder type on moisture damage of asphalt mixtures,

evaluate the effect of aggregate type on moisture damage of asphalt mixtures, and

evaluate the effect of moisture conditioning protocols on moisture damage of asphalt mixtures.

To achieve the objectives of this study, asphalt binder and asphalt mixture experiments were conducted. An unmodified asphalt binder and a styrene-butadiene-styrene (SBS) modified asphalt binder meeting Louisiana specification for PG 67-22 and PG 70-22, respectively, were utilized in the asphalt binder experiment. Five conditioning levels were considered in the asphalt binder experiment, namely a control (i.e., short-term aging via rolling thin-film oven test [RTFO]); single freeze–thaw (FT-1)-; triple freeze–thaw (FT-3)-; MiST 3500 (M-35); and MiST 7000 (M-70) conditioning cycles. Frequency sweep at multiple temperatures and multiple stress creep recovery (MSCR) tests were performed to characterize rheological properties of each conditioned asphalt binder.

Seven 12.5 mm Level 2 asphalt mixtures were utilized with two asphalt binder types (unmodified PG 67-22 and SBS-modified PG 70-22) and three aggregate types (limestone, crushed gravel, and a semi-crushed gravel). Similar to the asphalt binder conditioning levels, five moisture conditioning levels were considered in the asphalt mixture experiment. These conditioning levels consisted of short-term aging (STA) of loose mixtures; single freeze–thaw-; triple freeze–thaw-; MiST 3500 cycles; and MiST 7000 cycles of compacted mixtures. Asphalt mixture samples were then evaluated using the LWT test.

Methodology

Figure 2 presents the experimental plan for the study. The detailed experimental plan is discussed below.

Figure 2.

Experimental plan.

Materials

Two asphalt binder types, unmodified PG 67-22, and SBS-modified PG 70-22 meeting Louisiana Department of Transportation and Development (DOTD) standard specification for asphalt binders were selected ( 19 ). Further, three aggregate types, limestone (absorption <2%), crushed gravel (absorption >2%, natural sand content >15%), and a semi-crushed gravel (absorption >2%, and natural sand content >15%) meeting specification for 12.5 mm NMAS were selected. It is noted that the semi-crushed gravel was selected such that all particles passing the No.4 sieve (4.75 mm) were crushed whereas those particles retained on the No.4 sieve (50%) were smooth and round aggregates.

Asphalt Binder Experiment

The asphalt binder experiment comprised the conditioning of the asphalt binder specimens and the subsequent rheological evaluation of the conditioned asphalt binders.

Moisture Conditioning

Five conditioning levels were evaluated in the asphalt binder experiment. The first conditioning level is the control based on STA of asphalt binders following the RTFO ( 20 ). RTFO-aged asphalt binder was heated to 160°C until it was sufficiently fluid and then poured into pressure aging vessel (PAV) pans to achieve a uniform thickness of 3.2 mm, and with dimensions similar to those specified in AASHTO R28. The specimens in the PAV pans were then subjected to the remaining four conditioning levels (Figure 3). The second and third conditioning levels included single- and triple-freeze–thaw conditioning, respectively. The fourth and fifth conditioning levels were MiST 3500 (3500 Moisture-induced Stress-Tester cycles) and 7000 (7000 Moisture-induced Stress-Tester cycles) conditioning cycles, respectively. Details of each conditioning cycle procedure are described in “Freeze–Thaw and MiST Conditioning” sections.

Figure 3.

Freeze–thaw and moisture-induced stress-tester (MiST) conditioning of asphalt binders.

Asphalt Binder Rheological Characterization

Each asphalt binder specimen subjected to one of the five conditioning levels was then rheologically evaluated through multiple temperatures and frequencies and MSCR tests. A minimum of three replicates were used in each test.

