Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
5 A comprehensive literature review was previously conducted as part of NCHRP 09-40 (1). In this follow-up study, a review was conducted to identify various factors affecting interface bonding between pavement layers for both new HMA pave- ments and overlays on new, existing, and milled HMA and for PCC pavements. Results of the recently completed NCHRP Project 09-40 (1) study were also used in the experimental factorial design. 2.1 Tack Coat Materials According to ASTM D8-02, Standard Terminology Relating to Materials for Road and Pavements, âtack coat (also known as bond coat) is an application of bituminous material to an existing relatively nonabsorptive pavement surface able to provide a thorough bond between old and new surfacingâ (3). An adequate interface bonding between HMA overlay and underlying pavement layers is essential for satisfactory pave- ment performance. The main purpose of using tack coat is to provide necessary adhesive bonding between adjacent pave- ment layers to ensure that they behave as a monolithic system able to withstand traffic and environmental loadings. The surface receiving tack coat can be HMA or PCC pavement. The tack coat can also be applied onto a new HMA surface immediately before placement of the next layer, that is, a leveling or surface course, in a multilayer asphalt pavement structure. These tack coats are typically polymer-modified or unmodified water-based asphalt emulsions. Selection of tack coat materials in practice is mainly based on local experience, engineering judgment, or both. Historically, performance- graded or hot paving asphalt cement, cutback asphalt, and emulsified asphalt have been used as tack coat materials. However, cutback asphalts, a combination of asphalt cement and petroleum distillates, that is, kerosene or diesel oil, are not normally used because of numerous environmental concerns and loss of high-energy products (5). Hot paving or performance-graded asphalt cements are not popular for tack coat application, as they need to be fluid enough to spray and require immediate application in front of the paver to avoid quick cooling. Those cements are commonly used when a geosynthetic interlayer is used and for new rubber- ized HMA overlays (12). Emulsified asphalt or asphalt emul- sion is a nonflammable liquid substance that is produced by combining asphalt cement with water using emulsifying agents or surfactants, such as soap, dust, and certain colloidal clays. Emulsifying agents can act as a stabilizing agent, because asphalt globules or droplets in emulsion should be dispersed in colloidal form during storage, mixing, handling, and pumping. Emulsified asphalt is the most widely used tack coat material, because it offers various benefits, such as ease of handling, economic, energy saving, reduced environmental impacts, and higher personnel safety (2). Emulsified asphalts have advantages over hot asphalts and cutback asphalts, in that they can be used with cold or heated aggregate and with aggregate that is dry or damp. In addition, emulsified asphalt does not need to be at an elevated temperature for proper application, and thus fire hazard is eliminated, and setup is much quicker than cutback asphalt (2). 2.1.1 Classification of Tack Coat Materials Asphalt emulsions are typically classified on the basis of their particle charge and setting properties. According to ASTM D977, the two most commonly used types of emulsi- fied asphalts are anionic and cationic. An advantage of using ionic emulsion over nonionic emulsion is that ionic emul- sion offers fast and effective bonding when it is used with aggregates that hold an opposite charge on the surface (18). In addition, cationic emulsions are preferable to use in damp weather, because they are less sensitive to moisture and tem- perature. Results of a worldwide survey indicated that cationic emulsions were the most frequently used tack coat material (13, 14). Both anionic and cationic emulsions are further graded on the basis of their setting properties. The type and C H A P T E R 2 State of Practice
