8th RILEM International Conference on Mechanisms of Cracking and Debonding in Pavements

The contribution of an asphalt binder to the cracking performance of a mixture is strongly influenced by information relating to the relaxation properties and stiffness. In the 1980s considerable interest existed in the relationship between the relaxation properties and cracking. The relaxation spectra can be captured by the R-value over a limited range of stiffness. This can be related to bottom up fatigue, durability and thermal cracking when the stiffness information is known. Analysis has been presented that demonstrates the interrelationship between parameters used in France, United Kingdom and the USA through linear visco-elastic modeling of a limited region of the master curve. This has been used in several studies recently and has been shown to be useful for judging material used in airfield, high RAP products and other additives in the asphalt market. The use of the linear visco-elastic methods can also lead to a better understanding of methods used to describe fracture—for example the Vialit pendulum test and the direct tension test. While these tests are conducted often in a non-linear region the data can be better explained when interpreting the stiffness results with a frame work developed by linear visco elastic considerations. The paper discusses some aspects of the techniques for development of master curves, interrelationships, linear visco-elastic properties and the relationships between the various parameters proposed for use in different countries.

Aging of asphalt mixtures might play an important role in pavement structural behavior. That should be evaluated using mechanical models. This work aims at identifying the effects of aging on linear viscoelastic and damage behavior of asphalt mixtures, properties that influence the fatigue life of asphalt pavements. That is performed by characterizing a Hot Mix Asphalt (HMA) at four different aging states. The aged materials were obtained by heating loose asphalt mixture in oven at two different temperatures (85 and 135 °C) and aging times (2 and 45 days). Prony series were fitted to complex modulus results. For damage characterization, the Simplified Viscoelastic Continuum Damage (S-VECD) model was adopted. Damage characteristic curves and G R versus N f failure envelopes were obtained from controlled crosshead tension-compression test results at 19 °C. It was concluded that aging produces materials that fail for less evolved damage states. However, depending on pavement conditions and layer geometry, and considering the HMA hardening, aging does not necessarily reduce the fatigue life.

Since 2008, EIFFAGE has developed an aggregate optimization method (maximum contact and minimum space between aggregate particles) combined with the use of special binders (pure, multigrade or polymer-modified bitumen depending on the traffic concerned), to formulate the high-performance asphalt bituminous mix known as GB5®. During that time, numerous roads designed for heavy traffic have been built using this innovative technique, representing more than 1,000,000 tons at the end of 2014, mainly in France and, more recently, in South Africa. This article presents the result of collaboration between IFSTTAR and EIFFAGE aimed at scientifically relating the specific characteristics of the aggregate skeleton of GB5® to its outstanding mechanical properties, via 2D-image analysis and complete mechanical characterization.

This study presents the first attempt to use the Bouc-Wen model to model non-linear hysteresis loops of hot mix asphalt (HMA) samples tested in fatigue in controlled strain mode using a dynamic shear rheometer (DSR). A non-linear least squares optimization was used to identify the Bouc-Wen parameters. In this work, cylindrical samples (12 mm in diameter, 50 mm in height) fabricated from a HMA called dense bitumen macadam (DBM) were tested in fatigue. The outcome of this work confirmed the existence of a good agreement between modelled and experimental hysteresis loops.

Cracking behavior and fatigue performance of hot mix asphalt (HMA) are the important factors, when it comes to evaluate the durability and structural lifetime of bituminous pavement constructions. Both effects are strongly affected by resulting stresses in the bituminous layers due to traffic and/or temperature change, whereat HMA stiffness has huge impact on the magnitude of these stresses and, hence, on the durability. It is obvious that a reliable characterization and prediction of HMA stiffness is crucial. A suitable way for the description of the viscoelastic response of a material is continuum micromechanics, where the mechanical, volumetric and morphologic properties of the constituents of the material are considered to predict the homogenized, “overall” material behavior. Thereby, the material is observed on different, reasonably chosen lengths scales allowing for a description of mechanical effects where they occur. These model assumptions were validated extensively showing good accordance between experimental results obtained from 4 PB-PR tests and model predictions with only the mechanical properties of the constituents (bitumen, aggregate) and the volumetric composition as model input. The results of this investigation suggest that applying the presented technique can lead to a significant reduction of experimental efforts in mix design.

Use of the Discrete Element Method (DEM) to simulate the asymptotic elastic behaviours of bituminous mixtures (low-temperature-high-frequency or high-temperature-low-frequency) is investigated. Aggregates are represented in the numerical specimen by 10,000 spherical particles, with size distribution according to the grading curve of the reference material (the French GB3 bituminous mixture). Aggregates below 1 mm and bitumen are considered in the mastic phase, represented through interaction laws between particles. The preparation procedure of the numerical specimens ensures isotropy and enables detection of the networks of direct (contact) interactions between aggregates and of interactions via the mastic interface. Normal and tangential elastic interaction laws were used for the simulation of the asymptotic behaviours. Several numerical tests were performed to assess the capability of the discrete model to reproduce the asymptotic behaviours of the reference material in terms of complex Young’s modulus and complex Poisson’s ratio.

Behaviour of bituminous mixtures is very complex. Their rheological properties depend on strain amplitude level and temperature. We can observe linear viscoelasticity properties for very small strain amplitudes and non-linearities for larger strain level. Fatigue and rutting (permanent deformation) can appear for a great number of cycles. Moreover brittle or ductile failure can occur. All of these properties highly depend on temperature. The DBN (Di Benedetto-Neifar) model developed last years at LGCB-LTDS laboratory (University of Lyon/ENTPE “Ecole Nationale des TPE”) aims at describing the complex behaviour of bituminous mixtures with a unified formalism. The model is also versatile and may be adapted to take into account or not some specific properties of bituminous materials as failure mechanisms. First, equations of the DBNEPP (Elastic-Perfectly Plastic version of the DBN model) describing the whole three-dimensional DBN model in the linear and viscoplastic domains are recalled. Experiments made at LGCB-LTDS laboratory on bituminous mixtures in the large strain domain are then presented. Finally simulations are proposed based on the DBNEPP model in the viscoelastic and plastic domains.

A mechanism to explain the local contribution of fiber grid reinforcements to reduce tensile crack initiation and its propagation is proposed. In the first part, a local model of quasi-brittle rupture based on the fracture mechanics background is presented. This model is then compared to experimental results and predictions of the literature of V-notched samples in tension. In the second part, the effect of fiber grids as boundary reinforcements of the asphalt concrete layer is then analyzed under constant strain conditions. This study shows an improvement of the local performance against the propagation of cracks crossing the grid. The obtained results also indicate that the contribution of the fiber grid reinforcement depends on the stiffness ratio of the reinforcement and the asphalt and increases for higher quantities of fiber (per unity of length). However, this increase is bounded, which may define a limit quantity of fibers to be employed in practice.

The ability of grid-reinforced asphalt systems to improve pavement performance depends on the selection of the appropriate reinforcement product. In particular, the flexural properties of reinforced double-layered asphalt systems can be used to investigate the potential benefits produced by an interlayer reinforcement. This paper presents the results of monotonic three-point bending tests on grid reinforced double-layered asphalt specimens carried out as part of an interlaboratory test organized by the TG4 of RILEM TC 237-SIB focused on the evaluation of pavement grid effects on both flexural strength and interlayer bonding of asphalt pavements. The tests were carried out on beams of different geometry, at different test temperatures and speeds. Results were analyzed in terms of fracture work and by considering energy-based dimensionless indices, which allowed the direct comparison of apparently different bending test procedures. Results showed that grid reinforcement did not noticeably influence the crack-initiation toughness. However, the fracture energy was clearly influenced by the grid presence and by the grid type. This effect was related to the effect on the post-peak deformation phase and to the different failure mechanisms.

Low temperature cracking of an asphalt mix with specific design is primarily influenced by the loading rate and the temperature. For laboratory testing of hot mix asphalt (HMA), several test methods have been developed and standardised in the last decades to assess the resistance to low-temperature cracking. In order to characterise a mixture efficiently and economically, the test has to be kept within acceptable timeframes. Therefore, low temperature test methods apply fixed parameters, as Uniaxial Tensile Strength Test (UTST) and Relaxation Test (RT) according to EN 12697-46. To study impact of time and temperature on the low temperature behaviour, loading rate at UTST and initial stresses at RT are systematically varied at four temperatures (−20 | −10 | 0 | +10 °C). UTST are carried out at 3 strain rates (1.6 × 10−4–1.6 × 10−2 mm/s). The tensile strength βt and failure strain εfailure are determined and evaluated. Furthermore, RTs are performed at 3 different initial stresses, depending on the strength determined by UTST (0.75 | 0.5 | 0.35* βt). This will reveal how the level of initial stress influences the relaxation time and ratio. The results of this study can be used for modelling and simulation of pavement structures employed on bridges for expansions joints with low deformation rates and high absolute deformations.

Currently, Indirect Tensile (IDT) and Semi Circular Bending (SCB) tests are used for evaluating strength and fracture properties of asphalt mixture at low temperature. While IDT is performed on notchless samples, the SCB peak stress (strength) is obtained on pre-notched specimens which are not representative of initial undamaged pavement conditions. In this paper, the low temperature strength of asphalt mixture is investigated based on IDT and SCB tests, and on 2D finite element simulations. The values of maximum stress at crack tip in SCB simulation are correlated to the experimental IDT strength through non-parametric statistical analysis. The newly proposed relationship can be used to predict the IDT asphalt mixtures strength based SCB fracture tests and, thus, contribute to a better understanding of the low temperature cracking properties in the evolution of asphalt pavements from undamaged to damaged conditions.

It is known that asphalt pavement is subject to longitudinal cracking that begin at the surface and propagate downward. The main mechanism involved is considered to be bending-induced tensile strain away from the tire or shear-induced near-surface tensile strain at the tire edge. It is difficult to specify the location where such cracks occur because vehicles do not follow precisely the same path. Some cracks might form beneath the tire of a vehicle. However, no reliable method has yet been established. The purpose of the present study was to identify the mechanism involved in longitudinal cracking in asphalt mixtures using an improved wheel tracking test. By varying the temperature and loading conditions, it was determined that cracking occurred beneath the tire. The cracks had similar shapes to the longitudinal cracks that occur in road surfaces. The bottom surface of the asphalt showed no evidence of cracking. To determine the reason for this, stress relaxation in an asphalt mixture was investigated using compression tests. As the results, the stress-relaxation performance of an asphalt mixture becomes higher as temperature increases. Moreover the compressive stress was reduced immediately to about half of its maximum value. This suggests that if the compressive stress in the surface layer is released, the residual strain in the binder course layer would act as a tensile strain, which gives rise to the formation of cracks.

Cracking in asphalt layers is one of the most frequent distress mechanisms in flexible pavements due to thermal stresses and traffic loads. Furthermore, the ageing phenomenon along with moisture damage are the key factors for evaluating the asphalt mixture cracking resistance. The effect of ageing and water action on the cracking resistance of an asphalt mixture is studied in this paper. To this end, an asphalt concrete mixture, manufactured with two different bitumens (conventional and polymer modified binders), is subjected to ageing in the laboratory. Then, specimens are prepared with the aged mixture and subjected to the action of water. Unaged, aged and water-aged specimens are subjected to a direct tensile test (Fénix test) at different temperatures to obtain the variation of the fracture energy and toughness under these conditions.

Asphalt mixtures behave in a quasibrittle manner at low temperatures and, consequently, its nominal strength strongly depends on the structure size. Most of the research performed in the past has addressed this phenomenon either on notched or unnotched specimens; however, the evolution of the pavement conditions can lead to the formation of weaker stress singularities than the conventional −1/2 crack-tip singularity. In this paper an experimental study is performed to evaluate the effect of different stress singularities of notchless, V-notch and straight-notch beams of asphalt mixture on the strength size effect. The analysis is based on a generalized weakest link model that combines the energetic scaling of fracture with the strength statistics. The model captures the transition from the statistical to the energetic scaling as the magnitude of the stress singularity increases.