Asphalt Mixture Experiment

Seven 12.5 mm Superpave asphalt mixtures were designed utilizing two levels of asphalt binders and three levels of aggregates (Table 2). A Level 2 design (N_initial = 7, N_design = 65, N_final = 105 gyrations) was performed in accordance with AASHTO R 35, “Standard Practice for Superpave Volumetric Design for Hot Mix Asphalt (HMA),” AASHTO M 323, “Standard Specification for Superpave Volumetric Mix Design ( 21 ),” and Section 502 of the 2016 Louisiana Standard Specifications for Roads and Bridges ( 19 ). Specifically, the optimum asphalt cement content was determined based on volumetric properties (VTM = 2.5%–4.5%, VMA ≥13.5%, VFA = 69%–80%) and densification requirements (%G_mm at Ninitial ≤90, %G_mm at Nfinal ≤98). Among the asphalt mixtures evaluated, six (M1–M6) were laboratory produced and laboratory compacted, whereas one (M7) was plant produced and laboratory compacted. Mixtures M1, M2, and M3 consisted of unmodified PG 67-22 asphalt binder and limestone, crushed gravel, and semi-crushed gravel aggregates, respectively (Table 2). Mixtures M4, M5, and M6 included SBS-modified PG 70-22 asphalt binder and limestone, crushed gravel, and semi-crushed gravel aggregates, respectively (Table 2). Mixture M7 was plant-produced mixture prepared with PG 67-22 and limestone aggregate. It is noted that mixture M7 contained liquid anti-strip additive (Arr-Maz Products, Inc.) at a dosage rate of 0.6% by weight of mixture, and 19% recycled asphalt pavement (RAP) material (Table 2).

Table 2.

Louisiana Level 2 Asphalt Mixture Compositions

Mix id	Asphalt binder type	Aggregate type	Anti-strip additive	Moisture sensitivity
M1	PG 67-22¹	Limestone	na	L
M2		Crushed gravel	na	H
M3		Semi-crushed gravel	na	H
M4	PG 70-22¹	Limestone	na	L
M5		Crushed gravel	na	H
M6		Semi-crushed gravel	na	H
M7	PG 67-22^a	Limestone	0.6% LA-2	L

Note: % RAP = percent recycled asphalt pavement content; na =not applicable; LA-2 = liquid anti-strip additive; L = low moisture-susceptible aggregate (water absorption <2%); H = high moisture-susceptible aggregate (water absorption >2%).

Meeting 2016 Louisiana Department of Transportation and Development specifications for road and bridges.

Five conditioning levels were considered in the asphalt mixture experiment. The first conditioning level is the control and comprised STA of loose asphalt mixture samples according AASHTO R 30, “Standard Practice for Mixture Conditioning of Hot Mix Asphalt (HMA) ( 22 )” before compaction in the gyratory compactor. The other four conditioning levels were performed on compacted asphalt mixtures samples as follows.

Freeze–Thaw Cycle Conditioning

The freeze–thaw cycles were performed according to AASHTO T 283, “Standard Method of Test for Resistance of Compacted Asphalt Mixtures to Moisture-Induced Damage ( 23 ).” For the second and third conditioning levels, RTFO-aged asphalt binders and compacted short-term aged asphalt mixture samples were subjected to one and three freeze–thaw conditioning cycles, respectively. For each conditioning level, asphalt mixture specimens were partially vacuum saturated between 70 and 80%. Vacuum-saturated specimens were covered tightly with plastic wraps and placed in a freezer at a temperature of −18°C for 16 h. Further, asphalt mixture specimens were removed from the freezer and placed in a water bath at 60°C for 24 h. Asphalt binder specimens were conditioned without vacuum saturation or utilizing plastic wraps. It is noted for the three conditioning cycles, specimens were removed from water, tightly covered with plastic wraps, and then placed back in the freezer to repeat freeze–thaw cycles two more times. After conditioning, the specimens were removed from the 60°C water bath and placed in another water bath at 25°C before testing.

MiST Conditioning

The MiST conditioning was performed according to ASTM 7870, “Standard Practice for Moisture Conditioning Compacted Asphalt Mixture Specimens by Using Hydrostatic Pore Pressure ( 24 ).” For the fourth and fifth conditioning level, RTFO-aged and compacted asphalt mixture samples were conditioned at 3500 and 7000 cycles, respectively, in the MiST. Specimens were placed in the MiST and the chamber filled with water to the appropriate level. The specimens were kept in the machine at 60°C for 20 h to simulate adhesive failure in the mixture. Further, a pressure amplitude of 40 pounds per square inch was applied for 3500 and 7000 cycles, respectively, for the fourth and fifth conditioning levels. After conditioning, the specimens were removed from the MiST and placed in another water bath at 25°C before testing.

LWT Test

The LWT test was conducted in accordance with AASHTO T 324, “Standard Method of Test for Hamburg Wheel-Track Testing of Compacted Hot Mix Asphalt (HMA) ( 25 ).” This test is considered a torture one. It produces damage by rolling a 703 N (158 lb.) steel wheel across the surface of cylindrical specimens (150 mm diameter by 60 mm thick) that are submerged in 50°C water for 45 min. The test duration is 20,000 passes at a rate of 52 passes per minute. Four specimens (two specimens for each wheel) were tested. Rut-depth measurements were recorded at 11 locations across cylindrical specimen until failure. Then, rut-depth measurements at four middle locations were averaged. Further, rut depth at 20,000 cycles was recorded and used in the analysis. In addition, SIP is computed and reported as a measure of moisture damage for mixtures evaluated. Sample failure is considered at either 20,000 passes or 25 mm rut depth, whichever comes first. Further, LWT test results were analyzed according a procedure developed by Yin et al. ( 26 ) to separate the total rut depth into viscoplastic and stripping components of deformation. It is noted that the LWT device used in this paper was of a Hamburg type.