6 the amount of the emulsifying agent in the emulsion control the setting rate. The emulsions currently available in practice are SS, medium setting (MS), RS, high float, polymer modi- fied, and latex modified (L). Anionic grade emulsions are RS-1, HFRS-2, RS-2, MS-1, HFMS-2, MS-2, MS-2H, SS-1 and SS-1H; cationic grade emulsions are CRS-1, CRS-2, CMS-2, CMS-2h, CSS-1 and CSS-1H (5). RS emulsions are commonly used during night paving and cold weather because of their quick setting time. Studies have shown that SS grades (SS-1, S-1H, CSS-1, and CSS-1H) and RS grades (RS-1, RS-2, CRS-1, CRS-2, CRS-2P and CRS-2L) are the most common type of emulsions used as tack coats in the United States. Hot paving asphalt cements (AC-20 and AC-30) and cut- back asphalts are also occasionally used as tack coat materials in some states (6, 15, 16). A recent worldwide survey (17) on the current state of tack coat practices indicated that 100% of responding agencies permitted use of asphalt emulsions and 27% and 20% of responding agencies used performance- graded asphalt cement and cutback asphalts as tack coats, respectively. The most commonly used emulsified asphalts were SS-1 (41%), SS-1H (39%), CSS-1 (37%), and CSS-1H (41%). The most used hot paving asphalt cements and cutback asphalts were PG 64-22 (60%) and RC-70 (55%), respectively (17). 2.1.2 Tack Coat Dilution Diluted emulsion provides additional volume at a given residual rate needed for the distributor to function properly at low application rate while driving at normal speeds. This emulsion flowing easily from the distributor at ambient temperatures allows for a more uniform application (7, 8). An advantage of using SS grades over the RS grades is that they can be diluted. However, diluted SS emulsions are not recommended for use in cold weather, because they may take several hours to break or even days to set completely when compared to RS emulsions. In addition, an overlay tacked with SS emulsion and exposed to heavy traffic immediately after construction could experience excessive slippage in a short period (7). Different state DOTs have different opinions about tack coat dilution. Ohio DOT reported that only SS emulsion should be diluted (10). However, Texas DOT rec- ommends that emulsified tack coats should not be diluted; if diluted, they must be diluted by the supplier, not by the contractor at the site (11). 2.1.3 Tack Coat Breaking and Setting Time An asphalt emulsion is brown in color, as it contains asphalt cement, water, and an emulsifying agent. When the emulsified asphalt blends and reacts with the surface of aggregates, water separates from the emulsion, and the color of the emulsion changes from brown to black. Evaporation of water causes asphalt emulsion to break or set and to create a continuous coating of asphalt on a pavement surface (19). In general, an asphalt emulsion takes 1 to 2 hours to set completely; how- ever, the literature lacks consensus concerning how long a tack coat should remain uncovered before the subsequent asphalt layer is placed. The survey by the International Bitu- men Emulsion Federation indicated that the required time interval between the application of a tack coat material and the placement of the next asphalt layer typically ranges from 20 minutes to several hours, depending on the type of tack coat (14). Paul and Scherocman (15) reported that many state DOTs specified a minimum time between tack coat application and placement of HMA to provide adequate curing period for the emulsion to break and set. Three state DOTs specified a maximum time that a tack coat could be left before placement of asphalt concrete. Alaska DOT specified a maximum setting period of 2 hours for CSS-1 and a maximum of 72 hours for SS-1. Texas DOT specified a maximum of 45 minutes for SS-1 and MS-2. Four states indicated that paving was required on the day of tack coat application. It is generally recognized that an emulsion should completely set before the next asphalt layer is placed, as it has been verified in laboratory studies using both laboratory fabricated specimens and field extracted samples that greater ISS can be achieved with longer curing times (24, 26). However, experience has also shown that new HMA can be placed on top of unset tack coat and even over an unbroken emulsion with no detrimental effect on pavement performance (19). In Europe, the emulsified tack coat is often applied to pavement surface underneath the paver just before placing the HMA overlay. Several European companies have used this paving process with conventional dense graded HMA mixtures and normal emulsified tack coat application rates without negative consequences, despite concerns with water vapor passing through a dense graded mat (21). In the United States, this emulsion spray method is used in the Novachip⢠construction process, as reported by Estakhri and Button (48, 49). In addition, emulsified tack coats may also be applied on damp pavement, in cold weather, or both; however, the time required to set may increase (10). 2.2 Characterization of Interface Bond Strength Three basic modes of failures (shear, tension, and torsion) can occur at the pavement interface under traffic and environ- mental loading. Several methods, including laboratory testing, nondestructive testing, and theoretical analysis using field data, can be used to evaluate the interface bonding between