Although cracking resistance is one of the most important properties affecting the durability of asphalt mixtures, this property is often not considered as a requirement in the mixture design stage, and the bitumen content is determined from other properties (voids content, permanent deformation or water sensitivity, among others). The effect of bitumen content on the cracking resistance of an asphalt concrete mixture (type AC) is analysed in this paper by means of different types of tests. Firstly, a cyclic strain sweep test (EBADE test), to obtain the evolution of complex modulus and dissipated energy density with the number of cycles as well as the failure strain; the EBADE test can provided a first approximation to the fatigue law of the mixtures with different binder contents. Secondly, a direct tensile monotonic test (Fénix test) was used to determine strength and fracture energy of the mixture. The results from both tests were consistent between them.

Within the framework of the European Project SUP&R ITN a Ph.D. thesis is carried out to study the durability of asphalt mixtures made with the combination of high rates of reclaimed asphalt pavement (RAP) and warm mix asphalt technologies. For this purpose the complex modulus and fatigue resistance of three different asphalt mixtures, including surfactant modified and foamed warm mix asphalts, combined with RAP has been studied. The extra value is given by the application of an ageing procedure based on the oxidation of compacted materials in laboratory. It follows the recommendations of the RILEM TC-ATB TG5, which distinguishes between short and long term ageing. Fourier Transform InfraRed (FTIR) tests were carried out on the extracted bitumens to quantify the oxidation levels. An increase of the norm and a decrease of the phase angle of $$ \left| {{\text{E}}^{*} } \right| $$ E ∗ at 15 °C 10 Hz with ageing and RAP addition are experienced for all the mixtures. Similarly the slopes of the fatigue laws tend to increase with ageing. A consistent correlation is observed between these evolutions and the evolution of the carbonyl index. In general, the tendency is similar for all procedures, so the use of warm technologies combined with high RAP amounts may need to be considered.

Having an adequate coarse aggregate structure in any asphalt mixture may not be enough to accurately distinguish the cracking performance of an asphalt mixture: indeed, it has been proved that the interstitial volume (IV) can affect asphalt mixtures cracking performance. The IV is defined as the volume with in the coarse aggregate structure filled with finer material, binder and air voids (interstitial components). Several surveying activities on pavement sections made with mixtures designed with Dominant Aggregate Size Range (DASR), which is the coarse aggregate that forms the structural interactive network of aggregate, have been developed in the past. This paper presents an experimental study, recently developed at the University of Florida, aimed at investigating how interstitial components are influenced by different variables, as types of aggregates, aggregate gradation and by binders. The DASR-IC model was used to identify a range of mixtures to be tested by first designing the coarse aggregate structure with adequate inter-locking and then varying the fine portion of the gradation. Laboratory test results from Superpave Indirect Tension Test (IDT) clearly showed that the IV characteristics have a significant effect on asphalt mixture fracture performance.

The resistance of asphalt concrete pavements to thermal cracking is fundamentally related to the asphalt mix stiffness, strength, and contraction properties. These properties continuously change over time with the oxidative aging of the asphalt binder in the mix. Oxidation reaction in long-term mainly results in formation of carbonyl functional group that improves the association among polar components, increasing the asphalt binder stiffness and brittleness. Hence, thermal cracking is generally more prevalent in old pavements. In this paper, a methodology is proposed to refine the mechanistic model for prediction of thermal cracking events over pavement service life in order to account for continuous changes of asphalt mixture critical properties with oxidative aging. Particularly, the changes in asphalt mix relaxation modulus, crack initiation stress, and coefficient of thermal contraction (CTC) were related to the growth in carbonyl groups in the respective binder used in the mix. The approach was examined on few types of asphalt mixes having different air void levels, and with unmodified and polymer-modified binders to be used as a surface course in a specific asphalt pavement. Based on preliminary modeling results, significant differences in thermal cracking performances of mixes were observed when effect of aging was considered.

In this work, detailed study is carried out to develop heterogeneous micromechanical modelling based on multiscale technique to investigate complex response of asphalt mixture. The mechanical behavior of this material is very complex and one must take into account the geometry of the microstructure and the mechanical behavior of the different phases. The properties of such heterogeneous material are highly dependent on the volumetric fraction of bitumen, aggregate structure, air void distributions and interfacial bonding strength between bitumen and aggregates. In this study, the considered bituminous composites are mastic, mortar and asphalt mixes. These composites are treated as biphasic composites, composed of a viscoelastic bituminous binder and an aggregate skeleton with linear elastic properties. In the developed multiscale approach, the homogeneous equivalent properties of biphasic composite are transferred from one scale of observation to the next higher scale of observation. The dynamic modulus of the matrix and elastic properties of aggregates were used as input parameters into micromechanical finite element models qualified by heterogeneous micromechanical models. The complex modulus and phase angle were compared to experimental measurements and analytical values obtained by Generalized Self Consistent Scheme (GSCS).

Low temperature cracking is one of the major causes of failure of asphalt pavements in Kazakhstan where minimum pavement design temperature is between −28 and −46 °C. Thermal stresses in the asphalt pavement were calculated as a function of its relaxation modulus and temperature changes. The parameters of modified CAM model for relaxation modulus of bitumen were related with its penetration index and softening point by formulas based on Van Der Poel’s nomograph. Hot mix asphalt relaxation modulus has been described by Christensen–Bonaquist model. Hot mix asphalt tensile strength as a function was related to the bitumen stiffness by equation based on Heukelom’s data and Molenaar-Li formula. The strength of mix has been also studied by performing tensile tests. The results were analyzed considering the calculated stress increase with temperature drop, and critical temperature was estimated.

The results of asphalt pavement investigation regarding cracks formation on the road in the vicinity of the settlement Gabala and at the ramps and runway of the airport Gabala (Azerbaijan) are presented. Following the research of composition and properties of asphalt mixes, the impact of source materials on the mixes was determined. Mineral component of asphalt concrete is represented mainly by pebbles from mountain streams. Heterogeneity of mineral composition of pebbles causes heterogeneity of their adhesion to bitumen. The properties of original bitumen manufactured in Azerbaijan and of polymer modified bitumen were studied. The effect of viscosity of original bitumen, its susceptibility to cracks formation was studied. High viscosity of modified binder was noted that significantly affects the process of cracks formation in asphalt pavements. Type content before and after aging of original bitumen was studied. Significant decrease in aromatic hydrocarbons in the aged samples and high paraffin content was registered. Parameters that can impair crack resistance of pavement were defined. The properties of asphalt cores taken on the road in specific areas of cracking were studied. The composition of asphalt mixes placed in pavement, debonding in the materials of asphalt pavement, runway and ramps of the airport Gabala were studied. Difference in the nature and in the shape of cracks at these sites was noted. Basing on the research results, the proposals on preventing further cracking on the pavements that will be arranged and the countermeasures to fight cracks in pavements that are in operation were developed.

Field observations and mechanical analyses have shown that cracks accompany rutting in asphalt mixtures under external compressive loads. This study aims to model crack growth in asphalt mixtures under compressive monotonic and repeated loads. With hypothesizing energy equilibrium and viscoelastic Griffith fracture criterion, a damage density characterizing the cracking in a mixture is derived as a function of stress, nonlinear viscofracture strain, asphalt film thickness and bond energy. Crack evolution is modelled by pseudo J-integral Paris’ law. Six types of asphalt mixture were tested by monotonic compressive strength tests at 40 °C. Two were further tested at 5 temperatures and 5 loading rates, respectively. Repeated load test results for the same mixtures were obtained in previous studies. It is found that the damage density shows an S-shape curve under a strain-rate controlled monotonic load and an exponential curve under a stress controlled repeated load. Pseudo J-integral Paris’ law can capture the overstress softening behavior and model the crack growth in mixtures under a monotonic load. The Paris’ law coefficients (A and n) are independent of loading mode (monotonic or repeated), rate or temperature. Thus they are fundamental material properties and can be used to predict crack growth under varying loading and temperature conditions.

This work reports the development of a simple mechanistic fracture model based on a modified Paris’ law using the J-integral. An internal damage parameter, namely the fracture index (FIc), was calculated from this model and was used as an index for ranking the fracture performance of different hot mix asphalt materials tested in fatigue. The relaxation test coefficient (m) and the dissipated pseudo strain energy for the fine aggregates matrix (FAM) volume were used as fundamental parameters for deriving this model. The study revealed there is compatibility between the FIc and phenomenological approach using the number of load cycles at failure. In this work, limestone and granite aggregates were used with two binder grades: 40/60 and 160/220 to prepare four mixtures with two different gradations: gap-graded hot rolled asphalt and continuously graded dense bitumen macadam. The study showed that limestone mixes perform better in fracture than their granite counterparts.

This paper evaluates the effect of ground tyre rubber (GTR) bitumen pellets on the low temperature cracking properties of asphalt mixtures. Low temperature properties were evaluated at 0 °C using monotonic and cyclic three-point bending tests. Flexural strength was determined using beam specimens and fracture toughness was determined using two specimen geometries: single edge notched beam and semi-circular bending. Cyclic bending tests under controlled load conditions were also carried out using notched beams to determine crack propagation parameters. Dense surface course mixtures were produced with a paving grade binder, two different polymer modified binders and two pellets concentrations. It was found that the flexural strength increased when GTR bitumen pellets and polymer modified binders were used. Differences in fracture toughness between the mixtures were, however, relatively small. Only one of the mixtures with a polymer modified binder showed a significant higher fracture toughness value. Furthermore, good correlation was found between the fracture toughness values determined from the two specimen geometries. Cyclic test, on the other hand, showed that the mixtures produced with the bitumen pellets and with the polymer modified binders were able to sustain more load applications than the unmodified mixture. Furthermore, crack propagation parameters were affected by both the type of binder and the concentration of pellets in the mixture.

This paper reports on a series of experiments performed to investigate the effectiveness of different composite bonding configuration on reducing fatigue crack growth propagation. Fatigue crack growth behavior of steel plates with rivet hole in mode-I loading condition, which have been repaired with single and double-side composite patches was investigated. During fatigue tests the crack detection method called ‘beach-marking’ is adopted to follow the crack propagation shape developing during applied fatigue cycles. Furthermore, effect of patch modulus on the crack growth life of repaired specimens and differences between the crack-front shapes obtained for single and double side repaired are investigated. The experimental results demonstrated that the CFRP patches could reduce the crack growth and extend the fatigue life. Moreover, the CFRP plates with high modulus were found to be much more efficient. This study extends the understanding of CFRP repair and provides some interesting results on crack propagation in CFRP repaired elements.

The mechanical response of concrete is largely influenced by the interlocking coarse aggregate, which supplies the cement matrix with a resisting cohesive crack bridging force. Under low and high cycle fatigue loading, the cohesive stresses can influence the crack growth rate and the structural load capacity. A new method used to quantify the cyclic cohesive zone properties and effective crack resistance in three point bending single edge notch plain concrete specimens of different sizes under both low and high cycle fatigue loading is presented here. For validation, three point bend concrete specimens of two different sizes were tested under crack mouth opening displacement controlled low cycle quasi-static loading and force controlled high cycle fatigue using constant, variable, and random amplitude loading sequences. The results indicate that the cohesive stress-dependent cyclic crack resistance can be quantified and used to effectively characterize the high cycle fatigue non-linear crack growth and size effect.

In spite of the fact that many studies have been carried out to improve the understanding of fatigue cracking in bituminous materials, there are still some aspects that should be clarified and which prevent an accurate analysis. During cyclic loading, fatigue damage co-exist with other phenomena (plastic deformations, thixotropy, heating, etc.) that could distort its measurement and make more difficult its precise quantification. In addition, the damage caused by fatigue loads has global and local effects that cannot be separated. Because of this fact, it is not easy to establish a homogeneous failure criterion. Based on these considerations, this study presents a new approach that allows the identification of the different phenomena appeared during cyclic loading and therefore, to establish a homogenous failure criterion. In addition, using this approach is possible to make an accurate quantification of the cracking damage produced in bituminous mixtures, considering global and local effects.

