Contact between elastic bodies with self-affine rough surfaces is mostly studied with a focus on determining surface fields, despite body fields are of great importance to establish, for instance, when and where elasticity breaks down. This work aims at analyzing the effect of contact roughness on the body fields of compressible frictionless solids modeled using Green's function molecular dynamics. Although area-load curves are insensitive to changes in the Hurst exponent as long as they are correctly normalized and are clearly not affected by compressibility, the Von-Mises stress is found to depend on both Hurst exponent and Poisson's ratio.

@en

Although indentation of elastic bodies by self-affine rough indenters has been studied extensively, little attention has so far been devoted to plasticity. This is mostly because modeling plasticity as well as contact with a self-affine rough surface is computationally quite challenging. Here, we succeed in achieving this goal by using Green's function dislocation dynamics, which allows to describe the self-affine rough surface using wavelengths spanning from 5 nm to 100μm. The aim of this work is to gain understanding in how plastic deformation affects the contact area, contact pressure and hardness, gap profile and subsurface stresses, while the roughness of the indenter is changed. Plastic deformation is found to be more pronounced for indenters with larger root-mean-square height and/or Hurst exponent, and to be size dependent. The latter means that it is not possible to scale observables, as typically done in elastic contact problems. Also, at a given indentation depth (interference) the contact area is smaller than for the corresponding elastic contact problem, but gap closure is more pronounced. Contact hardness is found to be much larger than what reported by classical plasticity studies. Primarily, this is caused by limited dislocation availability, for which the stiffness of the deforming crystal is in between that of a linear elastic and an elastic-perfectly plastic material. When calculating hardness and nominal contact pressure, including very small wavelength in the description of the surface is not necessary, because below a given wavelength the subsurface stresses become invariant to a further decrease in true contact area. This is true for both elastic and plastic materials. Considering small wavelengths is instead required to capture accurately roughening and contact stress distribution.

@en

Persson's theory allows for a fast and effective estimate of contact area and contact stress distributions when a flat and a self-affine rough surface are pressed into contact. For elastic bodies, the results of the theory have been shown to be in very good agreement with rather costly simulations. The theory has also been extended to plastic bodies. In this work, the results of Persson's theory for plastic bodies are compared with those of discrete dislocation plasticity. The area–load curves obtained by theory and simulations are found to be in good agreement when the rough surface has a very small root-mean-square (rms) height. For larger rms heights, which are more realistic for metal surfaces, the agreement is no longer good unless in the theory, instead of a size-independent material strength, one uses a rms height- and resolution-dependent yield strength. A modification of this type, i.e., the use of a yield strength dependent on size, does however not lead to agreement between the probability distributions of the contact stress, which is much broader in the simulations than in the theory. The most likely reason for this discrepancy is that the theory, apart from neglecting plasticity size dependence, only applies to elastic-perfectly plastic bodies and therefore, neglects strain hardening.

@en

The Green's function molecular dynamics (GFMD) method for the simulation of incompressible solids under normal loading is extended in several ways: shear is added to the GFMD continuum formulation and Poisson numbers as well as the heights of the deformed body can now be chosen at will. In addition, we give the full stress tensor inside the deformed body. We validate our generalizations by comparing our analytical and GFMD results to calculations based on the finite-element method (FEM) and full molecular dynamics simulations. For the investigated systems we observe a significant speed-up of GFMD compared to FEM. While calculation and proof of concept were conducted in two-dimensions only, the methodology can be extended to the three-dimensional case in a straightforward fashion.

@en

Metals deform plastically at the asperity level when brought in contact with a counter body even when the nominal contact pressure is small. Modeling the plasticity of solids with rough surfaces is challenging due to the multi-scale nature of surface roughness and the length-scale dependence of plasticity. While discrete-dislocation plasticity (DDP) simulations capture size-dependent plasticity by keeping track of the motion of individual dislocations, only simple two-dimensional surface geometries have so far been studied with DDP. The main computational bottleneck in contact problems modeled by DDP is the calculation of the dislocation image fields. We address this issue by combining two-dimensional DDP with Green's function molecular dynamics. The resulting method allows for an efficient boundary-value-method based treatment of elasticity in the presence of dislocations. We demonstrate that our method captures plasticity quantitatively from single to many dislocations and that it scales more favorably with system size than conventional methods. We also derive the relevant Green's functions for elastic slabs of finite width allowing arbitrary boundary conditions on top and bottom surface to be simulated.@en