7 pavement layers. Moreover, several performance-related test tools are available to assess the bonding characteristics of tack coat materials. Each type of laboratory testing is intended to capture various interface failure modes that may occur in actual field conditions. Of the tests, the direct shear test is the most used and most representative test to measure interface bonding between pavement layers. In some cases, the direct tension test is also performed, because under service con- ditions, pavement interface failure is caused by both shear and tension (38). In fact, the interface bonding between pavement layers depends on several influencing parameters. The follow- ing factors must be taken into account during construction, the placement of an overlay to achieve satisfactory long-term pavement performance, or both. 2.2.1 Tack Coat Application The proper application of tack coat is important to high- quality asphalt pavement construction. To guarantee proper interface bonding during construction of overlays, tack coat materials are usually applied at interfaces. According to the Asphalt Institute, an application of tack coat material is rec- ommended when an HMA overlay is placed over an existing HMA or PCC pavement surface. However, tack coat may not be required when adequate bond strength can be developed, that is, when an additional layer is placed within 2 or 3 days of a freshly placed asphalt surface course that has not been exposed to traffic (2, 11). In general, tack coats are typically polymer-modified or unmodified water-based asphalt emul- sions. Mohammad et al. (27) reported that applying certain types of tack coat provided improved interface bond strength compared to the cases without tack coat application. Further, nontracking RS tack coats provided the highest bonding strengths, while CRS-1 yielded the lowest bonding strength. Akhtarhusein et al. (28) showed that the absence of tack coat severely hindered the development of bonding between adja- cent layers with undue slippage as a result. Leng et al. (32, 33) found that both performance grade (PG 64-22) and polymer- modified SS (SS-1HP) tack coat provided lower interface strains and better rutting resistance than RC-70 cutback asphalt. Many studies suggested that modified emulsions have the potential to improve interface bonding significantly, fol- lowed by nonmodified emulsions and nontacked interfaces (24, 25, 30, 36, 41, 53â60). In contrast, some studies showed that the improvement of interlayer adhesion property was not significant when the performance of tacked interface with various tack coat types was compared with nontacked inter- face condition (24, 29, 30, 61). In some cases, it was found that nontacked interfaces seemed to exhibit slightly higher shear resistance in comparison with tacked interfaces with unmodified emulsions (30, 61). It has been experimentally shown that modified emulsions yielded higher interface shear resistance than performance-graded emulsions (31), whereas other studies concluded that both interface treatments had comparable maximum interface bonding (27, 57). These observations could be affected by differences in the asphalt products, sample preparation techniques, and the mechanical performance of the residual asphalt cement (52, 55). A review of existing studies revealed the difficulties in select- ing the optimal quantity and quality of tack coat product to be applied at the interface. In particular, excessive tack coat may weaken the interface by introducing a slipping plane instead of working as a bonding agent and may thus promote shear slippage as it can act as a lubricating agent. Further, this situ- ation may cause compaction difficulties during construction because of movement of the HMA layer under heavy load of compactor. Furthermore, excessive tack may migrate into an overlaying HMA mat and thus affect mixture properties dur- ing construction. In contrast, insufficient tack coat may result in high tensile stresses at the bottom of the overlay because of poor interface shear resistance (2, 4, 11). Many authors confirm that an optimum amount of tack coat depends not only on properties of the