This paper is focused on the understanding of the origin of a bump a few centimeters high which suddenly popped up at the surface of a semi-rigid pavement composed of a bituminous layer placed over a cement concrete layer of 20 cm in thickness. Field investigations have shown that this bump developed right above a cement joint which had already undergone maintenance operations consisting in particular in the digging of trenches and resulting in a decrease of the cement layer thickness at the joint location. It was also noticed that the bump appeared during a warm day and in a wet area, which are both source of dilation. In this context, we suspect the mechanism at the origin of the bump to be triggered by compressive forces at the joint location whose effect is the bending of the cement layer and eventually failure by means of crack propagation in mode I (opening mode) due to excessive tensile stress. This paper aims at providing an accurate description of this field case which is modeled using the Thick Level Set (TLS) approach that proposes a straightforward transition from damage growth to fracture. The numerical implementation of the TLS approach is handled within the X-FEM framework. The results from the computations tend to confirm the suspected mechanism.

This paper presents a modelling approach to predict the fatigue damage of glass fibre grid reinforced asphalt concrete in the fully reverse four point bending test (4PB) configuration. The objective of the presented study is to predict the fatigue damage of a composite material made of asphalt concrete and geogrid, using a damage model proven for asphalt mixes. The non-local damage model of Bodin, implemented in a finite element code, is used. Parameter identification is performed with 4PB test results of non-reinforced asphalt concrete beams. A unique model is developed and three analyses are performed for both non-reinforced asphalt (NRA) and geogrid reinforced asphalt (GRA-100 for a 100 kN/m grid reinforced asphalt and GRA-50 for a 50 kN/m grid reinforced asphalt) in the 4PB fully reverse fatigue test configuration with five-layer beams. A comparison between the predicted and the experimental fatigue damage curves (GRA-100 and GRA-50) indicates that the finite element model is able to predict the fatigue behaviour of geogrid reinforced asphalt mixtures.

This paper presents a method of predicting the resistance to reflective cracking of an overlay system incorporating a Stress Absorbing Membrane Interlayer (SAMI). The method is based on simulation of crack growth, both top-down and bottom up, over a joint or crack in an underlying layer, utilizing a strain-based crack propagation law. It uses an incremental system such that the strain in the region of a crack has to be re-calculated after each increment of crack growth. Its simplified nature means that the method is suited to being incorporated into a spreadsheet. In support of this method of crack prediction, a series of laboratory tests are described. These comprise wheel track tests on rubber-supported beams representing a scaled-down pavement construction—a cracked lower layer, a SAMI (where present) and an overlay. The data from these tests is input into an adapted version of the crack prediction system as the first step in the process of validation. Validation is then taken a step further by making reference to a particular case study that compares a concrete pavement overlaid with and without a SAMI, and shows the pavement with a SAMI to outperform that without. The magnitude of the improvement, though as yet not fully quantified since failure of the SAMI system has not yet occurred, is compatible with the predictions made. Finally the prediction method is applied to selected design cases and the types of pavement that lend themselves to SAMI usage are identified and approximate performance benefits estimated.

The use of overlays in pavement rehabilitation is one of the most common techniques in pavement management systems. It can be used when a pavement structure is still in adequate structural condition and the observable distresses can be corrected, ensuring that pavement condition can be maintained for an additional service period. The main issue related to the performance of this technique is the reflection of existing cracks. In order to delay the propagation process, interlayer systems can be used (e.g. geosynthetics). Although research on this area has been conducted previously, it is still necessary to better understand the mechanical behavior of the hybrid system to improve the efficiency of its performance. The following paper shows the results obtained from the measurements and modeling based on core samples extracted from an experimental test section built using geotextiles as an interlayer system prior to overlaying. The samples were evaluated in the laboratory by means of the Overlay Tester. The fracture mechanics and viscoelasticity properties were determined from Indirect Traction Tests, in order to define the Cohesive Zone Model (damage), and Dynamic Modulus Tests (viscoelastic Prony Series parameters). The results helped describe the mechanical behavior associated with the reduction in the reflective cracking process when geotextile materials are used as interlayer system: (i) increase in fatigue life (up to 260 %, due to the energy dissipation capabilities of the material), and (ii) crack propagation trend (which generally follows geotextile-asphalt layer interface, and depends on the amount of binder content, testing conditions and existing material).

Bituminous mixtures are visco-elastic materials whose mechanical performance is highly influenced by temperature. In addition, during their service life, these type of materials suffer an ageing process which induce an increase in bitumen viscosity and therefore, a variation in the response of the mixture. The changes produced by these phenomena are mainly manifested in terms of stiffness, which could exert a considerable influence on the fatigue cracking resistance of asphalt materials. In this respect, this paper attempts to provide a better understanding of the effects caused by temperature and ageing on the resistance of bituminous mixtures to the propagation of cracking phenomena. For this purpose, the rheological response of the binder has been correlated with the mechanical performance offered by the bituminous mixture.

A viscoelastic-viscoplastic model based on a thermodynamic approach is developed under finite strain in this paper. By introducing a damage evolution, the proposed model is able to reproduce the behavior of Asphalt-Concrete materials until the complete fracture. Moreover, a recoverable part of the degradation is introduced to reproduce the self-healing observed under a sufficiently long rest period. The proposed model is implemented into a Finite Element code and good correlations between the numerical responses and the experiments have been observed.

Cracks in asphalt mixture can self-heal if enough resting time (hours or days) is given. This is a viscosity dependent phenomenon that can be accelerated by increasing the temperature of asphalt mixture. In the present paper, the healing performance of asphalt mixture heated using infrared heating to simulate the natural solar radiation, and induction heating, a new method to increase the temperature of asphalt pavements, were compared in terms of time and healing temperature. Healing was defined as the relationship between the 3-point bending strength of an asphalt beam before and after healing. The results show that both methods reach similar and satisfactory healing rates at around 90 %. However, induction heating is more energy-efficient, since the effect is concentrated on the binder, instead of heating the whole mix. This means much shorter healing times to reach the same healing rate than with infrared radiation. Finally, it was found an optimal radiation energy, from which on, infrared radiation reduces the material healing properties.

Damage in pavements is known to reduce over time when the material is left to rest, this phenomenon is known as healing. It has been shown that healing is an important influence factor in pavement performance. However, an accepted method to assess the healing capability of a pavement is currently not available. Healing of cracks is assumed to be the sum of two processes, cracked surfaces coming into contact (wetting) and strength gain of surfaces in contact (intrinsic healing). The paper describes influencing parameters of these two processes. The healing potential of bitumen is assessed using a novel test method. In this method two pieces of bitumen are brought together and left to heal under controlled conditions. After healing the amount of healing is assessed by testing the specimens using a direct tensile test. From the results it can be seen that normal force has a significant impact on the observed healing, indicating that the process of two surfaces coming into contact (wetting) has a significant impact on healing behavior of the bitumen.

Fatigue failure due to repeated traffic loading is a significant distress for asphalt pavements. Nevertheless, the damage associated to this phenomenon can be partially or entirely recovered during rest periods between loading repetitions, potentially resulting in a longer pavement service life. In this paper, the results from a comprehensive investigation on healing potential of asphalt materials are presented based on cyclic uniaxial tension compression fatigue tests under stress-control mode with and without rest periods. Then a new healing index parameter is proposed and evaluated using a testing procedure with different rest period durations. The analysis of the experimental results reveals that the newly developed healing index effectively asses the mixtures healing properties with an increasing trend for longer rest periods. A limit plateau value is then achieved when the duration of the rest period does not any longer affect the asphalt healing properties. The proposed healing index provides a simple and practical tool for optimal pre-selection of pavement materials, which can be used on a routine basis.

Historically, asphalt mixtures in Minnesota have been produced to be fine-graded in nature. Recently, there are increasingly more coarse-graded mixtures being produced with relatively low asphalt binder content as compared to fine-graded mixtures. The performance of the coarser low asphalt content mixes is not known; therefore, the objective of this study is to determine if the coarse asphalt mixes with lower binder content are prone to a reduced service life and increased performance problems. The present study focused on 13 low asphalt binder content, coarse-gradation mixes from actual field projects constructed mostly in the last 1–5 years. The performance of the pavement sections was quantified using Minnesota Department of Transportation’s (MnDOT) pavement management systems (PMS). Field cracking performance was compared to laboratory tests on core samples procured from the field sections. The field procured core samples were tested using the disk-shaped compact tension (DCT) fracture energy test and permeability measurements. The lab results were also analyzed with each pavement section’s field performance. The results from the study will not only provide answers to this question but also provide estimates of the pavement performances for low binder content mixes. Final outcomes will help revise future asphalt mix specifications. The modifications will ensure that pavements provide the service for as designed time duration and have minimal performance problems that reduce the future rehabilitation and maintenance costs.

Polymer modification of asphalt binders has gained quite popularity in many transportation agencies, primarily due to the superior crack- and rut-resistant performance. However, added cost of polymer modification results in an appreciable increase in the initial cost of an asphalt pavement. There are more economical and sustainable alternatives to polymers, such as the so-called “De-Vulcanized Rubber (DVR)”. Primary advantage of DVR technology is that, it is made from scrap tires and when mixed with asphalt binder, the particles completely dissolve within the binder. Therefore, the final product is a complete fluid, not a suspension. The main objective of this study was to investigate the relative performances of the SBS polymer and DVR in an asphalt binder. The impact of the modifications on performance grade (PG) and fatigue cracking resistance was investigated. The results have shown that more sustainable modification of asphalt binders can be achieved by replacing the entire or some amount of SBS with DVR.

The determination of the allowable percentage of Reclaimed Asphalt Pavement (RAP) that could be used in Hot Mix Asphalt without compromising fatigue cracking is a crucial task. The principal issue is related to the evaluation of the fatigue resistance of the aged binder contained in the RAP and how does it influence the fatigue resistance of the fresh and aged binder blend. In order to solve this issue a new approach is proposed to estimate the allowable RAP binder content in the asphalt mixture, with regard to fatigue resistance, by considering the Superpave parameter G*sin δ. This parameter is determined by using a recently proposed methodology, which avoids the extraction and recovery of the RAP binder to calculate the complex modulus and phase angle master curves of the blend of fresh and RAP binders. This procedure allows evaluating the rheological properties of the bituminous blend by the results of Dynamic Shear Rheometer tests carried out on mortars composed of a selected fine fraction of the RAP, totally passing the 150 μm sieve, mixed with fresh binder at different volumetric fractions of the selected RAP. These properties are estimated by using the Nielsen model specifically modified to take into account the influence of lower frequencies on the stiffening effects of fine particles in bitumen mortars. The allowable content of RAP, determined by this way, is then validated performing time sweep tests on mortars composed of RTFOT+PAV aged and fresh binders, confirming the reliability of the proposed procedure.

Bitumen type and content are two of the variables that most influence on the cracking resistance of an asphalt mixture, since both determine a more or less ductile fracture. Furthermore, this cracking resistance will be different depending on the temperature at which the mixture is exposed and is especially critical at low temperatures. The effect of bitumen type and content on the cracking resistance and fracture energy of an asphalt concrete mixture is analysed in this paper, by means of a new direct tensile test, Fénix test. With the aim of covering a wide range of performances, three different bitumens were used: a conventional 50/70 penetration bitumen, considered as the reference binder; a crumb rubber modified bitumen, which gives flexibility of the mixture; and a low penetration bitumen, which provides more stiffness to the mixture. The test was carried out at different temperatures to evaluate the effect of low temperatures at which the mixture may be critical against cracking phenomenon.

The use of recycled asphalt shingles (RAS) in asphalt concrete leads to questions regarding the mixing of recycled and virgin asphalt binders. These questions concentrate on: the degree of mixing, whether mixing or blending of the asphalt binders occurs, and how the degree of mixing affects performance. Researchers to date have focused efforts on vetting out the degree of blending through various asphalt binder and mixture tests. Research at the University of Illinois used mixture performance tests to evaluate blending. Mixture designs were developed using traditional Superpave methods with 2.5 and 5.0 % RAS. Then, samples with the same virgin aggregate structure and RAS pulp (material remaining post-extraction) and completely mixed asphalt binder were created for high and low temperature evaluation using Hamburg wheel and Disk-Shaped Compact Tension (DCT) tests. In addition, the recently developed Hamburg-DCT Performance Space Diagram was employed to consider mixture property shifts. Research findings demonstrated that up to 20 % asphalt binder replacement met volumetric and performance specifications for both high and low temperature tests. Furthermore, the completely mixed asphalt binder and RAS pulp mixtures performed approximately similar to the traditionally mixed samples at both high and low temperatures.

