Bituminous Interlayers Thermomechanical Behaviour under Small Shear Strain Loading Cycles with 2T3C Apparatus: Hollow Cylinder and Digital Image Correlation ^†

1. Introduction

A road/airport pavement is a multi-layered structure generally composed of several layers of bituminous materials, and cement-bound materials on unbound granular materials. In the design phase, the different bituminous layers are considered perfectly bonded and therefore expected to work as a unique structure throughout the service life of the pavement. However, due to environmental, traffic, and/or material-related conditions, the quality of the bond changes over time. The layers in the structure tend to work more and more independently with a degrading bonding capacity, which induces a reduction in the pavement life. Therefore, the mechanical behavior of the interface between bituminous layers has recently been studied through some original approaches. There were tests developed to focus on the interface study, for example, the Ancona shear testing research and analysis device at the Marche Polytechnic University [1,2,3], the shear-torque fatigue testing device at the University of Limoges [4,5], and other tests [6]. Most of the studies on this topic present several limitations. Only one or a few loading configurations can be applied to the sample (for example, pure shear of the interface) and studies focus only on the interface strength. Moreover, stress and strain fields within the sample are not homogenous, therefore not allowing investigation of the intrinsic mechanical behavior of the interface. In this study, the 2T3C (“Torsion, Traction/Compression sur Cylindre Creux” in French, Torsion, Traction/Compression on Hollow Cylinder) apparatus developed at ENTPE is used to investigate the behavior of a bituminous interface under shear loading and small strain cycles.

2. T3C Apparatus

The device consists of different basic parts: (1) a servo-hydraulic press capable of imposing axial and shear loading (cyclic or monotonic) on a hollow cylindrical specimen, equipped with a thermal chamber controlling the temperature; (2) four cameras, in pairs, used to perform digital image correlation (DIC) analysis, in order to determine the three-dimensional strain field in the upper and lower layers and to calculate the relative displacements at the interface between different layers; (3) several displacement sensors around the specimen to control its global deformation: one pair of noncontact sensors to control displacements in the vertical direction, and another pair to control torsional displacement. The sample has a total height of 125 mm, an outer radius of 86 mm, and an inner radius of 61 mm. The small thickness of the cylindrical wall of the sample allows consideration of quasi-homogeneous strain and stress fields.

3. Material and Experimental Procedure

The sample tested in this study is composed of two different bituminous layers (classically used in France as base and surface layers) with an interface in between made of a tack coat (bitumen emulsion). The sample was cored from a slab produced by successive compaction of the two layers using a wheel compactor. The sample was tested at 0 °C, 10 °C, 20 °C, 30 °C, and 40 °C by applying sinusoidal torsion at 5 different frequencies, 0.01, 0.03, 0.1, 0.3, and 1Hz. The global shear strain amplitude applied was 200µm/m, while normal stress was maintained at 0 MPa during the test. The two pairs of cameras were placed on opposite sides of the sample and each camera took 50 photos per loading cycle (but 35 photos per loading cycle at the loading frequency of 1 Hz). A specific analysis [7,8] was developed at ENTPE to compute the strains in both layers and the displacement gap at the interface.

4. Results

The complex shear moduli G_θz* of the two bituminous mixtures (upper and lower layers) were obtained. Since Time-Temperature Superposition Principle is validated for them, master curves are plotted at 20 °C (Figure 1a for the norm, and Figure 1b for the phase angle). Corresponding shift factors are plotted in Figure 1c. The interface also shows a viscoelastic behavior as the two mixtures. Its complex shear stiffness K_θz*, defined as the ratio of shear stress amplitude over the torsional displacement amplitude at the interface, could then be determined. This interface parameter exhibits a similar evolution with temperature and frequency as the viscoelastic parameter G_θz*. It is then possible to plot its master curves (norm and phase angle), which are also shown in Figure 1.

5. Discussion

The viscoelastic behavior of a multilayered sample composed of two mixtures, as well as their interface, was successfully investigated using the 2T3C apparatus. The DIC technology allows determining a displacement gap at the interface as low as 1–2 µm. These results show the potential of the device and the analysis.