Various trends in rate-dependent stress–strain behaviour caused by the viscous properties of a wide variety of unbound geomaterials that were observed under drained triaxial compression, plane strain compression, direct shear, and one-dimensional compression (i.e., oedometer testing) are summarized referring to those of bound geomaterials and others. The following findings are shown. As the interparticle contact points become more stable and the stability of the particles against rotation increases, the viscous properties become closer to the traditional type, namely, Isotach, whose stress–strain behaviour is determined by the instantaneous irreversible strain rate and whose strength during monotonic loading (ML) at a constant strain rate increases with an increase in the strain rate. A couple of non-traditional (i.e., non-Isotach) types of properties were found with granular materials (GMs) that exhibit noticeable creep deformation, stress relaxation, and changes in stress upon changes in the strain rate. Unbound poorly graded sub-angular to angular GMs exhibit the so-called TESRA type of properties (also called the viscous evanescent type), for which the stress–strain behaviour during ML at a constant strain rate is essentially independent of the strain rate. Unbound poorly graded sub-round to round GMs display the so-called Positive and Negative (P&N) type of properties, for which the strength during ML at a constant strain rate decreases with an increase in the strain rate. The combined type displays intermediate behaviour combining the Isotach and TESRA types of properties. These types of viscous properties are quantitatively characterized by a couple of functions and parameters, and they were incorporated into the non-linear three-component (NTC) model. All the observed trends in rate-dependent stress–strain behaviour are well simulated by the NTC model.
A high-performance composite material consisting of reinforced compacted cement concrete including Reclaimed Asphalt Pavement (RAP) has been investigated. This composite material is used as a base layer in combination with an asphalt overlay for heavy traffic roads. Steel fibers are used to minimize crack opening. RAP is added to preserve raw material and to comply with sustainable development, especially in the case of in-situ treatments. Composites with different RAP and fiber content and cement content were studied both in laboratory and on site. The laboratory study is divided into two parts. First, classical standard tests for road concrete were carried out: compressive strength, compressive modulus and tensile splitting strength, with a view to the design of pavement structures to be used on site. Secondly, more comprehensive rheological tests, such as complex modulus and fatigue tests, were performed to obtain a better knowledge of material behavior. Then, on-site behavior was studied from two experimental sites. The presence of bitumen in cement-treated composites induces an increase of viscous properties, observed mainly on the phase angle. These composites respect the time-temperature superposition principle and can be fitted by a visco-elastic rheological model.
This study presents the results of direct tensile tests performed at three different temperatures on mixtures containing crumb rubber (CR). These tests were conducted on GB5® mixtures, which included varying percentages (0
This study was conducted to assess the time evolution of the linear viscoelastic behaviour of bituminous mixtures containing crumb rubber added by dry process. For this purpose, three mixtures containing 0
The bonding at the interface is seen in most road structure design methods as perfectly bonded and this ideal condition is supposed to last during the whole service life. However, the real working capacity at the interface is complex and gradually degrades because of repeated traffic load and weather conditions. A new apparatus called 2T3C (“Torsion, Traction, Compression sur Cylindre Creux” in French, or “Torsion, Tension, Compression on Hollow Cylinder” in English) was developed at the ENTPE/University of Lyon to investigate the thermomechanical behaviour of the interface between bituminous layers. The sample has two layers of similar or different materials. It has a total height of 125 mm and a wall thickness of the 25mm. The thickness was chosen small enough in order to validate the assumption of quasihomogenous stress and strain fields between inner and outer surface. The DIC technology is performed considering the three listed elements: (1) two couple of cameras are situated on two opposite sides of the sample (4 cameras in total) to take 3D images; (2) a software is employed to calculate the displacements in 3 dimensions of all points in the area of interest; and (3) the strain fields in the 2 layers and the displacement gap at the interface are computed using a method developed at the University of Lyon/ENTPE. In this paper, a focus is proposed on the viscoelastic and fatigue behaviours of the two bituminous layers and of the interface. Small and large numbers of loading cycles are successively considered in the small strain domain.