This paper presents the results of long term investigations that evaluated the impacts of various additives on the resistance of asphalt mixtures to the two types of cracking distresses; fatigue and thermal. Even-though these investigations did not evaluate all additives on the same asphalt mixture and to the same cracking distress, the cracking resistance of asphalt mixtures with and without each type of additive are compared within the same asphalt mixture. Resistance to fatigue and thermal cracking were evaluated using flexural beam fatigue and uniaxial thermal stress and strain test (UTSST), respectively. Since cracking is a problem occurring in the latter part of pavement life, all asphalt mixtures were subjected to long-term oven aging prior to performance testing. The analysis of the data generated from this extensive study indicates that the cracking resistance of AC mixtures is impacted by several factors, including: aggregate absorption, type of WMA additive, asphalt binder modification process, and anti-strip additive.

Many methods and approaches have been developed to evaluate the asphalt performance in the laboratory. The laboratory testing can be conducted on mix design (lab produced) and QA/QC (plant produced) stages. To predict the asphalt field performance, an understanding of differences between the behavior of lab and plant produced mixtures is required. There are many factors in the design and production of recycled mixtures that can potentially impact the performance of the in-place material. The project involves testing of both laboratory and plant produced mixtures containing 20–30 % recycled materials (RAP and RAS) by total weight. The mixtures have different gradations (12.5 and 19 mm NMSA), and contain different virgin binder PG grade (PG 58-28, and PG 52-34). The characteristic damage curves and failure criterion diagrams of mixtures are developed based on fatigue testing following the simplified viscoelastic continuum damage (SVECD) protocols. The results show that the most of lab produced mixtures have better fatigue properties in damage characteristics, but they do not follow a consistent trend in fatigue life.

The advantages of alkali activated slags (AAS) with adequate types and dosages of activators are high strength development, rapid setting, lower permeability, excellent durability, low hydration heat, high early-age strength and high resistance to chemical attack. The hydration products found in AAS are C-S-H with a low Ca/Si ratio related to GGBFS and the components of activators used. However, high autogenous and drying shrinkages of alkali-activated slag concrete are frequently reported. Therefore, researchers have continued their efforts to mitigate shrinkage strains and stresses. Two types of alkali activators and superabsorbent polymers were used to investigate autogenous shrinkage, and porosity characteristics of AAS. To mitigate shrinkage of AAS mortars, superabsorbent polymers (SAP) were applied to create an internal curing effect on the mixtures. As a result, autogenous shrinkage strains and AAS mortars that SAP were applied was significantly decreased, therefore, it is concluded that SAP played an important role on shrinkage mitigation of all mixtures.

During winter season of 2012 numerous transverse cracks developed in high-modulus asphalt concrete (HMAC) base of newly constructed motorway. Pavement cracked both in transverse joint locations and in the area between them. Research which was conducted during investigation of the causes and mechanisms of cracking consisted of: field examination, laboratory testing of specimens cored out of the existing pavement, computational analyses and effect of pavement homogeneity on transverse crack frequency. This paper focuses mainly on impact of homogeneity of asphalt layer on number of transverse cracks. The field investigation of analyzed motorway section includes visual assessment of homogeneity and number of cracks. Laboratory test conducted on specimen cored out of the pavement allowed to assess volumetric properties: binder content, voids content and compaction degree, and mechanical properties: indirect tensile stiffness modulus and strength. Analyses of HMAC layer properties revealed their impact on the number of transverse low-temperature cracks observed in field. It was found that a less effective compaction contributes to increase in the number of cracks. Quality and homogeneity of pavement courses have a considerable effect on mechanical properties of HMAC. As analysis showed, intensity of cracks is well correlated with mechanical properties and homogeneity of asphalt layer.

The use of both cold asphalt and cold recycled asphalt mixtures produced with bitumen emulsion is constantly increasing. The failure properties of these low-energy materials are normally improved by using mineral additions or active fillers. These additions are a key component of the bituminous mortar, which binds the coarse aggregate fraction of the mixture. The objective of the present study is to investigate the evolution of the failure properties bitumen-emulsion mortars obtained by using three different types of additions: Portland cement (active filler), hydrated lime and calcium carbonate (inactive filler). The mortars were prepared with standard sand and a fixed bitumen/addition rate. Cylindrical samples were compacted with a shear gyratory compactor, the selected procedure also allowed evaluating the effect of the different additions and water content on workability. The indirect tensile strengths were measured after 7, 14, 28 and 100 days of curing in two conditions: sealed (i.e. no evaporation) and unsealed, at 50 % relative humidity. Each bituminous mortars was also characterized in terms of low temperature fracture resistance. Results showed the influence of the mineral additions on the emulsion breaking and the different curing conditions allowed highlighting the impact of both evaporation and cement hydration on strength and stiffness evolution.

Over the last decades, reinforcement grids have been used to prolong the service life of pavements. Although there is a broad consensus about their overall efficiency, there are still many open questions about the best alternatives regarding the materials to use, the shapes of the grid, their installation method or their position in the pavement among others. One controversial aspect is the use of Stress Absorbing Membranes Interlayers (SAMI) together with the application of a grid. Many manufacturers claim that the SAMI helps sealing any preexisting cracks in the underlying layer, retarding their propagation to the surface. This paper reports about the performance of a reinforcement grid installed with SAMI in a pavement with artificial cracks. One of the parameters analyzed to evaluate the performance, is the propagation of artificial cracks from the base course to the pavement surface. Other important consideration is the rutting development. To that end, the same cracked pavement, reinforced with a fiber grid and SAMI and without any reinforcement is loaded with identical accelerated trafficking, carried out with the traffic simulator MLS10. Results show that, although the grid with SAMI is able to slow down the formation of cracks in the surface, the density of cracks at the end of the tests is as high as if no grid was used. Moreover, resulting rutting in the reinforced pavement is higher than in the pavement without grid, indicating that the use of SAMI has a counterproductive effect on the rutting performance of the reinforcement.

The effectiveness of SAMI systems in combatting cracking in pavements is frequently questioned and is often found to vary greatly from case to case. This paper will present the results of two different sets of simulative tests: the first set comprises pilot-scale wheel track tests of an overlaid cracked pavement; the other set is of slow strain tests that mimic the effects of a thermal cycle in an overlay over an underlying crack. In both cases the apparent mechanism provided by the SAMI has been investigated, based on the evidence of crack pattern and speed of crack growth, allowing deductions to be made regarding the way the presence of the SAMI changes the stresses and strains present in the system as a whole. These observations also include a limited appraisal of the differences in effectiveness given by different types of SAMI, allowing comment on the role of both the thickness and the viscosity of the SAMI to be made. In general terms it was found that all the SAMI types investigated gave enhanced performance, both against traffic-induced damage and against thermal action. In the case of effectiveness against thermal action, the critical parameter was the viscosity of the material; in the case of resistance to traffic damage, the main influencing factor was fatigue resistance.

Anti-reflective cracking interlayers for the renovation of roads have been extensively used for the last decades. Today different anti-cracking interface systems exist: SAMI’s, non-woven geotextiles, geogrids (plastic, glass or carbon), combigrids and steel reinforcing nettings. The different interlayer systems all have their specific properties with advantages and disadvantages. Moreover the performance of a specific product is not only depending on the product properties and the origin of the cracks but also on the proper installation of the product. Past research revealed that steel is an ideal reinforcement material for an anti-reflective cracking interlayer. Steel nettings as they are known today are very rigid and require a good fixing by nails or a slurry seal. Due to their dimensions, it is also advisable to use an overlayer thickness of at least 5 cm in order to prevent the reflection of the crack/joint through the asphalt surface. In this paper we present a new steel based anti-reflective cracking interlayer with improved installation properties compared to traditional steel nettings, enabling a fast and easy installation.

During the past decades, it has been experienced that reflective cracking is a very complex phenomenon. Not only which one of the possible mechanisms behind the reappearance of cracks in new pavement surfaces (traffic, temperature variations in time or uneven settlements) is dominant, depends on the typical circumstances of a specific project, but also a variety of maintenance solutions often seems to be applicable. Examples are: (combinations of) thick overlays, use of modified asphaltic mixtures, application of stress-relieving systems or the incorporation of reinforcement. At motorway and airfield (maintenance and rehabilitation) projects there usually is time, budget and information. At those large projects it pays tribute to include reflective cracking into the routine design, because a cement treated base or concrete slabs are quite often present. From these specific pavement layers, cracks or joints propagating into and through the asphaltic overlay is the dominant mechanism. Ooms Civiel has developed an analysis method to include the crack driving mechanisms temperature variation and traffic into the routine design of new asphaltic pavements. The temperature influence is analysed by means of ARCDESO®; the traffic loading requires finite elements simulations. This paper explains the design method, which meets the challenge mentioned above. The method has been calibrated with long term field experience (crack mapping data) and is being validated continually.

The final aim of this work is to build a tool dedicated to the calculation of the mechanical fields in pavements incorporating possible vertical cracks in some layers or partial debonding at the interface between layers. The development of this tool is based on a specific layer-wise modeling of the structure so-called M4-5n. In this model the stress fields are approached through polynomial approximations in the vertical direction for each layer. Its construction is based on the Hellinger-Reissner (H-R) variational principle of continuum mechanics. One advantage of the M4-5n is to reduce by one the dimension of the problem. Moreover this model leads to finite values of the generalized interface stresses at the crack lips of the structures studied. This approach is thus particularly adapted to parametric studies and might be considered for analyzing crack growth in layered structures such as pavements. The contribution of the present paper to this model is focused on the computation of its numerical solution by means of the mixed Finite Element Method (FEM). The developed method is based on the maximum of the complementary energy theorem using Lagrangian multipliers to ensure the equilibrium equations. The resulting formulation is equivalent to the H-R variational principle applied to the generalized displacement and stress fields. This approach is applied to a beam structure composed of four elastic homogenous layers resting on Winkler’s springs. Vertical cracks across some layers are introduced. The results obtained are compared with those from an earlier approach using the Finite Difference Method (FDM).

This paper presents a study about the effect of thicknesses and resilient moduli of upside-down pavement layers on the stresses that occur on surface course. For this, a full factorial experiment was developed considering three layers upside-down pavement structures with thicknesses and resilient moduli variations, generating 20,160 experimental conditions, whose elastic behaviors were simulated using Elsym5 software. Factorial statistical analysis was used to evaluate the results and another simulation set was made evaluating variable effect separately. The results indicated that there is increase on tensile stresses on surface course when (i) there is decrease of the surface course thickness or increase of granular layer thickness; (ii) an increase of resilient modulus of surface course or a decrease of resilient modulus of granular base course. Cemented base course thicknesses, resilient modulus and the subgrade modulus have no or slight influence on surface coarse induced stresses.

Fatigue cracking in asphalt pavements results in decreased ride quality, decreased fuel economy, and provides an avenue for intrusion of water. Design of the mixture is predominantly focused on volumetrically balancing the amount of air, asphalt, and aggregate in the system. Such an approach only implicitly considers performance and so the natural progression is to design through direct material property assessment and correlations to field performance. The objectives of this paper are to relate mixture stiffness, fatigue, and pavement system characteristics together for use in performance-based mixture design; identify a Simplified-Viscoelastic Continuum Damage model output parameter which produces the most separation between poorly and satisfactorily performing structures when combined with dynamic modulus and phase angle information; and evaluate the impact of reclaimed asphalt pavement on the performance of the indicator. Results show a relationship between fatigue life of the pavement system and an energy-based index. This approach holds promise because of its reliance on material attributes that can be derived on one testing machine. The constitutive model parameters can be found from the direct tension cyclic fatigue test and can be incorporated into prediction software, further enhancing the appeal of a performance specification.

Fatigue cracking is one of the dominant failure modes of asphalt concrete pavements. There are a number of analysis and design methods that can be used to optimize pavement sections for this kind of distress. Most of these methods incorporate advanced material property predictive models. However, traffic loading, which has been identified as a primary contributing factor in causing fatigue cracking, is characterized relatively simplistically. There is a concern in light of recent advancement in traffic characterization, and tire inflation pressure surveys that existing methods might not be adequate. The objective of this paper is to evaluate and quantify the effects of truck traffic characterization in axle load spectra and high tire inflation pressure levels on predicted fatigue cracking performance. This was achieved by evaluating a number of pavement sections using the mechanics-based fatigue cracking analysis framework. The studied traffic characterization approaches are ESALs, axle load spectra with and without traffic seasonal variations and three levels of tire inflation pressures. It is evident from the result that higher tire inflation pressure and traffic characterization using axle load spectra induce more damage and subsequently early crack initiation time.