residual asphalt binder but also on pavement surface characteristics, mixture properties, temper- ature, and many other influencing factors (22, 24, 26, 27, 31, 55, 62). Therefore, all aspects of tack coat application should be carefully considered and controlled during construction for optimum pavement performance. The selection of an optimum tack coat material and appli- cation rate is critical to the development of proper interface bonding; however, currently, the selection is mainly made on the basis of experience and engineering judgment. Most important, it is the residual amount of asphalt, not the quan- tity of diluted asphalt emulsion, that should be specified in the practice of tack coat application. It has been widely rec- ognized that different pavement surface typesâfor example, new, existing, and milled HMA and PCCârequire different application rates to produce the maximum interface bonding (1). Further, the amount of tack coat to be applied in terms of residual asphalt content is a function of the existing pavement surface conditions (24). Paul and Scherocman (15) reported that the residual application rates of the emulsions varied between 0.01 and 0.06 gallons per square yard (gsy), depend- ing on the type of pavement surface receiving tack coats. Leng et al. (32, 33) reported that 0.05 gsy was found to be the optimum residual application rate for the HMA-PCC inter- face. According to the survey by the International Bitumen Emulsion Federation (13, 14), the residual asphalt content of tack coats applied on conventional asphalt surfaces ranged between 0.02 and 0.09 gsy. The Asphalt Institute specifications (4) for tack coats reported that the total application rates ranged from 0.05 to 0.15 gsy for an emulsion diluted with water at
8 a 1:1 ratio, which was equivalent to residual application rates between 0.02 and 0.05 gsy. Lower application rates are recommended for new layers, while the intermediate range is for normal surface conditions on an existing relatively smooth pavement. The upper limit is for old, oxidized, cracked, pocked, and milled asphalt pave- ments and grooved or tinned PCC pavements. The residual asphalt contents, as specified in the Hot-Mix Asphalt Paving Handbook (19), should be within the range of 0.04 to 0.06 gsy. Open-textured surfaces require more tack coat than surfaces that are tight or dense. Dry or aged surfaces require more tack coat than those that are fat or flushed. A milled surface would require even more residual asphalt because of the increased specific surface area, as much as 0.08 gsy. Only half as much residual asphalt is typically required for new HMA layers, that is, 0.02 gsy (16, 19). Ohio DOT reported that typical tack coat application rates for various pavement types using SS asphalt emulsions (SS-1, SS-1H) varied from 0.03 to 0.08 gsy (10). The Asphalt Handbook states that the tack coat should be heated to the specified temperature so that it is fluid enough for proper application. The coat must uniformly cover the entire pavement surface and be cured readily before subsequent construction to generate adequate interface bonding (4). Moreover, proper application of tack coats requires properly calibrated application equipment. If the distributor has not been used for a while, the operator should place a trial tack coat application over some convenient, unused area to ensure that all nozzles are open and working properly. In addition, the distributor application rate needs to be calibrated both in longitudinal and transverse directions with the procedure described in ASTM D2995 (20). Operators should adjust the spray bar height throughout the day, depending on the amount of emulsion in the tank, truck speed, nozzle con- figuration, and application pressure. One perpetual problem often associated with tack coat application using distributor trucks is tracking, which is picking up a tackâdirt mixture on the tires of construction vehicles and equipment and thereby leaving existing pavement surface with little or no tack coat in the wheel paths, where it is most important. Many methods address the tracking problem (69). One method is to apply tack coat to pavement surface underneath the paver just ahead of the screed. This type of application can be achieved with a special paver fitted with a tack coat spray bar. A material trans- fer vehicle may also be used to address the haul truck pickup problem. Further, the most effective solution to the tracking problem is the use of modified tack coat material that is free of stickiness or pickup. An example of such a tack coat material is a patented procedure called COLNET, developed by Colas in France (21). The COLNET procedure was reported to allow immediate trafficking after spraying by employing a clean- bond cationic asphalt emulsion (Colacid R 70 C) with very fast and controlled breaking agents. 