Results and Discussion

Results were statistically analyzed using analysis of variance (ANOVA) procedure provided in the Statistical Analysis System (SAS) 9.4 program ( 27 ). Multiple comparison (Tukey test) procedure with a confidence level of 95% was performed on the means. Groupings represent the mean for the test results reported by asphalt binder or asphalt mixture type. Results of the statistical grouping are reported with letters: A, B, C, and so forth, representing statistically distinct performance from best to worst. Multiple letter designations, such as A/B (or A/B/C) indicate that the difference in the means is not statistically significant.

Frequency Sweep Test

Effect of Conditioning on Asphalt Binder Stiffness

Figure 4a presents frequency sweep master curves for asphalt binders evaluated. For the two asphalt binders evaluated, freeze–thaw conditioning (FT-1 and FT-3) resulted in an increase in stiffness as compared with the RTFO-conditioned asphalt binder. Similar increase from RTFO was observed for MiST-conditioned (M-35 and M-70) asphalt binder. However, PG 67-22 asphalt binder exhibited higher increase in stiffness from RTFO for each conditioning level as compared with PG 70-22 asphalt binder, Figure 4a. It is noted that an increase in conditioning level (FT-1 to FT-3, and M-35 to M-70) resulted in increased stiffness.

Figure 4.

(a) Frequency sweep master curves, and (b) G*/Sinδ, _{50° C, 10 rad/s} for asphalt binders.

Figure 4b shows the rut factor, G*/Sinδ,_{50°C,10 rad/s}, values for asphalt binders considered. For PG 67-22 binder, it is noted that the rut factor values increased as conditioning level progressed from RTFO to FT-1 and FT-3. However, a slight increase in the rut factor was observed as conditioning level progressed from RTFO to M-35 and M-70. Further, for PG 70-22 asphalt binder, FT and MiST conditionings had minimal effect on rut factor values.

MSCR Test Results

Effect of Conditioning on Elastic Response

Figure 5, a and b, show percent recovery (R) at stress level of 3.2 kPa for asphalt binders evaluated. For the asphalt binders evaluated, FT-1 and FT-3 conditionings resulted in a minimal Jnr decrease and R increase as compared with RTFO-conditioned asphalt binder (Figure 5, a and b). A similar trend was observed for M-35 and M-70 conditionings.

Figure 5.

(a) Non-recoverable creep compliance, (b) percent recovery and (c) elastic response curve for multiple stress creep recovery at 67°C.

Figure 5c presents the elastic response curve for the asphalt binders evaluated. Two clusters for each binder type were identified: PG 70-22 in the passing zone and PG 67-22 in the failed zone. For the two clusters of asphalt binders in Figure 5c, freeze–thaw (FT-1 and FT-3) and MiST (M-35 and M-70) conditioning had no effect on the capability of the asphalt binder to meet the delayed elastic response criteria ( 28 ).

LWT Test Results

Effect of Conditioning on Moisture Resistance

Figure 6 presents the LWT rut depth of asphalt mixtures evaluated. Moisture-susceptible mixtures are expected to show higher rut depths in the LWT test. For each asphalt mixture evaluated, freeze–thaw (FT-1 and FT-3) and MiST (M-35 and M-70) conditioning resulted in increased rut depth as compared with the control asphalt mixture. An increase in conditioning level (FT-1 to FT-3, and M-35 to M-70) resulted in a significant increase rut depth. The increased rut depth is in conflict with that observed in asphalt binder test results (Figure 4). This observation may be attributed to the dominance of adhesive failure in the mixtures, and is consistent with that reported by other researchers ( 29 – 31 ). It has been demonstrated that for wet-conditioned specimens, the predominant failure mode in the bitumen bond strength test was adhesive failure ( 29 – 31 ). Cohesive failure is mainly dominated by asphalt binder response ( 3 , 32 –34).

Figure 6.

Loaded wheel tracking results—total rut depth for mixtures (a) M1–M3 and (b) M4–M7.