Asphalt pavements exhibit strongly temperature-dependent viscoelastic behaviour resulting in response to load varying with both temperature and traffic speed. To design against fatigue cracking for structural pavement design applications the linear-elastic material response is typically assumed to simplify the calculation of the critical strains in the pavement structure layers. To be representative the simplified pavement model needs to be defined in such a way that it accurately reflects the effect of both temperature and traffic speed on the critical strains used to compare with the material’s tolerable strains. Against this background, this paper presents a method to determine an equivalent asphalt modulus (EAM) for the asphalt layer which represents the effect of temperature and loading speed on the critical tensile strains. The EAM is determined from viscoelastic modelling. Two thick asphalt pavement configurations representative of typical French pavement designs are considered. Results show expected trends of the equivalent asphalt modulus increasing with increasing traffic speed and decreasing with increasing temperature. The application of the asphalt material shift factors allowed building pseudo-master curves for the EAM dataset. Finally, the effect of temperature and pavement structure on the relationship between traffic speed and complex modulus frequency is examined. Results support the use of 10 Hz for 70 km/h at intermediate temperatures currently used for pavement designs in France.

Cementation stabilization of unbound granular materials often offers a feasible solution for the strengthening of existing degraded unbound pavements. The primary determination mode of cemented pavement materials is fatigue cracking. Flexural properties including flexural modulus and tensile strain at break are incorporated into the fatigue criteria of cemented materials. However, there are no universally accepted testing protocols available for cemented pavement materials to determine the aforementioned properties in the laboratory. The four-point bending test is chosen in this paper to study the flexural properties of two different cemented pavement materials as it more closely simulates the stress/strain gradients generated in service. The results from this study revealed that, the strain at break could not be determined with sufficient precision as the tensile strain at the bottom of the specimen just prior to the point of fracture increases uncontrollably without further increase in tensile stress. Based on the response of cemented materials to different loading conditions, an equation is tentatively proposed for determination of the modulus for pavement design.

The immediate benefit of producing Warm Mix Asphalt (WMA) mixtures is the reduction in the consumption of energy required to produce the traditional hot mix asphalt (HMA). WMA additives also improve workability and compactability of mixtures to a point where they can be produced at lower temperatures. With the decreased production temperatures comes the benefit of reduced emissions, fumes, dust production and odors, as well as an extended mix haul distance. This study evaluated the long term performance of two different WMA technologies with neat, polymer modified and terminal blend tire rubber asphalt mixtures, designed according to Nevada Department of Transportation (NDOT) and California Department of Transportation (Caltrans) specifications using the Hveem mix design, using one source of aggregates sampled in two different batches from a local quarry. An experimental program was designed to cover the impact of WMA additives on the performance characteristics of the different mixtures using the flexural beam fatigue test and a mechanistic analysis of mixtures in simulated pavements using a 3D-Move analysis on two pavement structure; thin (100 mm) and thick (200 mm). The findings will be presented in the paper below.

The Mechanistic-Empirical approach in rigid pavements design, allows to achieving the superstructure damage by calculating incremental degradation. This study has examined two different pavements typologies: JPCP (Jointed Plain Concrete Pavements) and CRC (Continuously Reinforced Concrete Pavements). Referring to the AASHTO Design Guide the superstructure performances are evaluated in terms of Joint Faulting and Transverse Cracking for JPCP, Punchouts for CRCP, and IRI (International Roughness Index) for both pavements typologies. The performances have been determined using ME-PDG software that is able to evaluate, for JPCP design, the structural fatigue distresses related to Transverse Cracking of PCC slabs and differential deflection related Transverse Joint Faulting. For CRCP, the principal structural distress considered is edge Punchouts. The authors propose the comparison between the different typologies of pavements with the objective of identifying the design solutions more effective with equal materials performance and traffic conditions. These comparisons were carried out by varying the soil class, the climatic conditions and the type of cement concrete. The numerous analyzes performed have enabled to evaluate the influence of different design parameters and then to define useful suggestions that can be used by rigid pavement designers to reduce the occurrence of premature cracking, so as to increase the service life of pavement system.

Weak interlayers within the upper structural layers of road pavements are specifically prohibited in most road-building specifications. However, such layers are extremely common and often lead to premature pavement distress. Heavy Vehicle Simulator (HVS) evaluation indicates that the presence of such layers within the structural layers of a flexible/semi-flexible pavement is far more deleterious than is commonly appreciated. These effects are modelled using examples based an HVS testing and simplistic mechanistic pavement analyses. In particular, weak upper base courses of lightly cemented pavement under thin bituminous surfacing may lead to severe surfacing and upper base failure within a matter of months after opening to traffic, not excluding failure during construction. The causes of these adverse conditions, together with simple detection methods during construction and analyses of their effects on the structural capacity of flexible and semi-rigid (cemented) road pavements, are briefly discussed. Methodologies are available to detect and investigate the existence of these weak layers in cemented pavement layers. Analysis were done on a typical recycled hot mix asphalt (HMA) pavement, cemented base pavement and a granular base pavement, with and without these weak layers and interface conditions. The analyses focused on the strain energy of distortion (SED) as a pavement response parameter indicating the potential for structural damage. SED shows some potential for quantifying relative distortional effects of these weak layers and/or weak interfaces within flexible and semi-flexible pavements. In this paper a very brief overview of above is given.

Enrobé à module élevé Class 2 (EME2) is currently being introduced into Queensland which has a warm sub-tropical climate. While the French design system for flexible pavements directly applies inputs from performance-based mix design, immediate implementation of such a system in Queensland was inhibited by the existing Australian pavement design methodology which does not make such links. In addition, there has been limited application of EME2 in similar climates as in Queensland. The execution of a manufacturing and paving trial, including collection of pavement temperature at various depths over a one-year period, provided vital input into the technology transfer. Flexural modulus temperature-frequency sweep tests, and a series of asphalt fatigue tests at 10, 20 and 30 °C were carried out. The results provide input into the LCPC-developed methodology for calculating asphalt fatigue properties at a given equivalent temperature, and now is utilized and validated for a warm sub-tropical climate.

There is a need to incorporate long term pavement performance evaluation into the current specification through development of performance-related specification (PRS). PRS calls for the prediction of pavement life and development of pay adjustment factors based on expected life differences between as-design and as-built conditions. Different distress models have been developed and are in use by different researchers. Testing and analysis features, simplicity, test equipment and knowledge availability play a significant role in selection of the model for use. Nevertheless, to date, a comparative assessment of how the different analysis approaches affect pavement performance evaluation has not been undertaken. In this study fatigue performance evaluation is done on pavements with different surface mixture condition, varying asphalt and air void content to replicate a range of conditions in the field. Material characterization is done by performing complex modulus and fatigue testing. Pavement response is computed using three layer elastic analysis (LEA) programs. Fatigue life evaluation is done using three different fatigue life prediction models. The results from dynamic modulus testing show that there is an increase in stiffness with a decrease in asphalt and air void content which infers a reduction in tensile strain at the bottom of the asphalt layer. The tensile strain outputs from the different LEA programs were comparable and there is about a 20 % increase in strain for a change in bond condition from full bond to full slip. There is a distinction between predicted fatigue life due to the use different models. The highest variation is observed at low air void content level whereas the impact was lower at optimum and high air void content levels.

The fatigue performance of asphalt materials is a key parameter for asphalt pavement structural design. It is determined in the laboratory as a function of the loading strain magnitude under standard continuous sinusoidal signals. However, real loading signals measured in asphalt pavements differ in shape from sinusoidal signals and many studies showed that the shape of the loading signal has an important effect on the fatigue life of asphalt mixtures. Moreover, to improve road transport efficiency and reduce its environmental impact, new regulations allow longer trucks to circulate on highways and allow current trucks to transport heavier loads which implies a need of evaluating the impact of these changes on pavement fatigue performance. In this paper, a multi-linear fatigue model taking into account the shape of the loading signal is coupled with a viscoelastic model to evaluate the fatigue life of asphalt pavements and determine the effect of loading conditions and trucks configurations.

Reflective cracking is not addressed in the current Federal Aviation Administration (FAA) Advisory Circular for asphalt concrete (AC) overlaid rigid pavements. This paper presents the development of a reflective cracking model using full-scale test data. The prediction model reflected all three-stage process involving crack initiation, propagation, and final failure. Regression models were established to relate varies parameters including overlay temperature, cracking strain, crack propagation rate, number of loading cycles, and crack length. Based on the validation study results, it was shown that a reasonable prediction of the overlay fatigue life can be forecast. It was also demonstrated that lack of considering fracture healing effect during rest periods could result in substantial overestimate of the crack propagation rate and therefore premature overlay failure.

In view of the fact that pavements are multilayer systems, achieving high bonding between layers is a key element to increase service life. Interface debonding is mainly responsible for the slipping failure of pavements that leads to high rehabilitation and maintenance costs. The bonding between asphalt layers is usually evaluated by testing the interlayer shear strength and is affected by several parameters such as test speed, test temperature, normal stress applied and specimen diameter. This paper focuses on the effect of test speed and specimen diameter on the shear strength evaluated through the Leutner equipment, for a typical dense graded asphalt mixture. Leutner tests were carried out on double-layered specimens with a diameter of 100 and 150 mm and with interlayer deformation rates corresponding to nominal test speeds of 1, 2.5, 5, 10, 25 mm/min. The effective interlayer deformation rate was calculated by measuring the deformation through an external transducer in order to perform a reliable data analysis. Results showed a steady increase in the shear strength with the increase in the interlayer deformation rate. Moreover, a clear scale effect was observed at any test speed resulting in higher values for shear strength measured on specimens with diameter of 100 mm.

An effective and durable bond between the various constituent layers is an absolute prerequisite for the durability of a road pavement. To ensure the adhesion between successive pavement layers, cationic bitumen emulsions are the most frequently used type of tack coats. The intrinsic characteristics of tack coats play an important role in the adhesion between layers, but the conditions of application of these coats are equally crucial. In this context, the Belgian Road Research Centre actively participates in a Belgian working group on tack coats initiated by the Walloon federation of road contractors and the Walloon public service. The objective of this joint working group is to carry out a “field” study about adhesion between layers while evaluating the influence of different parameters—such as type and rate of spread of emulsion, spraying equipment, nature and preparation of the binder course, breaking and curing times of the emulsion before overlaying, etc. With a view to this objective, a test site was constructed in the summer of 2014, consisting of four test sections differing in type of tack coat, milling speed and cleaning operation of the binder course. The bond strengths were investigated by direct shear test. This article describes the conditions of application, the measurements made on site (tack coat application rate, binder course texture and cleanliness, etc.) and the results of the interlayer adhesion test performed in the laboratory on specimens taken from the four test sections.

The interface mechanical behavior in asphaltic pavement structures represents a key parameter for the computational design. Several degradation mechanisms are observed in surface layers due to the de-bonding despite the implementation of tack coats or reinforcing systems. The interface modeling becomes more important in the pavement field during the last ten years, even if this concept is widely studied in composites, masonry structures… Furthermore, the experimental research highlights the sensitivity of the imperfect interface behavior in respect to the underlayer roughness. Due to these concerns, parametrical studies are proposed in this paper, by using a macro scale cohesive zone model. This model allows taking into account the damage behavior of the interface, in pure or mixed mode. Damage mechanisms are governed by internal variables. The numerical results based on a finite element method are compared with experimental tensile and shear data. The geometrical periodic roughness (triangular, crenel) and the mixed mode failure are investigated under monotonic loading. A parametric numerical analysis is presented and shows clearly the roughness influence of these structures.