Evaluating the shear fatigue behaviour of the interface between two bituminous layers is important in order to assure the structure can withstand repeated loadings. The 2T3C (Torsion-Traction-Compression sur Cylindre Creux, in French) hollow cylinder apparatus developed at the University of Lyon/ENTPE can apply torsion, tension, and compression on hollow cylinder samples. In this study, the apparatus is combined with 3D Digital Image Correlation (3D-DIC) to study the fatigue behaviour of the double-layered hollow cylinder samples with the interfaces with a tack coat made of emulsified bitumen. The fatigue failure was observed at the interfaces and not in layers. Six fatigue criteria are proposed to determine the fatigue life of an interface. They are: (1) the 50% reduction of stiffness criterion; (2) the maximal phase angle criterion, (3) maximal dissipated energy criterion, (4) slope change in the dissipated energy evolution curve criterion, (5) slope change in the stress evolution curve criterion, and (6) slope change in the displacement gap at the interface evolution curve criterion. The first criterion is more conservative in comparison with the last five criteria.
Rheological properties of a Colombian bituminous binder (60/70 pen grade) and its corresponding SBS blends were studied. A total of 4 binders were tested, including the neat bitumen and three blends with different SBS dosages (4%, 7% and 10% by total mass). Standard tests were performed to determine penetration, softening point, elastic recovery (tension and torsion), and viscosity (dynamic viscosity curves by Brookfield viscosimeter). Finally, Linear viscoelastic (LVE) behaviour was analysed for the neat bitumen and the 4% SBS blend. Norm and phase angle of complex shear modulus were obtained from Dynamic Shear Rheometer (DSR) frequency sweep tests on a wide range of temperatures and frequencies. Black and Cole-Cole plots and complex modulus (norm and phase angle) master curves were plotted. Experimental data were successfully simulated using the analogical 2S2P1D model.
Thermomechanical behaviour of interface of pavement structures was experimentally characterized using the 2T3C (French acronym for “Torsion-Tension-Compression on Hollow Cylinder”) apparatus developed at ENTPE, combined with 3D Digital Image Correlation (3D-DIC). Tested samples are double-layered hollow cylinders with an interface composed of bitumen emulsion and two layers made of different materials. In this study, cyclic torsion is applied on tested samples at combinations of five temperatures (from 0°C to 40°C) and five frequencies (from 0.01 to 1 Hz). Two different levels of global strain amplitudes are applied on samples (50 and 200 µm/m). The experimental results show that Time-Temperature Superposition Principle holds for all tested conditions. The linear viscoelastic model 2S2P1D perfectly simulates the behaviour of interfaces for the lower global strain amplitude, but not for the higher global strain amplitude because of an underestimation of phase angle. Successful simulation of phase angle values were obtained using the DBNPDSC model.
In this study, the influence of crumb rubber added by dry process on the dissipative properties of bituminous mixtures has been investigated. Six mixtures were tested: two reference mixtures without crumb rubber were produced, one containing a pure binder and the other one containing an SBS-modified binder. Four other mixtures containing 2 % and 4 % (in mass) of crumb rubber and the same two mentioned binders were also produced. First, complex modulus tests were carried out at different frequencies and temperatures and for an axial strain amplitude of 50 mu m/m. The 2S2P1D analogical linear viscoelastic model was used to model experimental data. From these results, the dissipated energy during one sinusoidal loading cycle was calculated, both for constant strain amplitude loading and for constant stress amplitude loading. Simulations at constant strain amplitude show lower energy dissipation for bituminous mixtures containing 2 % and 4 % CR and the SBS-modified bitumen with respect to the corresponding reference mixture without CR. However, the opposite is found at constant stress amplitude. Mixtures containing pure bitumen and CR show lower dissipated energy for any loading mode. Numerical simulations were carried out to determine energy dissipation at the scale of a pavement structure. The linear viscoelastic properties determined experimentally for the different materials were implemented in a finite element model to perform the calculations in the frequency domain. Results show that pavements made with bituminous mixtures containing CR added by dry process dissipate more energy with respect to the reference mixtures without CR.