Mixtures M2, M3, M5, and M6 were prepared with moisture-susceptible aggregates per Louisiana DOTD Standard Specification ( 19 ). Further, LWT test results of mixtures M2, M5, and M6 showed compliance with specifications per AASHTO T 324. However, as samples were conditioned, an increase in rut-depth measurements was revealed that exceeded specification criteria. Conditioning levels considered on mixture M3 had significant effect on sample integrity (samples were damaged during conditioning), see Figure 6a. This observation is attributed to the use of 50% uncrushed material in M3.

Further, the addition of anti-strip additive in mixture M7 improved the moisture damage resistance as compared with Mixture M1 at all conditioning levels evaluated. Although mixtures M1, M4, and M7 were moisture resistant ( 19 ), an increase in freeze–thaw (FT-1 to FT-3) and MiST (M-35 and M-70) conditioning levels exposed them as moisture-susceptible mixtures. Therefore, it is relevant to incorporate a moisture conditioning protocol in AASHTO T324 test for moisture damage.

Figure 7 shows the SIP values of mixtures evaluated. Higher SIP values are desired for moisture-resistant asphalt mixtures. Mixtures with SBS polymer-modified PG 70-22 asphalt binder (M4, M5, and M6) performed well with no stripping damage. However, mixtures with unmodified PG 67-22 asphalt binder (M1, M2, M3, and M7) did see a reduction in SIP values with a progressive increase in moisture conditioning level. This observation is consistent with the field performance of asphalt mixtures containing SBS polymer-modified asphalt binder ( 35 , 36 ).

Figure 7.

Loaded wheel tracking results: stripping inflection point for mixtures (a) M1–M3, and (b) M4–M7.

Conditioning of mixture M3 at FT-3 and M-70 caused severe damage and sudden disintegration within few cycles of wheel passes (Figure 8). Thus, SIPs for these mixtures were selected as the number of passes to failure (Figure 7). For mixtures with no clearly defined stripping slope, selection of the total number of passes to total failure may be misleading. To address this issue, SIP is rendered invalid whenever the ratio of stripping slope to creep slope is less than 2.0 ( 37 ).

Figure 8.

(a) Loaded wheel tracking test results for M3, (b) disintegrated conditioned specimens.

Figure 9 presents LWT rut depths separated into viscoplastic and stripping components of deformation ( 26 ). It is noted that LWT rut depths were separated into stripping and viscoplastic components utilzing a procedure developed by Yin et al. ( 26 ). Among mixtures evaluated, only mixtures M1 (FT-3, M-35, and M-70) and M7 (FT-3 and M-70) exhibited stripping deformations. Further, mixture M1 had much higher stripping deformation than M7. In addition, progressive increase in MiST conditioning from M-35 to M-70 yielded an increase in the stripping and viscoplastic rut componenets for mixtures M1 and M7 (Figure 9). This observation is consistent with that reported by Santuci (38 –40). Pore pressure build-up in asphalt mixtures has the potential to emulsify and soften asphalt binder films and therefore result in increased rutting and stripping in asphalt mixtures (38 –40). Analysis reported by Yin et al. ( 26 ) could not separate total rut depth into stripping and viscoplastic component for certain mixtures with high rut depths (Figure 9). This may be attributed to the sudden disintegration of these mixtures within a limited number of wheel cycle passes (Figure 8).

Figure 9.

Viscoplastic and stripping rut-depth components for mixtures (a) M1–M3, and (b) M4–M7.

Table 3 presents a summary of computed parameters from LWT test results used by California, Iowa, Louisiana, Massachusetts, Texas, Utah, and Washington states. Those parameters are summarized in Table 1. Generally, performing the LWT test according to AASHTO T324 on samples that went through moisture conditioning was effective in capturing moisture damage as opposed to conducting the test as currently specified with no moisture conditioning. For example, results of LWT tests on mixtures M1, M2, M3, and M7 that were freeze–thaw (FT-3) and MiST (M-70) conditioned did not meet the corresponding State criteria for moisture damage (Table 3). However, those same mixtures did meet the specified State criteria when the test was conducted according to current AASHT T 324 protocol. It is noted that the use of SBS polymer-modified asphalt binder is a major contributor to moisture damage resistance.

Table 3.