Reinforcement of asphalt pavements has become a valuable constructive method for preventing reflective cracking and prolonging the service life of asphalt pavements. Although the advantage of reinforcement overall seems beyond doubt, there is still insufficient information about the actual effect on the prolongation of the pavement’s life-span. Further, it is not clear how the different grids affect the bonding performance between layers. This paper investigates the performance of different reinforcement grids on the performance of two layered asphalt pavements, mainly considering: (a) Reflexive crack propagation, (b) Shear bonding strength. The paper compares the performance of different non-reinforced and reinforced pavements, constructed in the laboratory and in the field and trafficked with a down-scaled (Model Mobile Load Simulator MMLS3) and a full-scale traffic load simulator (Mobile Load Simulator MLS10) respectively. Loading the pavements with rolling tires (cyclic loading) could show differences in the durability between the different reinforcement types as well as the improvement in the flexural bonding strength when using reinforcements compared to non-reinforced structures. Grid styles with different mesh opening sizes of 12.5 and 25 mm plus different tack coats or tack films were also considered. The investigation of the interlayer bonding strength was done by means of the layer parallel direct shear tester (LPDS) and 150 mm cores. It was found that the bonding properties do depend on the reinforcement type, but that in most cases the interlayer bond is according to the requirements.

Bonded Concrete Overlay of an Asphalt pavement (BCOA) is a rehabilitation technique consisting of 50–175 mm thickness portland cement concrete overlay on an existing flexible, semi-rigid or composite pavement. This technique, that has also been known as thin (minimum 100 mm) or ultrathin whitetopping (thinner than 100 mm) in the past, relies on the composite action of the concrete and asphalt layers acting together with a third phase of the system being the interface between the two materials. For this study, the stiffness and strength/fatigue resistance of this interface have been characterized by means of a series of laboratory tests conducted on asphalt and composite cylindrical specimens under different loading and environmental conditions. Loading conditions were intended to reproduce the rapid traffic pulses and the slow temperature and shrinkage related actions, in both shear and vertical tensile modes. Environment related testing conditions included wet and dry, and temperatures between 4 and 40 °C, which is a range applicable to asphalt bases located under 100–175 mm thick bonded concrete overlays in California. Several conclusions were extracted that provide insight into the mechanical nature of the interface concrete-asphalt as well as the basis for designing a laboratory protocol for interface mechanical characterization.

Reflective cracking is a common issue present in recently rehabilitated pavements, especially when hot mix asphalt overlay is applied over the cracked pavement. It can cause the cracks to propagate to the new asphalt layer in a short period of time, wasting most of the resources spent in the rehabilitation project. Among many existing techniques to control reflective cracking, there is the application of an asphalt interlayer capable of relieving the stresses that reach the overlay, also known as Stress Absorbing Membrane Interlayer (SAMI). This solution can be highly efficient but it is essential to ensure adequate adhesion at the interface between layers, otherwise movement at the interface will eventually cause cracking at the pavement surface. The described rehabilitation solution was applied in a test site at BR-116 highway. As for the SAMI, an asphalt mixture produced with fine granite aggregates and Highly Modified Asphalt (HiMA) binder was used. The hot mix asphalt (HMA) overlay was produced using SBS-modified asphalt binder and the same aggregates source. Tack coat was applied between the asphaltic layers to ensure proper bonding. In this study, field samples were taken and the Leutner Shear Test was used to assess the bond strength at the interlayer-overlay interface. Results have shown that the bond strength was adequate in the test section. Therefore, the efficiency of the solution can be studied without influence of any debonding effects. The values obtained for the shear strength were higher than recommended, so lower rates of tack coat could still be effective.

It is recognised that the adequacy of bond between an asphalt surface layer and the underlying pavement material is fundamental to good pavement performance. This is even more important in airport pavements where shear forces imparted by braking and turning aircraft are high. Various measures of interface shear resistance are available to characterise the bond of asphalt surface layers. Advanced test methods were developed to measure the shear resistance of the interface between asphalt layers. These methods include monotonic testing in direct shear, as well as repeated load testing in inclined shear mode. Both methods have been utilised to compare the interface shear resistance of runways of different field performance, as well as assessing the impact of rain on interface shear resistance achieved during runway asphalt overlay work. Further, the repeated load test has been used to identify a shear susceptible asphalt mixture, with poor field performance, using the monotonic test results to compare cores from two runways.

The pavement lifetime depends in particular on the presence and quality of the tack coat at the interfaces. Various methods are proposed in the draft European project 12697-48:2014. The USIRF, Professional Union of the French Road Industries, by means of two internships (2012–2014) supervised by the ESTP and ENSAM, two French engineering schools, suggested analyzing the relevance of the tests of the draft standard, by realizing a literature review but also a laboratory work. The use of experimental design relieved the tests number. The tests were conducted within the Scientific Campus of Colas on bi-layer cores. The pavement structure was made up of two asphalt layers bonded with a tack coat (French BBSG 0/10 and GB 0/14). Five main factors were taken into account in experimental design: temperature, loading rate, density of the two layers, and type of emulsion, with 2 interactions. The three principal tests methods proposed by the draft standard were applied: Torque Bond Test (TBT), Shear Bond Test (SBT), Tensile Adhesion Test (TAT). The results turn out complementary between them, even sometimes contradictory, with not always the same evolution. Consequently, we have numerous questioning on the relevance of the tests proposed by the draft standard: can we select one and which? Besides, we have noticed feasibility and security problems on the manual Torque bond test described by the standard. In addition, mechanical Torque bond tests were led, with more reliable results.

The rehabilitation of flexible pavements aims the recovery of its characteristics and is usually accomplished either by the application of additional bituminous mixtures (reinforcement), or by removing the layers whose function is compromised and subsequent placement of new bituminous layers, seeking to increase load capacity, serviceability and extend its life cycle. The bonding between the applied layers is one of the key aspects in pavements performance, reflecting the capability of pavements to work as a whole. A comparative study of the behaviour of the interface between the bituminous top layer and an underlying layer by applying different reinforcement elements has been developed. In order to evaluate the interface bonding behaviour for national conditions, namely in what concerns the type of actions, applied materials and construction techniques, a laboratory study was made with destructive tests such us pull-of test and shear type test allowing the proposition of reference values.

Poor bond between two asphalt layers is one of the main causes of premature failure of pavement surface layers, such as slippage cracking, fatigue cracking, shoving or surface layer delamination. These distresses become more serious in pavement areas often adversely affected by horizontal loads due to traffic wheels. There are many factors affecting the quality of bond between asphalt layers including temperature and application rate of tack coat. In an elemental study, the modified Leutner shear test, which allows determining shear strength and shear stiffness modulus at the interface of double-layered cylindrical specimen, was used. Tests were carried out at a range of temperatures between 20 and 60 °C and with different application rates between 0.0 and 0.9 l/m2 (non-residual) using a Cationic Rapid Setting 1 (CRS-1) emulsion. Results of the elemental study are then used to establish regression equations, relating bonding strength, temperature and application rate which are used as part of larger studies to investigate the influence of interface bond on premature failures in pavement structures, such as premature fatigue cracking or rutting. Some interesting results of these studies cast additional light on the behavior of bonding between asphalt layers.

Specialist surfacings such as high-friction and coloured traffic calming surfaces have gained huge popularity since their introduction. However, the reputation of these specialist systems in New Zealand is also plagued by premature failures due to cracking and other related modes. Many of the failure modes are in fact originated from or at least associated with the performance of the underlying pavement substrate. The purpose of this work is to assess test methods where the emphasis is on the performance of the underlying substrate and its interaction with the specialist surfacing systems to ensure best outcome. A number of resin systems, namely epoxy, polyurethane, and methyl methacrylate were tested using ASTM standard test methods such as a pull-off bond test and % elongation measurements. Thermal effects were also investigated by measuring the coefficient of thermal expansion and conducting thermal cycling experiments. Part of the experimental work is to develop tests which can be implemented in the field. Using these results, it is intended that a practical specification can be developed to include a test or tests to better understand the performance of the surfacing and the underlying pavement and thus ensure future surfacing systems meet their life expectancy.

In the context of road network rehabilitation, private companies develop multilayered systems to prevent reflective cracking problems in pavements. These systems are submitted to a mixed mode of cracking propagation due to thermal (mode I) and traffic (modes I and II) effects. In addition, a fatigue mechanism has to be considered under the repetition of these solicitations. In order to evaluate the performances of these anti-reflective cracking systems, several laboratory devices have been developed. In this paper, we focus on two of these tools in order to perform round robin tests on various types of anti-reflective cracking systems. The first test was developed at CEREMA laboratory in Autun (former LRPC). The second was developed at the Ecole de Technologie Supérieure (ETS) in Montréal and is used at Eurovia’s Research Centre in Bordeaux for the purpose of the study. This paper first proposes an analysis of the two test procedures which particularly highlights a main difference in the horizontal and vertical ways of solicitations. Horizontally, a monotonic stretching displacement (mode I) is applied for the first test whereas it is cyclic for the other. Vertically, shear loading (mode II) is applied with the second test device whereas it is flexural (mode I) with the first one. The author’s analysis will focus on seeking correlations between results of the two tests.

The present paper deals with the effect of interlayer bonding conditions between Hot Mix Asphalt (HMA) and Cement Bound Material (CBM) layers on pavement design reinforcement in a case of a semi-flexible pavement. For this purpose a field experiment was conducted in a motorway by the Laboratory of Pavement Engineering of the National Technical University of Athens (NTUA). Non-Destructive Testing (NDT) was applied using a Falling Weight Deflectometer (FWD) and a Ground Penetrating Radar (GPR) system in order to assess the existing pavement condition and take into account reasonable considerations towards the repair and upgrade of the existing motorway. The pavement analysis was conducted considering two approaches: (a) full-bonding and (b) no-bonding conditions between HMA and CBM layers. The overall investigation indicated that the differences between the two approaches range for the different cases of the new road elevation. Moreover elevation increase results in less significant differences between the two approaches. The related findings and results are displayed and discussed thoroughly.

It is now generally recognized that good adhesion between different layers of an asphalt pavement is of great importance for its lifetime. Poor interfacial bond strength leads to a higher probability of cracking, early fatigue failure, delamination, etc. Many factors have an impact on the bond strength between layers. Despite all these factors, it is important that the performance of tack coats can be evaluated independently of the asphalt complex for the purpose of comparison, optimization and selection of these products. Consequently, the development of a reproducible quantitative test method to characterize the bond strength of a bituminous tack coat is crucial. This paper describes the development and validation of a pull-off test on a bituminous tack coat film. In a first step, the test device, the breaking and curing conditions and the pull-off test conditions were optimized to ensure that failure occurs within the film and not at the interface between the film and the contacting plates of the test device. In a second step, a validation of the pull-off device and procedure was conducted by comparing the results of the pull-off tests for different emulsions with direct tensile tests carried out on multi-layer asphalt samples in which the same tack coats were applied. This paper is an overview of BRRC’s work in this field.

Pavement distresses are external indicators of pavement deterioration caused by loading, environmental factors, construction deficiencies, or combinations thereof. The Hot Mix Asphalt (HMA) design for commercial airports in United States of America (USA) is performed as per Federal Aviation Administration (FAA) Advisory Circular AC 150/5370-10G “Standards for Specifying Construction of Airports”, Item P401—Plant Mix Bituminous Pavements (2014). According to FAA study “Operational Life of Airport Pavements”, pavements that meet FAA standards and specification (for design, and construction) generally performed well over their design life. However, recently cases of debonding and slippage have been observed at airports in USA, and other countries. This distress was also observed in the full-scale accelerated pavement tests during flexible pavement construction cycles 1 and 3 at the FAA’s National Airport Pavement Test Facility (NAPTF) and was attributed to tack coat and environmental factors. Asphalt strain gages were installed at the top of milled HMA surface and bottom of HMA overlay at Runway 4R-22L at Newark Liberty International airport (EWR) to study the debonding and slippage distress. Sensor responses were used to detect the onset of debonding and subsequent slippage in the HMA overlay. This paper summarizes the results from two case studies—HMA behavior under full-scale accelerated pavement tests at NAPTF, and the pavement instrumentation project at EWR airport.

Aircraft tyre inflation pressures and wheel loads have increased over time and this trend is not expected to abate in the future. With increased tyre inflation pressures has come increased shear forces induced by braking aircraft. An increase in reports of interface debonding and delamination failures of asphalt surfaces on runways has followed in Australia, Japan and South Africa. Octahedral shear and normal stresses were calculated for extreme braking aircraft of various tyre inflation pressures and wheel loads. The historical increase in tyre inflation pressure of around 50 % was shown to have resulted in a moderate 10 % increase in the stress-strength ratio associated with the surface layer interface. The factor of safety associated with interface shear strength was shown to be negligible. New interface construction techniques or tack coat materials with increase adhesion are required to reduce the risk of further shear-related delamination of asphalt surfaces in high stress applications such as runways.

