Interfacial adhesion and friction are important factors in determining the performance and reliability of microelectromechanical systems. We demonstrate that the adhesion of micromachined surfaces is in a regime not considered by standard rough surface adhesion models. At small roughness values, our experiments and models show unambiguously that the adhesion is mainly due to van der Waals dispersion forces acting across extensive non-contacting areas and that it is related to 1/Dave2, where Dave is the average surface separation. These contributions must be considered because of the close proximity of the surfaces, which is a result of the planar deposition technology. At large roughness values, van der Waals forces at contacting asperities become the dominating contributor to the adhesion. In this regime our model calculations converge with standard models in which the real contact area determines the adhesion. We further suggest that topographic correlations between the upper and lower surfaces must be considered to understand adhesion completely.

No abstract

An elementary theory for a rigid spherical indenter contacting a thin, linear elastic coating that is bonded to a rigid substrate was developed. This theory predicts that contact area varies as the square root of the compressive load in contrast to Hertz theory where contact area varies as the two-thirds power of the compressive load. Finite element analysis confirmed an approximate square root dependence of contact area on compressive load when the coating thickness-to-indenter radius ratio is less than 0.1 and when the coating Poisson’s ratio is less than 0.45. Thin-coating contact mechanics theories that use either the Derjaguin-Muller-Toporov (DMT) approximation or the Johnson-Kendall-Roberts (JKR) approximation were also developed. In addition, a finite element simulation capability that includes adhesion was developed and verified. Illustrative finite element simulations that include adhesion were then performed for a thin elastic coating (rigid indenter/substrate). Results were compared with the thin-coating contact theories and the transition from DMT-like to JKR-like response was examined.

No abstract

No abstract

No abstract

A method for modeling the initiation and growth of discrete delaminations in shell-like composite structures is presented, The laminate is divided into two or more sublaminates, with each sublaminate modeled with four-noded quadrilateral shell elements. A special, eight-noded hex constraint element connects opposing sublaminate shell elements. It supplies the nodal forces and moments needed to make the two opposing shell elements act as a single shell element until a prescribed failure criterion is satisfied. Once the failure criterion is attained, the connection is broken, creating or growing a discrete delamination. This approach has been implemented in a threedimensional finite element code. This code uses explicit time integration, and can analyze shelllike structures subjected to large deformations and complex contact conditions. The shell elements can use existing composite material models that include in-plane laminate failure modes. This analysis capability was developed to perform crashworthiness studies of composite structures, and is useful whenever there is a need to estimate peak loads, energy absorption, or the final shape of a highly deformed composite structure. This paper describes the eight-noded hex constraint element used to model the initiation and growth of a delamination, and discusses associated implementation issues. Particular attention is focused on the delamination growth criterion, and it is verified that calculated results do not depend on element size. In addition, results for double cantilever beam and end notched flexure specimens are presented and compared to measured data to assess the ability of the present approach to model a growing delamination.

No abstract