The main results of the French collaborative project Improvmure, funded by the National Research Agency are presented. A very wide study was performed on bituminous mixtures produced on experimental sites or in laboratory at different scales (semi-industrial process or lab process). Objectives were to evaluate the influence of the recycling rate (0%, 40% and 70%), the number of successive recycling cycles (from 1 to 3 cycles), and the manufacturing process (hot mix or warm mix using additives or foamed bitumen). The impact was studied in laboratory on the emitted fumes during the production process, on the properties used for mix design (compactability, water sensitivity, complex modulus at 15 degrees C/10 Hz, fatigue), on advanced thermomechanical behaviour (linear viscoelasticity, crack behaviour at low temperature, crack propagation) and on surface properties (skid resistance). The paper gathers many valuable results, showing the possibility to recycle bituminous mixtures multiple times, at high rates and with warm manufacturing process.
The fatigue properties of asphalt mixes are usually established according to a series of laboratory cyclic loading tests at fixed temperature (θ) and frequency (f). Unfortunately, the results are different when considering different types of tests. This paper proposes a rational method to link the 4-point bending (4PB) and tension/compression (TC) fatigue test results. First a theoretical analysis of the 4PB fatigue test is presented. This analysis allows to introduce a link between the fatigue curves (modulus or damage versus number of cycles) of this type of test with the results from uniform TC tests. The 4PB fatigue curve is superimposed with the TC fatigue curve if the reference strain amplitude is correctly chosen. This amplitude must not be the maximum strain at the boundary fibre of the beam (ε0max), as usually considered, but a lower proposed value (ε0h0). Then, the previous approach is extended to the fatigue failure Wöhler’s law, which is a line in the Log–Log plot of the number of cycles at failure versus the loading amplitude (stress or strain). An experimental campaign on a Canadian asphalt concrete (HMAC) validates the developped approach. It is an important output from this research as fatigue life from one type of test can be obtained from the other type of test in the case of strain (or displacement) control tests.
Validation of time-temperature superposition principle (TTSP) in the fatigue domain for a high modulus asphalt concrete (HMAC) is presented in this paper. All tests were performed in tension-compression under strain control mode. First, TTSP was validated in the linear viscoelastic domain. Then, fatigue tests were performed under three loading conditions, 9.2 degrees C and 5 Hz, 11.0 degrees C and 10 Hz and 12.9 degrees C and 20 Hz, which are equivalent regarding TTSP. Two fatigue protocols were adopted: continuous fatigue test (FT) and fatigue test with rest period (FTRP). For FT, three samples were tested at 180 mu m/m for each loading condition whereas for FTRP, one sample was tested at 100 mu m/m. The data were analysed by comparing the dynamic modulus evolution as a function of time or the fatigue life duration. The results showed that HMAC with fatigue damage remains thermorheologically simple (i.e., respects the TTSP) in the studied temperatures range.
This study investigated the effect of temperature on the fatigue behavior of high-modulus asphalt concrete (HMAC). The impact of temperature on the W & ouml;hler's curve parameters, especially with respect to the slope and intercept, was studied to evaluate the accuracy of current methodologies used to estimate them in mechanistic-empirical pavement design tools (M-E PD). Four series of tension-compression (TC) fatigue tests were conducted on HMAC samples at four temperatures (-8.3 degrees C, 1.4 degrees C, 11.3 degrees C, and 21.0 degrees C). The findings indicated a significant impact of temperature on the intercept; the slope did not exhibit any statistically significant effect.