Summary of Loaded Wheel Tracking Results

State	Test condition	M1		M2		M3		M4		M5		M6		M7
State	Test condition	R, max.	SIP, min.	R, max.	SIP, min.	R, max.	SIP, min.	R, max.	SIP, min.	R, max.	SIP, min.	R, max.	SIP, min.	R, max.	SIP, min.
Cal.	T 324	P	P	P	P	F	F	P	P	P	P	P	P	P	P
	T324/FT-1	P	P	P	P	F	F	P	P	P	P	P	P	P	P
	T324/FT-3	F	F	F	F	F	F	P	P	P	P	P	P	P	F
	T324/M-35	P	P	P	P	F	F	P	P	P	P	P	P	P	P
	T/324M-70	F	F	F	F	F	F	P	P	P	P	P	P	P	F
Iowa	T 324	P	P	P	P	F	I	P	P	P	P	P	P	P	P
	T324/FT-1	P	P	P	P	F	I	P	P	P	P	P	P	P	P
	T324/FT-3	F	F	F	I	F	I	P	P	P	P	P	P	P	F
	T324/M-35	P	F	P	P	F	I	P	P	P	P	P	P	P	P
	T/324M-70	F	F	F	I	F	I	P	P	P	P	P	P	P	F
La.	T 324	F	na	P	na	F	na	P	na	P	na	P	na	P	na
	T324/FT-1	F		F		F		P		P		P		P
	T324/FT-3	F		F		F		F		F		P		F
	T324/M-35	F		F		F		F		P		P		P
	T/324M-70	F		F		F		F		P		F		F
Mass.	T 324	P	P	P	P	F	F	P	P	P	P	P	P	P	P
	T324/FT-1	P	P	P	P	F	F	P	P	P	P	P	P	P	P
	T324/FT-3	F	F	F	F	F	F	P	P	P	P	P	P	P	F
	T324/M-35	P	F	P	P	F	F	P	P	P	P	P	P	P	P
	T/324M-70	F	F	F	F	F	F	P	P	P	P	P	P	P	F
Texas	T 324	P	N/A	P	N/A	F	N/A	P	N/A	P	N/A	P	N/A	P	N/A
	T324/FT-1	P		P		F		P		P		P		P
	T324/FT-3	P		F		F		P		P		P		P
	T324/M-35	P		P		F		P		P		P		P
	T/324M-70	F		F		F		P		P		P		P
Utah	T 324	P	na	P	na	F	na	P	na	P	na	P	na	P	na
	T324/FT-1	P		P		F		P		P		P		P
	T324/FT-3	F		F		F		P		P		P		P
	T324/M-35	P		P		F		P		P		P		P
	T/324M-70	F		F		F		P		P		P		P
Wash.	T 324	P	P	P	P	F	F	P	P	P	P	P	P	P	P
	T324/FT-1	P	P	P	P	F	F	P	P	P	P	P	P	P	P
	T324/FT-3	F	F	F	F	F	F	P	P	P	P	P	P	P	F
	T324/M-35	P	F	P	P	F	F	P	P	P	P	P	P	P	P
	T/324M-70	F	F	F	F	F	F	P	P	P	P	P	P	P	F

Note: Cal. = California; La = Louisiana; Mass = Massachusetts; Wash. = Washington; T324 = AASHTO T 324; / = And; FT-1 = one freeze–thaw cycle; FT-3 = three freeze–thaw cycles; M-35 = MiST 3500 cycles; M-70 = MiST 7000 cycles; R, max. = maximum rut depth at specified number of passes; SIP, min. = minimum stripping inflection point; P = passes specification; F = failed specification; na = not applicable; I = invalid stripping inflection point value ( 37 ).

Summary and Conclusion

The objective of this study was to evaluate the capability of the LWT test to evaluate the moisture susceptibility of asphalt mixtures with different moisture conditioning protocols. Asphalt binder and asphalt mixture experiments were conducted in the study. An unmodified PG 67-22 and a SBS-modified PG 70-22 asphalt binders were utilized in the asphalt binder experiment. Further, five conditioning levels, a control based on STA of asphalt binders following RTFO test; one freeze–thaw-; three freeze–thaw-; M-35; and M-70 conditioning cycles were considered. Frequency sweep at multiple temperatures and frequencies, and MSCR tests were performed to characterize rheological properties of asphalt binders.

Level 2 asphalt mixtures were prepared utilizing two asphalt binder types and three aggregate types. The two asphalt binder types selected consisted of unmodified PG 67-22 and SBS-modified PG 70-22. The three aggregate types included limestone, crushed gravel, and a semi-crushed gravel. Consistent with the asphalt binder conditioning levels, five moisture conditioning levels were considered in the asphalt mixture experiment. These conditioning levels included STA of loose mixtures; one freeze–thaw-; three freeze–thaw-; MiST 3500; and MiST 7000 conditioning cycles of compacted mixtures. The LWT test was used to evaluate the asphalt mixture samples.