Research on the potential link between debonding and cracking has primarily focused on changes in location and magnitude of maximum tensile stress/strain under the load center for various interface bonding conditions. Prior modeling efforts have evaluated smeared bonding conditions, which represented the entire interface as either bonded, partially bonded or debonded. However, field evidence indicates debonding is a local phenomenon that starts at a critical location and gradually progresses through a portion of the interface. A parametric study identified a critical zone of high shear stress coupled with low confinement where the onset of debonding is likely. This critical zone is located around the mid-depth of the asphalt layer and extends for nearly 5 cm from the edge of the tire. Localized debonding should be introduced at the identified critical zone to better evaluate stress redistribution resulting from interface failure. Localized debonding was shown to potentially lead to stress states of shear-induced tension in the portion of the interface near the tire edge, which can explain the initiation of near-surface longitudinal cracking.

Top-down cracking is known to be initiated near the surface of the asphalt layers. The aim of this paper is twofold: (i) to show experimental evidence of the viscoelastic behavior of interface in asphalt pavements under some temperature conditions, (ii) to show that the integration of such a behavior could provide an explanation of the mechanism involved in the initiation of bottom-up and top-down cracking. This paper documents the methodology used to investigate the behavior of the upper interface from experimental tests carried out at the IFSTTAR’s track facility. The mechanical behavior of the experimental pavement is evaluated using three different models: (1) elastic model, (2) Huet-Sayegh viscoelastic model to account for the behavior of asphalt layers, and (3) same as 2, but with additional very thin viscoelastic layers to represent interfaces between the asphalt layers. Computations are performed for two load configurations using software ViscoRoute 2.0© to evaluate the stresses and strains at different depths of the pavement for the 3 models. The comparison between the test track results and the models clearly show that model 3 is that yielding the best fit. This model shows that significant tensile stresses and strains occur near the surface and at the interface between two asphalt layers. The transposition of the viscoelastic behavior of interfaces to real traffic conditions could explain top-down cracking as one of the modes of failure of asphalt pavements.

The mechanical properties of surface courses inter-layers are a key parameter for the design of repaired or reinforced multi-layered asphalt pavement structures. The insufficiency of the interlayer bonding leads to significant decrease of pavement durability and recurrent distresses (slippage failures, top-down cracking). Hence, the main objective of this paper is to identify the input parameters for interlayer modeling, using DIC techniques. On the first hand, direct tensile test (DDT) and double notched shear test (DNST) are performed on two layered bituminous concrete specimens. The full field analysis allows to characterize an interphase thickness, the constitutive laws of both interphase and interface without thickness and the critical energy release rate in mode I and II. On the other hand, the contact surface morphology is analyzed by a fringe projection technique in order to determine the roughness parameters. A strong degree of similarity seems to be obtained between the interphase thickness and the roughness.

The mechanism of moisture-induced debonding in asphalt mixtures is described using data obtained from pull-off and direct tensile tests of aggregate-mastic bonds and miniature 20 mm diameter cored asphalt mixture specimens, respectively. Moisture conditioning using deionised water was conducted at multiple temperatures for periods of up to 40 days. Analyses of the results led to the development of a novel framework for describing moisture-induced debonding in asphalt mixtures in moisture-susceptible aggregates. Damage models describing debonding as cohesive damage in the bulk asphalt and/or adhesive failure at the aggregate-mastic interface resulting from monotonic tensile loading were derived based on load versus conditioning time. Debonding at the aggregate-bitumen interface was found to be the main mechanism of moisture-damage in the moisture-susceptible acidic aggregates that were tested. The trends in debonding with conditioning time was found to be neither test-mode (pull-off versus direct tension) nor conditioning temperature (20 °C versus 60 °C) dependent which suggest the presence of moisture at the aggregate-bitumen interface is a key factor influencing the mechanism of moisture-induced debonding in asphalt mixtures.

Concrete overlays constitute one of the most frequent measures to repair and strengthen concrete bridge decks, concrete pavements, and industrial concrete floors. Good, secure, and durable bond between overlay and substrate provides the prerequisite for monolithic action and improves the durability. 30 years of Swedish research on bonded overlays thought beam tests, slab tests, and field measurements shows that a good bond can be obtained if some important demands are fulfilled. The Swedish research also includes long-term studies on repaired concrete bridge decks. These studies show that the bond is of the same magnitude ten years after concrete repair independent of structural system of the bridge, climate zone, traffic volume, or use of de-icing salt. Laboratory studies show that good bond is also able to resist fatigue loading. This paper summarizes these research activities and provides heavy arguments for answering the question in the heading negatively. Debonding is not unavoidable.

Externally bonded carbon Fiber Reinforced Polymers (FRPs) are now commonly used for the strengthening and repair of Reinforced Concrete structures. However, if the effectiveness of this technique has been widely demonstrated, the durability of the adhesive bond at the concrete/composite interface is still a matter of investigation and remains a critical issue to be addressed in order to assess the long-term performance of FRP strengthening methods. The proposed paper aims at presenting the first results of an ongoing investigation on the time evolution of the concrete/composite adhesive bond strength. Such an evolution was studied by performing double lap shear tests, while changes in the mechanical properties of the polymer adhesive were investigated by means of tensile tests. Preliminary results show that shear tests are able to reveal evolutions of both the bond strength and the failure mode of concrete/composite assemblies subjected to various accelerated ageing conditions. The weakest part of the assembly, initially assigned to the concrete substrate (cohesive failure in concrete), is progressively transferred to the polymer joint (adhesive failure at the bonded interface).

In order to evaluate water sensibility on the interface between layers of composite pavements, a four-point bending (4PB) test on bilayer structure in a water bath is proposed. Using the virtual crack closure technique, the individual strain energy release rates are calculated with a specific model. The debonding mode I should be recognized as the main failure mode. For bilayer specimens made of a cement concrete overlay on bituminous material, the specific test has shown a competition between the different failure mechanisms. A very good bond resistance between layers compared to the fracture tension resistance of the cement concrete layer is observed. In this work, first results of the effect of water on the behavior of such a material interface are presented. An aquarium is built in order to submerge under water the bi-layer specimen. The final fracture length of the specimens, curves of force-displacements and first digital image correlation results show the influence of water immersion on the debonding failure mode.

After certain period of time, the degradation of concrete structures is unavoidable. For large concrete areas, thin bonded cement-based overlay is a suitable rehabilitation technique. Previous research demonstrated that durability of such applications is always a problem and one of its main reason is debonding at interface. Laboratory and field researches show that fiber reinforcement in repair material can be a solution for controlling crack opening and also to enhance the durability of thin bonded cement-based overlays. In other respect, previous researches also show that by addition of rubber aggregates obtained from grinding of used tyres is also a suitable solution for improving strain capacity of cement based materials. This present research mainly focuses on synergetic effect of using rubber aggregates and fiber reinforcement in mortar as a composite for the repair work. For this study four mortar compositions to be used as overlay material were prepared: one control mortar, second with fibers at dosage of 30 kg/m3, third containing rubber aggregates replacing 30 % sand by equivalent volume and fourth one containing fibers and rubber aggregates. Direct tension tests were conducted in order to obtain the tensile strength, strain capacity, residual post peak behaviour of the repair material and bond tensile strength of the repair-substrate interface. Results showed that although by incorporating rubber aggregates in mortar reduce compressive strength and modulus of elasticity but improvement in straining capacity is observed. Moreover, fiber reinforcement in repair significantly improves residual post peak tensile strength.

Recently, bonded concrete overlay has been considered as a possible alternative material for use in pavement rehabilitation as its material properties are similar to those of the existing concrete pavements. In a bonded concrete overlaid pavement, both the overlay layer and the existing pavement should perform as one monolithic pavement and bonding between the overlay layer and the existing pavement is essential. Therefore, bonded concrete overlays are carefully constructed to achieve and maintain the bond between the overlay layer and the existing concrete pavement using milling, blasting method and bonding agent. Generally, typical debonding failure modes of bonded concrete overlay observed in the loss of bonding, failure by normal tensile stresses, and horizontal shear stresses. Previous study indicated that normal tensile stress at the interface is main cause of debonding in bonded concrete overlay. However, the bonded concrete overlay at the traffic congested section and toll booth of expressway experience due to stopping and starting of vehicle. This may lead to debonding by horizontal shear stress at the interface of bonded concrete overlay. This study aimed to investigate the debonding mechanism of bonded concrete overlay due to horizontal shear stress. A set of numerical analysis was performed in order to evaluate the horizontal shear stress at the interface of bonded concrete overlay due to vehicle stopping and starting of vehicle in the traffic congested section and toll booth of expressway.

Specific products are designed for truck traffic lanes on orthotropic steel bridge decks. These products are multilayered systems that must meet requirements for two functions: waterproofing and wearing course. This paper focuses on a specific test described in the standard NF P 98 286. This durability test reproduces mechanical behavior of the steel deck in the vicinity of a stiffener. A typical test specimen is comprised of a 10–14 mm thickness steel plate covered by a multilayered waterproofing system including asphalt layer from 50 to 80 mm thickness. After surface preparation of the steel plate in a blowing chamber, the test sample is manufactured in laboratory. A waterproofing layer is applied on the steel plate. An asphalt mix is compacted over substrate elements with a wheel track compactor. After sawing, strain sensors are glued on the asphalt mix sides of the specimen near the stiffener. The specimen is subjected to fatigue flexural load at control temperature that produces two kinds of distress: delamination and cracking. The authors present ten years of feedback on the use of this test at Eurovia Research Center in Bordeaux. This paper focuses on the study of the slab calibration and proposes as a conclusion to extend the standard specifications to any slab thickness.

Latex modified concrete (LMC) has many advantages against an ordinary Portland concrete. Hundreds of projects have been completed using LMC for overlay in Korea. A sound bond is an essential requirements of bridge deck pavements, and the bond property of a bonded overlay to its substrate concrete during the life-time is one of the most important performance requirements. This paper compare and evaluate the effect of the hydrodemolition and cutting diameter of the pull-off test into the bond strength of latex-modified concrete overlay by a series of experimental variables. The concrete surface preparation methods included hydrodemolition, cold milling and jack hammer. The core cutting diameters of pull-off test included four levels: 50, 75, 100 and 150 mm.

The shear strength of concrete/rock interface must be clearly assessed to design the rock foundations. However, the concrete/rock interface has different failure modes which depend on the mechanical characteristics of the materials and the morphology of the rock surface. Consequently, to study thoroughly the shear behavior of concrete/rock interface and to define the cracking mechanisms and debonding propagation along the interface, different instrumentation devices were applied during a large scale direct shear test. A large sample formed by pouring concrete over a large rock surface (shear surface of 1.5 m2), was instrumented using the following techniques: strain gauges, acoustic emission and distributed optical fiber sensor. The different observations made during the test were combined to determine the successive stages of interface failure. The strain gauges, which measure local strains within the materials, showed their ability to detect the crack propagation in the materials and the debonding at the interface. The acoustic emission technique was able to detect the damages in the materials and at their interface. The optical fiber sensor measured strains evolution of the concrete near the interface showing the crack propagation during the test and the progressive debonding at the interface.

An acoustic emission (AE) approach to evaluate low-temperature cracking susceptibility of asphalt binders and asphalt concrete mixtures is presented. In the approach thin films of asphalt binders were bonded to granite substrates and exposed to temperatures ranging from 15 °C to −50 °C. Differential thermal contraction between granite substrates and asphalt binders induces progressive higher thermal stresses in the binders resulting in the thermal crack formation, which is accompanied by a release of elastic energy in the form of transient stress waves (AE). These AE-based Tcr predictions showed excellent correlations with predictions based upon AASHTO TP1 and with AASHTO MP1A protocols. Similar results were also obtained when asphalt concrete mixtures samples were exposed to temperatures ranging from 15 °C to −50 °C. The AE-based approach for low-temperature characterization of binders and asphalt concrete mixtures is faster and has a lower coefficient of variance than the traditionally used methods based upon the binder rheological properties.