2.2.2 Pavement Surface Characteristics Mechanical bonding between adjacent pavement layers is highly dependent on the characteristics of existing pave- ment surface before the construction of overlays. It has been experimentally confirmed that the surface micro- and macro- textures of the bottom layer seem to play important roles in the development of proper interface bonding (31, 39, 59, 61). Raab et al. (37, 38) reported that the bond between pavement layers depends on not only the mechanical properties of binder but also aggregate interlock at the interface. In this regard, lower interlayer shear resistance can be related to lower surface roughness (59). It was found that, for equivalent test condi- tions, milled surfaces provided significantly higher interface bond strengths than nonmilled surfaces because of high sur- face roughness (9, 26, 31). Mohammad et al. (1) showed that surface texture was observed to correlate well with the ISS test results. Milled HMA surface yielded the highest inter- face bonding, followed by PCC, existing HMA, and new HMA. Akhtarhusein et al. (28) reported that similar pavement sur- faces (AC-AC) yielded higher bond strength than other surfaces (AC-PCC). According to A Basic Asphalt Emulsion Manual (2), the pavement surface receiving tack coat must be clean and free of loose materials to obtain the best bonding performance. Existing and milled HMA or PCC pavement surfaces can be fairly dirty and dusty. Leng et al. (32, 33) reported that sur- face cleaning method had a significant influence on interface bonding between HMA-PCC pavements, and the direction of tining had a negligible effect on interface bonding. McGhee and Clark (34) showed that poor interface bond strength was associated with unsound or dirty underlying surface or both. If the tack coat is sprayed without properly cleaning or washing of existing surface, then tack may adhere to the loose materials rather than working as an adhesive agent. Then the tracking problem can be exacerbated (10); sweeping the surface using a power broom is recommended before tack coat applica- tion (8, 19). Ohio DOT reported that slippage cracking and delamination were typical distresses observed when tack coat had been applied without proper surface preparation. In sum- mary, the literature suggests that, to ensure the best possible interface bonding between layers, pavement surfaces should be thoroughly cleaned before construction of overlays. 2.2.3 Field-Related Factors Interface bonding between pavement layers and the effec- tiveness of tack coat materials can be potentially influenced by several other field-related factors, such as temperature, tack coat curing period, presence of moisture, and mixture prop- erties, as well as shear load and loading rate. Many studies investigated the effect of temperature on interface bonding,
9 since asphalt cement is a thermoplastic material. Researchers reported that both interface shear strength and shear stiffness are highly temperature dependent for interfaces with or with- out tack coats (1, 22, 23, 24, 30, 31, 38, 47, 57). Canestrari and Santagata (30) stated that an increase in shear resistance was observed with decreases in test temperature. Chen and Huang (35) and Collop et al. (36) reported that shear resistance and reaction modulus increased with decreasing temperature and increasing normal stress. Further, results of repeated loading showed higher fatigue life and greater sensitivity to shear stress level at lower temperature (36). Tschegg et al. (23) investi- gated fracture characteristics of interface bonding between asphalt concrete layers at different test temperatures. Their study reported results of plastic ductile fracture behavior of the bonding area at 10.5°C; however, with decreasing tem- perature, linear elastic behavior was observed because of the higher modulus of elasticity at the interface caused by embrittlement of the tack coat. Hachiya and Sato (24) and Sholar et al. (26) conducted several field studies to investigate the effect of tack coat curing period on interface bonding condition. They reported that longer tack coat curing time yielded greater interface bond strengths (26). However, the time interval between the con- struction of wearing course and binder course