Freeze–thaw and MiST conditioning of asphalt binders resulted in an increase in stiffness as compared with the RTFO-aged asphalt binders. For the mixtures evaluated, freeze–thaw and MiST conditioning resulted in an increase in rut depth compared with the control asphalt mixture. It is worth noting that the stiffening of the asphalt binders resulted in progressive damage of the asphalt mixtures in the LWT test. The stiffening effect of the asphalt binder may have contributed to moisture damage. The conditioning levels evaluated were effective in exposing moisture-sensitive mixtures, which initially showed compliance with LADOTD specification per AASHTO T324. Therefore, it is recommended that a moisture conditioning protocol is incorporated in AASHTO T324 test to ascertain moisture susceptibility of asphalt mixtures. SBS polymer-modified PG 70-22 asphalt mixtures performed well with no stripping damage, whereas unmodified PG 67-22 asphalt binder showed a reduction in SIP values with a progressive increase in moisture conditioning level. Specific observations include:

Generally, rut factor values for unmodified PG 67-22 asphalt binders increased with an increase level of freeze–thaw and MiST conditioning.

Freeze–thaw and MiST conditioning had no effect on rut factor of SBS-modified PG 70-22 asphalt binders.

PG 67-22 asphalt binder exhibited higher increase in stiffness from RTFO for each conditioning level as compared with PG 70-22 asphalt binder.

Freeze–thaw (FT-1 and FT-3) and MiST (M-35 and M-70) conditioning resulted in a minimal Jnr decrease and R increase as compared with RTFO-conditioned asphalt binder.

Two clusters for each binder type were identified in the MSCR elastic response curve: PG 70-22 in the passing zone and PG 67-22 in the failed zone.

For the two clusters of asphalt binders in Figure 5c, freeze–thaw (FT-1 and FT-3) and MiST (MiST 3500 and MiST 7000) conditioning had no effect on the capability of the asphalt binder to meet the delayed elastic response criteria.

An increase in conditioning level (FT-1 to FT-3, and M-35 to M-70) resulted in a significant increase rut depth.

The addition of anti-strip additive in mixture M7 improved the moisture damage resistance as compared with Mixture M1 at all conditioning levels evaluated.

Only mixtures M1 (FT-3, M-35, and M-70) and M7 (FT-3 and M-70) exhibited stripping deformations.

Mixture M1 showed much higher stripping deformation than M7.

Progressive increase in MiST conditioning from 3500 to 7000 yielded an increase in the stripping and viscoplastic rut componenets for mixtures M1 and M7.

For certain mixtures with higher rut depths, total rut depth could not be separated into stripping and viscoplastic component because of sudden disintegration of these mixtures within few cycles of wheel passes.

As the increased stiffness associated with moisture conditioning of the asphalt binders did not translate into improved moisture resistance in the conditioned asphalt mixtures, it is recommended that agencies rely on asphalt binder and mixture conditioning to ascertain the moisture susceptibility of asphalt mixtures. It is further recommended that this study is expanded to include a wide variety of asphalt mixture types to select the appropriate conditioning protocol to be included in AASHTO T 324 protocol for ascertaining moisture susceptibility of asphalt mixtures. In addition, the researchers recommend that additional test protocols such as asphalt binder bond strength (AASHTO T 361) and atomic force microscopy tests are used in the future to estimate the extent of cohesive and adhesive failure in each mixture caused by moisture damage.

Footnotes

Acknowledgements

The presented study was part of LTRC Project No. 19-2B, “Development of a Moisture Sensitivity Test for Asphalt Mixtures.” The authors would like to acknowledge the support of the Louisiana Transportation Research Center (LTRC).

Author Contributions

The authors confirm contribution to the paper as follows: study conception and design: L. Mohammad and M. Akentuna; data collection: S. Sachdeva, M. Akentuna, and L. Mohammad; analysis and interpretation of results: L. Mohammad, M. Akentuna, M., S. Sachdeva, and S. Cooper III; draft manuscript preparation: M. Akentuna, L. Mohammad, S. Sachdeva, and S. Cooper, Jr. All authors reviewed the results and approved the final version of the manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received Funding for this research from the Louisiana Department of Transportation and Development through the Louisiana Transportation Research Center.

ORCID iDs

Moses Akentuna

Louay N. Mohammad

Samuel B. Cooper, Jr.

References

Kim

Y. R.

Lutif

J. S.

Bhasin

Little

D. N.

Evaluation of Moisture Damage Mechanisms and Effects of Hydrated Lime in Asphalt Mixtures through Measurements of Mixture Component Properties and Performance Testing. Journal of Materials in Civil Engineering, Vol. 20, No. 10, 2008, pp. 659–667.

Camacho-Garita

Aguiar-Moya

J. P.

Ávila-Esquivel

Loría-Salazar

L. G.

Effect of Moisture on Full-Scale Pavement Distress. Journal of Testing and Evaluation, Vol. 48, No. 1, 2020, pp. 235–246.