Geosynthetics have been a reinforced solution for pavement structures for more than 80 years and could be effective in extending its service life. There is a lack of consolidated design methods for pavement with this reinforcement. Therefore, this work aims at proposing a new rational design approach for reinforced structures based on the French design method. In this approach, the geogrid contribution was included by improving the fatigue and rutting properties of some layers, using coefficients named kmaj and kmaj_Z. Three hypotheses were considered concerning the condition of an old bituminous layer remaining from rehabilitation works. The first one considered this layer in healthy condition to simulate a new reinforced structure. The second one considered it as cracked and the third one as disintegrated to simulate the design for rehabilitation. Two placement positions and two geogrid-interface conditions (bonded and not bonded) were analyzed. The results indicated that the geogrid was most effective in a completely deteriorated structure and it should be placed in the lowest possible position in the bound layers. This method can be used for any geogrid position within the structure. Lastly, the reinforcement by geogrid allows a reduction of the thickness of the layer above it.
To improve pavement design methods, the behaviour of the interfaces between pavement layers must be precisely taken into account. This paper describes the thermomechanical behaviour of an interface in the small strain domain. Thanks to the innovative 2T3C Hollow Cylinder Apparatus (2T3C HCA), the behaviour of an interface was observed directly with 3D Digital Image Correlation (3D DIC). The studied samples were constituted of two layers of bituminous mixtures with a bitumen emulsion at the interface. From cyclic tests performed at four different frequencies (from 0.01 to 0.3 Hz) and four different temperatures (from 10 to 40 °C), the complex interface stiffnesses in the tension–compression mode and in the shear mode were obtained. The time–temperature superposition principle was verified for the abovementioned stiffnesses. Knowing a 2S2P1D (2 Springs, 2 Parabolic elements, 1 Dashpot) model of the bitumen of the tack coat, the norms of the complex interface stiffnesses were modelled with 2S2P1D models by using a SHStS (Shift, Homothety, Shift and time Shift) transformation.
Pavement layer interfaces have a major influence on pavement structure service life and bearing capacity. This study aims at understanding how interfaces behaviour can affect the pavement response in the frame of airfield pavement assessment with the Heavy Weight Deflectometer device. Based on laboratory viscoelastic characterisation of an interface, numerical simulations allow comparing various interface conditions: fully bonded, perfectly sliding and with elastic or viscoelastic stiffnesses. It is shown that interface viscoelasticity participates to the pavement surface deflection response. This contribution is significantly affected by pavement temperature and load frequency content. Computations also demonstrate that, in the studied case, interface viscoelasticity does not question current rational pavement design methods for new roads, whose behaviour stays close to the 'fully bonded' case.
Black Space diagrams representing rheological data of asphalt materials in the form of complex modulus (|G*| or |E*|) versus phase angle (delta) have been successfully used for interpretation of material behavior and performance. Previous studies have used Black Space for identification of testing geometry compliance errors when testing over multiple temperatures and loading times (frequencies), screening of the "thermo-rheological simplicity" of various binders and mixtures, and detailed evaluation of the performance balance in term of "stiffness" versus "re-laxation" needs. This paper provides an overview of how Black Space can be further used to provide a greater understanding of the concepts of damage and healing and cracking susceptibility and fracture, and to also quantify the complex rheological response of alternative binders. In terms of the damage assessment, cyclic loading tests were analyzed using Black Space to identify additional physical phenomena such as nonlinearity, self-heating, and thixotropy. The cracking analysis has included thermal, fatigue, and durability cracking as well as the use of Black Space to access the performance of asphalt mixtures subjected to aging as well as rejuvenation and materials with recycled asphalt. Concepts such as the Glover-Rowe parameter that are based around Black Space and linked to other forms of rheological indices such as the low-temperature stiffness and relaxation rate parameters are introduced. The results in the paper show that Black Space provides a critical means of rheological characterization to investigate and evaluate the properties and performance of both binders and mixtures. This is particularly relevant at a time when there is a concerted move within the asphalt paving industry toward more sustainable solutions and increased demand for reuse and recycling of materials in asphalt mixtures.