Scaling laws for opening partially adhered contacts in MEMS

Paolo Decuzzi, David J. Srolovitz

Research output: Contribution to journalArticlepeer-review

21 Scopus citations

Abstract

MEMS switches represent a promising technology for a wide range of communications applications. They operate at high frequency, opening and closing the contact between metal pads. Surfaces are rough in nature, therefore a predictive description of contact switching requires understanding of asperity contact and adhesion. A simple model is presented to estimate adhesion forces in MEMS switches and the total forces required to detach the pads. The interface between two rigid pads is modeled as an array of uniformly distributed elastic asperities mimicking the interface roughness. Two types of asperities are considered: bonded asperities and adhered asperities, as described by the JKR adhesion model. The effects of asperity size, density, and load-displacement relationship are analyzed as a function of the loading conditions (i.e., tensile load and bending moment). The minimum force for detachment is proportional to the asperity size and asperity density, except when bending moments and tensile forces are combined (which leads to a more complex relationship). The results are summarized in a loading-detachment map. The most effective method to detach the contacts is through a combination of a bending moment and a tensile force. The implication of these results to MEMS designs are discussed and approaches to modifying the moment and tensile force ratio are suggested.

Original languageEnglish (US)
Pages (from-to)377-385
Number of pages9
JournalJournal of Microelectromechanical Systems
Volume13
Issue number2
DOIs
StatePublished - Apr 2004

Keywords

  • Contact adhesion
  • Detachment force
  • Roughness

ASJC Scopus subject areas

  • Electrical and Electronic Engineering
  • Mechanical Engineering

Fingerprint

Dive into the research topics of 'Scaling laws for opening partially adhered contacts in MEMS'. Together they form a unique fingerprint.

Cite this