This paper presents a new approach for the continuous health monitoring of asphalt concrete pavements using self-powered wireless sensors. Numerical and experimental studies were carried out to evaluate the damage detection performance of the proposed self-sustained sensing system. A three-dimensional finite element analysis was performed to obtain the pavement responses under moving tire loading. Damage was introduced as bottom-up fatigue cracks at the bottom of the asphalt layer. Thereafter, features extracted from the simulated dynamic strain data for a number of sensing nodes were used to detect damage. Laboratory tests were carried out on an asphalt concrete specimen in three point bending mode to verify the sensor response. The results indicate that the proposed method is effective in detecting different damage states including crack propagation.

A pavement structure, which contains artificial interface defects, has been built on the full scale accelerated pavement testing facility of IFSTTAR in Nantes. This test section is made of two bituminous layers (8 cm thick base layer, and 6 cm thick wearing course), over a granular subbase. Several types of defects have been included at the interface between the two asphalt layers. Rectangular debonded areas of different size (of longitudinal or transversal direction) have been created artificially, using different techniques (sand, textile, absence of tack coat). The construction has been carried out by a road construction company, using standard road works equipment. Then, the pavement fatigue testing facility has been used to apply traffic loading on this pavement, to study the effect of such sliding interfaces on the mechanical behaviour of the pavement, and the evolution of the defects with traffic. The pavement structure has been monitored with several non-destructive methods. The paper presents the survey of the test section with an Ultrasonic Pulse Echo method. This Non Destructive Technique used high frequency wave propagation. It has been able to detect and locate interface defects. The method can also be used to evaluate material properties.

The existence of a fundamental energy threshold for meso-scale crack initiation is investigated using micromechanical modeling techniques. X-ray Computed Tomography (CT) is used to acquire the internal structure of an asphalt concrete mixture while Digital Image Processing (DIP) techniques is used to segment and analyze the different phases present in the mixture. Finite Element (FE) modeling is used to simulate a tensile loading condition to establish a critical micromechanical criterion for meso-scale crack initiation. The meso-scale asphalt concrete mixture is subjected to different loading rates to obtain the global strain energy density at the instance when the critical micromechanical crack-initiation criterion threshold is attained at different deformation rates. The result from the study shows that there exists a fundamental global strain energy density threshold that is invariant of the rate of loading at the instance of meso-scale crack initiation. The result of this study also shows the potential of the use of X-Ray computed tomography in understanding the cracking phenomenon in asphalt mixtures.

The microstructure of an asphalt mix is influenced by the aggregate content, orientation and contacts. In several previous studies, the asphalt concrete is treated as homogeneous material and its microstructures are ignored. However, it has been reported that the deformation and strength of asphalt concrete are not only influenced by volume fraction of its components but also affected by the spatial distribution of its microstructures. Thus, to investigate the mechanical behaviors of asphalt concrete, it is imperative to notice more accurately a micromechanical model containing information of components and microstructures. The objective of this study is to investigate the influence of the microstructure characteristics on the mechanical behaviour of asphalt mixes. The details of the microstructure are obtained by X-ray computed tomography (CT). A comparison with results issues from digital models of random aggregate generation will be considered. The dynamic modulus of the matrix and elastic properties of aggregates were introduced as input parameters into numerical models qualified by heterogeneous microstructure. Finally, the influence of the microstructure on the mechanical response of material at local level is also highlighted.

This paper deals with the characterization of mode I fracture parameters using a kinematic approach integrating the experimental displacement measured by optical techniques. Tests are carried out using a wedge splitting sample made in asphalt concrete. The WST-cube specimen is a bi-layered asphalt concrete AC12 with Carbon Fiber grid at the interface. The tested material was provided in the RILEM-SIB. During the test the displacement field evolution close to the crack tip was recorded by using the optical techniques. An adjustment procedure was also used to improve the displacement fields and avoid experimental noise. Based on the experimental optical measurements, the energy release rate was performed by means the Crack Relative Displacement Factor and Stress In-tensity Factor. The Mark Tracking method is employed in order to measure the Crack Opening Displacement. This approach allows to consider the assessment of fracture parameters for the real structures.

This study is an experimental investigation of the mechanical response of asphalt mixtures using a full-field measurement technique: the grid method. This measuring technique is able to detect small strain amplitudes (up to hundreds of microstrains) with a spatial resolution which is often compatible with the high strain gradients which generally occur in thin binder layers between aggregates. Compression tests were carried out on cylindrical specimens. Displacement and strain fields were measured during these tests. The main conclusion is that the mixtures exhibit highly concentrated strain peaks in the binder, while the aggregates remain almost undeformed. In addition the analysis of the local temporal response of the binder shows that it exhibits a nonlinear local response.

The transportation industry has become increasingly focused on thermal cracking in asphalt concrete. As such, fracture tests such as a single-edge notched beam test have become the norm. The most important output from this type of test is the crack mouth opening displacement (CMOD) based fracture energy, but it only takes into account the applied load and displacement at a discrete location on the test specimen. Full field displacement evaluation such as digital image correlation (DIC) and advanced material models used in discrete element modeling (DEM) provide further avenues to characterize the complex fracture process associated with asphalt mixtures. An ongoing research study couples DEM with DIC displacement fields to study the Mode I response of asphalt concrete in the beam fracture test. Previous research studies defined local properties to match the global load-CMOD behavior using elastic contact bonds. However, improved property calibration is possible through the optimized use of DIC displacement fields and viscoelastic DEM traction-separation laws. Research findings in the proposed work demonstrate the applicability of the DIC-DEM optimization scheme.

This paper presents the two image processing techniques that have been developed within the scope of the TRIMM project to automatically detect pavement cracking from images. The first technique is a heuristic approach (HA) which was originally developed to process pavement images from the French imaging device. The Minimal Path Selection (MPS) method is a new technique which provides the accurate segmentation of the crack pattern along with the estimation of the crack width. HA has been assessed against the field data collection over UK roads provided by Yotta and TRL using the Tempest 2 device. MPS has been assessed against Aigle-RN pavement images over the French network. The benchmarking of five automatic segmentation techniques has been provided at both the pixel and the grid levels. Among others, MPS reached the best performance at the pixel level while it is matched to the FFA method at the grid level.

The Road Engineering/Sealing Components Laboratory at Empa has been working during the last four years in an innovative experimental approach aiming at developing a new kind of healable asphalt road via induction heating. The last step of this research focused on the analysis of 1.8 m long test slabs damaged by the Model Mobile Load Simulator MMLS3 in order to prove the feasibility of the healing concept at larger scale. It is known that visible cracks could not be completely healed by this technique and therefore, the recovering of the mechanical performance was not significant. In this context, it seems clear that the healing treatment must be applied not later than the initiation of microcracks to avoid their propagation. For this reason, a digital image correlation system has been used successfully to monitor the damage level of a number of test slabs during the loading phase. This system employs two digital cameras which record the deformation process by means of high resolution images. Then, these images are analysed with correlation algorithms in order to calculate the 3D displacements and strain components for every object point. This optical method allowed to visualize the accumulated damage as well as to select the right moment for initiating the healing process. In addition, this method was useful to confirm that the strength is partially recovered after the healing process resulting in an increase of life of the road. Finally, it was shown that the healing procedure by induction heating can be a feasible alternative for maintenance purposes by acting before irreversible damage of the pavement occurs.

In civil engineering, the survey of reinforced concrete structures is performed mostly from visual indicators by specialized operators. This paper focuses on the internal swelling of concrete, which is revealed by the cracking at the material surface. The manual survey method consists in estimating the average elongation of the material from the cumulative measurement of cracks thicknesses in 4 directions. To overcome some drawbacks of the manual method, this paper proposes to develop a semi-automated measurement of the cracking index from digital images and image processing tools. This allows us to obtain a more reliable crack index and refined crack characterization. This method has been first tested in laboratory and on a bridge near Nantes in 2012. Finally, the authors have been involved in 2015 on a long-term survey of the stacks of a larger bridge in Nantes. The results of the first-year survey will be reported by the time of the congress in 2016.

Three or four point bending tests are able to initiate controlled mixed mode cracking between two layers of different materials (Hun in Influence de l’eau sur le décollement d’une chaussée urbaine, 2012). The monitoring of such phenomena is a difficult challenge as if global force-displacement measurements are able to detect from the apparition of a non linearity the existence of a defect, they cannot determine its nature or location. More recent Digital Image Correlation (DIC) techniques can help however their application is limited because they require an image of a speckled side of the beam. Due to the form factor of the plate or beam, this image cannot cover all the side of the test except if one owns a ultra-high definition camera and lens. On the contrary the Virtual Image Correlation (VIC) method (François et al. in Eur Phys J App Phys 56:1–10, 2011) allows the measurement of the plate shape without any speckle thus does not require ultra high definition imaging. This global image is correlated to a virtual image whose parameterized shape is simply issued from the beam or plate theory. The delamination or the cracking is respectively introduced in the computation from simple reduction of the beam stiffness or a plastic joint. It will be shown that both their magnitude and location can be measured by using the VIC.

The paper gives an overview of the ongoing experiment to survey debonding areas within pavement structure during accelerated pavement testing on the Ifsttar’s fatigue carrousel. Several defects have been embedded during the construction phase and have been probed by different NDT techniques. Among them, the paper focuses on radar NDT&E techniques which have been used to detect and locate the artificial defects at the early stage of the experiment and then to perform the survey at different loading cycles. Besides, the test-site has been used to test different GPR materials, including the 3D GPR technology. The experiment has also motivated some related studies in the field of GPR data modelling and data processing, which are summarized in the paper. The experiment is expected to be pursued beyond 300 loading kcycles to reach larger pavement degradations.

There is a need to assess the realistic tyre contact pressure created by a tyre in contact with a pavement. A traditional representation of tyre-pavement contact pressure is a circular uniform contact patch. This is overly simplistic, the contact pressure is a function of tyre type, axle loading and tyre inflation pressure (TiP). The research carried out considered dual tyre and wide-based tyre assemblies across a range of axle loading and TiP. These contact pressures were incorporated into a finite element package (CAPA-3D) and modelled on a thin pavement structure. The strains from this modelling were sorted to produce key shear strains associated with the key mechanisms of near surface pavement distress. The main distress mechanisms being top down cracking and asphalt cracking/rutting. This gave a method to fairly compare the dual and wide-based tyre assemblies with the same axle loading and TiP. The analysis gave interesting results for the different distress mechanisms of the pavement. The wide-based tyre gives consistently higher shear strains for all the areas of distress investigated. There is great variation in shear strains due to the different combinations of axle loading and TiP. It is clear that the wide-based tyre is a more damaging tyre for all combinations of TiP and axle loading. It is also apparent that how these factors interact has a great influence on the damaging potential of a tyre.

Usual Falling or Heavy Weight Deflectometer (F/HWD) backcalculation methods assume that pavement layers are fully bonded. In order to assess the ability of HWD testing to detect debonding of the interfaces between pavement layers, the French civil Aviation technical center (STAC) performed an HWD campaign on the circular Accelerated Pavement Testing (APT) facility of IFSTTAR in Nantes, on an experimental pavement with artificially created local sliding interface between bituminous surface and base layers. Both the sound and the defective tested pavement sections present a common structure in terms of subgrade, materials and layer thicknesses. HWD tests have been performed along a large circular arc covering the two sections. A fine spatial measure discretization (10 cm) was used. The experimentation shows that the HWD is a valuable tool for the detection of extended interface defects detection, the central deflection being significantly affected by the interface quality, the outer deflections being less sensitive to this parameter. This paper presents raw deflection results, allowing to precisely locate the underlying defects in the test section. Then a numerical study is detailed, which aims at assessing the effect of the bonding conditions over deflections. The theoretical analysis of this HWD survey is performed in two steps: firstly, on the basis of the finite elements method (FEM) dynamical modeling of the HWD test developed by the STAC, a backcalculation of material moduli is performed using the deflections measured in the sound area. Secondly the backcalculated moduli are introduced in a forward calculation, taking into account the debonded interface. The numerical results are compared with experimental deflections measured over the defects. They are consistent with in situ results.