Hicks

R. G.

Santucci

Aschenbrener

Moisture Sensitivity of Asphalt Pavements: A National Seminar. San Diego, CA, 2003, pp. 2–21.

Terrel

R. L.

Shute

J. W.

Summary Report on Water Sensitivity. Report No. SR-OSU-A-003A-89-3. Oregon State University, 1989.

Cooley

L. A.

Kandhal

P. S.

Buchanan

M. S.

Fee

Epps

Loaded Wheel Testers in the United States: State of the Practice. Transportation Research Board, National Research Council; 2000.

Aschenbrener

Evaluation of Hamburg Wheel-Tracking Device to Predict Moisture Damage in Hot-Mix Asphalt. Transportation Research Record: Journal of the Transportation Research Board, 1995. 1492: 193.

West

Rodezno

Leiva

Yin

Development of a Framework for Balanced Mix Design. NCHRP Project 20-07(406). Washington, D.C., 2018.

Apeagyei

Buttlar

W. G.

Dempsey

Moisture Damage Evaluation of Asphalt Mixtures Using AASHTO T283 and DC(T) Fracture Test. Proc., 10th International Conference on Asphalt Pavements. International Society of Asphalt Pavements, Red Hook, NY., 2006, pp. 862–873.

Saadeh

Eljairi

Development of a Quality Control Test Procedure for Characterizing Fracture Properties of Asphalt Mixtures. Project Report. Project 10-24. California State University, Long Beach, 2011.

10.

Azaril

Mohseni

Improvement of the Test Method for Determining Moisture Damage Resistance. Multi-Scale Modeling and Characterization of Infrastructure, Vol. 8, 2013, pp. 109–125.

11.

Dave

Baker

Moisture Damage Evaluation of Asphalt Mixes that Contain Mining Byproducts. Transportation Research Record 2371. TRB, National Research Council, Washington, D.C., 2013, pp. 113–120.

12.

Al-Qadi

I. L.

Abuawad

I. M.

Dhasmana

Coenen

A. R.

Trepanier

J. S.

Effects of Various Asphalt Binder Additives/Modifiers on Moisture-Susceptible Asphalt Mixtures. Report FHWA-ICT-14-004. Illinois Center for Transportation, Illinois Department of Transportation, Springfield, 2014.

13.

Mohammad

L. N.

Elseifi

M. A.

Raghavendra

Hamburg Wheel-Track Test Equipment Requirements and Improvements to AASHTO T 324. National Cooperative Highway Research Program, Report No. Web-Document 219. Transportation Research Board, Washington, D.C., 2015.

14.

Solaimanian

Harvey

Tahmoressi

Tandon

Test Methods to Predict Moisture Sensitivity of Hot-Mix Asphalt Pavements. In Moisture Sensitivity of Asphalt Pavements-A National Seminar California Department of Transportation, Federal Highway Administration, National Asphalt Pavement Association, California Asphalt Pavement Alliance and Transportation Research Board, 2003.

15.

Stuart

K. D.

Mogawer

W. S.

Effect of Compaction Method on Rutting Susceptibility Measured by Wheel-Tracking Devices. Presented at the 76th Annual Meeting of the Transportation Research Board, Washington, D.C., 1997.

16.

Aschenbrener

Currier

Influence of Testing Variables on the Results from the Hamburg Wheel-Tracking Device. Final Report No. COOT -D1D-R-93-22. Colorado Department of Transportation, Denver, 1993.

17.

Walubita

L. F.

Fuentes

Prakoso

Pianeta

L. M.

Komba

J. J.

Naik

Correlating the HWTT Laboratory Test Data to Field Rutting Performance of In-Service Highway Sections. Construction and Building Materials, Vol. 236, 2020, p. 117552.

18.

Walubita

L. F.

Fuentes

Lee

S. I.

Dawd

Mahmoud

Comparative Evaluation of Five HMA Rutting-Related Laboratory Test Methods Relative to Field Performance Data: DM, FN, RLPD, SPST, and HWTT. Construction and Building Materials, Vol. 215, 2019, pp. 737–753.

19.

Louisiana Department of Transportation and Development. Louisiana Standard Specifications for Roads and Bridges. Louisiana Department of Transportation and Development, Baton Rouge, 2016.

20.

AASHTO T 240. Standard Method of Test for Effect of Heat and Air on a Moving Film of Asphalt Binder (Rolling Thin-Film Oven Test). American Association of State Highways and Transportation Officials, Washington, D.C., 2013.