may influence the interface bond strength because of an accumulation of dirt at the interface (24). Johnson et al. (41) stated that insufficient tack coat curing and the presence of contamination during construction could lead to premature tack failures. A few studies investigated the effect of moisture on interface bonding; however, there is no agreement between authors. Sholar et al. (26) experimentally showed that presence of water (i.e., rainwater) on pavement surface before tack coat application was potentially detrimental to the performance of asphalt mixture and could reduce the interface bonding strength significantly. With respect to tack coat curing time, the bonding at a wet interface may increase with time but not at the same rate when compared with equivalent sections without water. Raab et al. (37, 38) reported that the pres- ence of moisture at the interface could reduce interface shear resistance. Further, the effect of moisture at low temperature (â20°C) was found to be critical in terms of interlayer bond properties and led to weakening of pavement structure. Raab et al. (37, 38) revealed that the influence of age and traffic volume below design limits was beneficial to the inter- face bonding. Hence, traffic, in terms of vertical loading, influences bonding between pavement layers. To identify the effect of traffic conditions, several studies investigated the effect of normal stress on the interface frictional perfor- mance. Uzan et al. (22) reported that shear resistance of the interface increased significantly with increasing normal stress and decreased with increasing temperature. Canestrari and Santagata (30) showed that the dilatancy phenomenon was dependent upon the applied normal stress and temperature. An increase in the applied normal stress caused an increase in the peak shear stress with reduced dilatancy. Similarly, many authors (1, 30, 31, 59, 62) reported that interface bond- ing increased with increasing normal stress as a result of improved frictional characteristics of the interface. West et al. (31) evaluated the effect of mixture properties on interface bonding. It was reported that fine graded asphalt mixture [0.19-in. nominal maximum aggregate size (NMAS)] exhibited higher bond strength when compared with a coarse graded asphalt mixture (0.75-in. NMAS). In contrast, Sholar et al. (26) showed that fine graded overlay mixtures were found to produce significantly lower bonding strength than coarse graded mixtures. Leng et al. (32, 33) found that a 9.5-mm (0.37-in.) NMAS surface mixture yielded greater interface bonding than 19-mm (0.75-in.). NMAS surface mixture on the HMA-PCC pavements. Chen and Huang (35) showed that dense graded asphalt concrete yielded high shear resistance compared to open and gap graded mixtures. Further, an increase in film thickness led to decrease in peak shear stress and reaction tangential modulus. Mohammad et al. (1) reported that at high temperature ISS depends on mixture characteris- tics not tack application rates. West et al. (31) also investigated the effectiveness of three distribution methodsâhand wand sprayer, distributor truck spray bar, and Novachip⢠spreaderâin tack coat application. Results of this study showed that the Novachip spreader tack coat distribution method yielded better interface bonding performance than other application methods. Salinas et al. (40) evaluated the effect of two types of paving proceduresâ conventional paver and spray paverâand reported that influences of paving procedures and cleaning methods on interface bonding were insignificant. 2.2.4 Nondestructive Testing Interface bond condition can considerably affect stress and strain distribution in a multilayer pavement structure. A few attempts using specialized devices have been made to assess the magnitude of interface bonding nondestructively, with various degree of success. Hakim et al. (42) used FWD deflec- tion data for prediction of the bonding condition between asphalt pavement layers. They reported that the FWD back- calculated stiffness was lower than that measured in the laboratory. This difference was attributed to the fact that the conventional backcalculation procedure assumed full bond- ing between asphalt concrete layers. Results of theoretical analysis showed that assuming full bonding instead of using the actual condition would have changed the backcalculated stiffness values of asphalt layers and subbase by up to 50%. To address this issue, the interface shear bond stiffness was considered in a modified FWD backcalculation method.
