The adhesive bonding strength of the membrane layers between the asphalt concrete surface layers and the decks of steel bridges has a strong influence on the fatigue life of orthotropic steel deck bridges (OSDBs). The interfacial properties between the membrane and the layers bonded to it have not been extensively studied in the current orthotropic steel deck bridge system. For the adequate characterisation of the adhesive-bonding strength of various membranes and surrounding materials on OSDBs and for the collection of the necessary parameters for finite element model, details of the membrane adhesion test (MAT) are introduced and simulated by using the adhesive traction-separation interface element which was developed in a companion paper to this contribution (Liu, X., Kasbergen, C., Li, J., & Scarpas, A. (2019). Modelling of membrane bonding response: part 1 development of an adhesive contact interface element. International Journal of Pavement Engineering). Parametric studies of the adhesive contact element utilised for modelling the membrane bonding layer in the MAT test have been performed on the basis of the combination of different critical strain energy release rates and the characteristic opening length in the constitutive model. Comparison of membrane deformation profiles and the in-time debonding force distribution between experimental observations and finite element simulations have been presented.

@en

The orthotropic steel deck bridges (OSDBs) in the Netherlands consists mostly of multilayer system: top porous asphalt layer, guss asphalt layer, steel deck and two membrane layers. The five-point bending (5PB) beam test is the standard device in France for characterization of fatigue response of asphalt concrete surfacings used on orthotropic steel deck bridges. In this paper, an analytical solution for the 5PB beam test setup is presented first. In order to better understand the influence of geometrical, mechanical and structural parameters on the performance of the typical multilayer surfacing system of OSDBs, the 5PB test specimens with five structural layers have been investigated. The parametric study is performed at the numerical platform CAPA-3D that was developed at the Section of Structural Mechanics of TU Delft. The influences of the thickness of the asphalt layers and the mechanical properties of both top and bottom membrane layers are studied. The sensitivities of those influence factors are evaluated by the examination of the maximum tensile stress at the top surface of the porous asphalt layers and the strain distributions through the entire thickness of the specimen at two cross-sections.

@en

In the Netherlands asphaltic surfacings on orthotropic steel deck bridges (OSDB) mostly consist of two structural layers. The upper layer consists of what is known as very open porous asphalt (ZOAB) for noise reduction. For the lower layer Guss Asphalt (GA) is used. Earlier investigations have shown that the bonding characteristics of membrane layers to the surrounding materials have a very strong influence on the overall response of the steel bridge decks. Rijkswaterstaat, an agency of the Dutch Ministry of Infrastructure and the Environment, has commissioned Delft University of Technology to investigate and rank the performance of various commercially available membranes. In order to obtain insight into the response of membranes and their interaction with the surrounding materials on orthotropic steel decks, a project of evaluation of the performance of modern surfacing systems on OSDBs has been undertaken. Currently, there are various kinds of membranes provided by various companies. Thereby it was necessary to examine the bonding strength of these membrane products and to develop a ranking methodology. The research project focused on membrane performance and the effects hereof on the bridge deck as a whole. The methodology used was a multi-phase approach, which consisted of three main phases. In Phase 1, a Membrane Adhesion Test (MAT) device was developed at Delft University of Technology for the characterization of the adhesive bonding strength of membranes with the surrounding materials on OSDBs on the basis of a fundamentally sound, mechanistic methodology. Several membrane products were tested monotonically in this phase. In Phase 2, the MAT device was utilised for investigation of the fatigue response of the various membrane products on various substrates and under two different temperature conditions and three different cyclic load levels. A ranking methodology consisting of a combination of experimental (via MAT) and computational investigations was also developed and utilized for the ranking of the various membrane products. In Phase 3 of the project, four typical Dutch multilayer surfacing systems, constructed with five selected membrane products from Phase 1 and 2, were studied by means of five-point bending (5PB) beam tests and FE simulations. The findings of the 5PB beam tests were used for calibration and validation of the finite element predictions and for further ranking of the performance of the various membranes in Dutch OSDBs. On the basis of the project results, the top two ranking membranes were selected for subsequent testing by means of the LINTRACK facility available at Delft University of Technology. @en

The French five-point bending (5PB) beam test provides a laboratory scale test that allows studying the fatigue resistance of surfacing systems on orthotropic steel deck bridges (OSDB). The surfacing structure for OSDB in the Netherlands consists mostly of a multilayer system: top porous asphalt layer, Guss asphalt layer, steel deck, and two membrane layers between the asphalt surfacing layers. In this paper, the 5PB beam test was employed to evaluate the performance of the whole multilayer structure on a steel deck and also to evaluate the mechanical properties of both the top and bottom membrane and the influence they have on the structure. Four membrane products were used as the top and bottom membrane layers in the 5PB beams. Furthermore, in the last part of this paper, the dissipated energy was used to explain the incremental damage during the testing.@en

In order to adequately characterize the fatigue response of the various membranes with surrounding multilayer surfacing layers on orthotropic steel decks and collect the necessary parameters for FE modeling, the details of the cyclic membrane adhesion tester (MAT) were introduced. The fatigue damage in membrane interface was related to the amount of dissipated work computed by using the measurement of actuator load and piston deformation during the loading cycle. The dissipated work, which is equivalent to the lost part of the total potential energy of the membrane, was utilized to explain the incremental damage during the testing. Furthermore, using the experimental data obtained from MAT, ranking of the bonding characteristics of various membrane products was demonstrated as well as the role of other influencing factors, such as the types of substrate and test temperatures.@en