21.

AASHTO

323. Standard Specification for Superpave Volumetric Mix Design. American Association of State Highways and Transportation Officials, Washington, D.C., 2017.

22.

AASHTO

30. Standard Practice for Mixture Conditioning of Hot Mix Asphalt (HMA). American Association of State Highways and Transportation Officials, Washington, D.C., 2002.

23.

AASHTO T 283. Standard Method of Test for Resistance of Compacted Asphalt Mixtures to Moisture-Induced Damage. American Association of State Highways and Transportation Officials, Washington, D.C., 2014.

24.

ASTM D7870. Standard Practice for Moisture Conditioning Compacted Asphalt Mixture Specimens by Using Hydrostatic Pore Pressure. American Society for Testing Materials, West Conshohocken, PA, 2013.

25.

AASHTO T 324. Standard Method of Test for Hamburg Wheel-Track Testing of Compacted Asphalt Mixtures. American Association of State Highways and Transportation Officials, Washington, D.C., 2019.

26.

Yin

Arambula

Lytton

Martin

A.E.

Cucalon

L.G.

Novel Method for Moisture Susceptibility and Rutting Evaluation Using Hamburg Wheel Tracking Test. Transportation Research Record: Journal of the Transportation Research Board, 2014. 2446: 1–7.

27.

SAS Institute Inc. Base SAS 9.4 Procedures Guide: Statistical Procedures. SAS Institute Inc., Cary, NC, 2013.

28.

Kabir

M. S.

King

Jr.

Validity of Multiple Stress Creep Recovery (MSCR) Test for DOTD Asphalt Binder Specification. Final Report 564. Louisiana Transportation Research Center, 2017.

29.

Kanitpong

Bahia

H. U.

Role of Adhesion and Thin Film Tackiness of Asphalt Binders in Moisture Damage of HMA (With Discussion). Journal of the Association of Asphalt Paving Technologists, Vol. 72, 2003, pp. 502–528.

30.

Zhang

Airey

G. D.

Grenfell

Apeagyei

A. K.

Moisture Sensitivity Examination of Asphalt Mixtures Using Thermodynamic, Direct Adhesion Peel and Compacted Mixture Mechanical Tests. Road Materials and Pavement Design, Vol. 19, No. 1, 2018, pp. 120–138.

31.

Apeagyei

A. K.

Grenfell

J. R.

Airey

G. D.

Moisture-Induced Strength Degradation of Aggregate–Asphalt Mastic Bonds. Road Materials and Pavement Design, Vol. 15, Supp1ement 1, 2014, pp. 239–262.

32.

Caro

Masad

Bhasin

Little

D. N.

Moisture Susceptibility of Asphalt Mixtures, Part 1: Mechanisms. International Journal of Pavement Engineering, Vol. 9, No. 2, 2008, pp. 81–98.

33.

Lytton

R. L.

Masad

E. A.

Zollinger

Bulut

Little

D. N.

Measurements of Surface Energy and Its Relationship to Moisture Damage. FHWA/TX-05/0-4524-2. Texas Transportation Institute, Austin, 2005.

34.

Moraes

Velasquez

Bahia

H. U.

Measuring the Effect of Moisture on Asphalt–Aggregate Bond With the Bitumen Bond Strength Test. Transportation Research Record: Journal of the Transportation Research Board, 2011. 2209(1): 70–81.

35.

Blankenship

P. B.

Myers

A. H.

Clifford

A. S.

Thomas

T. W.

King

H. W.

King

G. N.

Are All PG 70-22s the Same? Lab Tests on KY I-64 Field Samples. Journal of the Association of Asphalt Paving Technologists, Vol. 67, 1998, pp. 493–552.

36.

Harvey

J. T.

Evaluation of Hamburg wheel-tracking device test with laboratory and field performance data. Transportation Research Record: Journal of the Transportation Research Board, 2006. 1970: 25–44.

37.

Iowa Department of Transportation. Standard Specifications for Highway and Bridge Construction. Iowa Department of Transportation, 2015.

38.

Santucci

Moisture Sensitivity of Asphalt Pavements. Tech Topics, 2002.

39.

Wang

Luo

Yan

Wang

Evaluation on Moisture Sensitivity Induced by Dynamic Pore Water Pressure for Asphalt Mixture and Its Components Using the Bending Beam Rheometer Method. Construction and Building Materials, Vol. 251, 2020, p. 118942.

40.

Yang

Investigation of the Excess Pore Water Pressure Inside Compacted Asphalt Mixture by Dynamic Triaxial Tests. Construction and Building Materials, Vol. 138, 2017, pp. 363–371.