10 Further, improved bonding condition with service time was reported after retesting the evaluated test sections. An objective of the Strategic Highway Research Program 2 (51) was to identify and to develop rapid nondestructive techniques (NDTs) to determine the extent and depth of delamination and discontinuities in HMA pavements. Several NDT methods including ground penetration radar, infra- red thermography, mechanical wave methods, and deflec- tion measurement methods were evaluated. To predict the responses for debonding and stripping, theoretical modeling was utilized for each test method. Authors of this study con- cluded that none of the NDT technologies could conclu- sively distinguish between types of pavement discontinuities. Results of FWD backcalculation showed that pavement deflec- tion changed with depth of delamination; FWD time history data were recommended as an NDT tool to identify pavement discontinuities. 2.2.5 Theoretical Analysis Several theoretical analyses were conducted to evaluate the advantages of strongly bonded HMA overlays in a multi- layer flexible pavement system (1, 22, 47). The finite element method (FEM) is able to provide a more realistic and closer approximation of how a pavement responds to loading com- pared to multilayer elastic theory analysis, because the FEM can model complex systems with fewer assumptions. In addi- tion to vertical load, the development of a three-dimensional computer program takes into account the horizontal shear stresses induced on the pavement surface because of vehicle acceleration and deceleration. Several studies derived inter- face constitutive models for characterizing the bonding con- dition of a pavement structure through numerical simulation. The BISAR program (44) considers Goodmanâs constitutive law (45) for the surface and base interface, which relates the shear stress at the interface to the difference in horizontal dis- placements. In this model, shear stress is proportional to the difference in horizontal displacements of the bonding layer. Uzan et al. (22) reported that the interface reaction modulus used in the Goodman model was independent of normal stress at the interface. Crispino et al. (46) proposed the use of the Kelvin model to predict the viscousâelastic phenomenon of interlayer reaction under dynamic loading. Romanoschi and Metcalf (47) reported that, in the direct shear test, shear stress and displacement were proportional until the shear stress equaled the shear strength and the inter- face failed. On the basis of this observation, they proposed a constitutive model for the asphalt concrete layer interface using three parameters: (1) the interface reaction modulus, which is the slope of the shear stressâdisplacement curve; (2) the maximum shear strength; and (3) the friction coefficient after failure. Romanoschi and Metcalf concluded that the values of interface reaction modulus and shear strength were not affected by normal stress for an interface with a tack coat; however, the values were affected for an interface without a tack coat. The study showed that the interface bond might fail in fatigue and that the permanent shear displacement had a linear relationship with the number of load repetitions. Mohammad et al. (43) investigated the effects of interface shear bond characteristics of tack coats on pavement response at the interface. With the use of commercial FE software, Abaqus Version 6.9-1, a detailed parametric study was con- ducted to investigate the effect of system parameters, including layer thickness and stiffness on the stressâstrainâdisplacement fields induced in the pavement. A two-dimensional FE model- ing approach incorporating laboratory-measured bond char- acteristics of tack coats was performed to simulate the stress, strain, and displacement responses of composite pavement to wheel loading and to describe the constitutive behavior at the interface. The pavement was modeled primarily as a layered system of linear elastic materials with the capability of treat- ing the surface asphalt layer as a linear viscoelastic material. Anisotropy and temperature effects were incorporated in the FE model. For the delaminating problem in a multilayered pavement, it was found that decreasing loading rate, increas- ing overlay thickness, or both would reduce maximum ISS (43). Maximum ISS was identified at the edge of a tire where both normal and shear stresses were applied to a pavement surface. After the maximum ISS is identified, it can be com- pared with the bond strength obtained through simple direct shear testing, so that an appropriate tack coat material can be selected. Roque et al. (71) examined the potential impact of inter- face debonding on near-surface longitudinal cracking in the wheel path of asphalt pavements. A detailed FE parametric study based on maximum tension and maximum Von Mises stress was conducted to identify mechanisms of near-surface longitudinal cracking in an asphalt pavement with localized interface debonding. Results of this study showed that bend- ing caused by repeated traffic could initiate a crack below the edge of the debonded area, which could propagate to the sur- face because of traffic wander and thermal cycles. Further, it was found that internal tension due to partially restrained dilation could result in a crack that propagated upward through the more aged and less fracture-tolerant mixture near the surface. The authors defined a critical zone (associated with high shear stress coupled with low confinement) at a depth of about 2 in. and extending to 2 in. from the edge of the tire, regardless of asphalt layer thickness.
