Abstract
In this paper a model is presented that describes the distribution of adhesion values typically experimentally observed for different MEMS devices that have been fabricated in the same way. This spread is attributed to the fact that different devices differ in the details of their surface roughness, even if these surface roughnesses are modeled as coming from the same ‘parent’ stochastic process.Using Monte Carlo simulations, the effect of surface roughness and relative humidity has been evaluated in detail, both on the expected mean value of the surface interaction energy between the MEMS surfaces, and the expected spread on this value from device to device.By comparing the new model to existing literature reporting this experimentally observed spread, we have found excellent agreement between the experimental spread observed, and the spread calculated with the theoretical model using Monte Carlo simulations.This work paves the way to detailed adhesion failure predictive modeling. It may be used to assess the reliability of MEMS designs that rely on contacting surfaces for their operation, but have a limited restoring force available to separate the surfaces when in contact.
Highlights
It has long been recognized thatadhesion failures, called stiction in the MEMS community, pose both a yield and a reliability hazard [1]
To understand the adhesion caused by capillary condensation and van der Waals molecular forces, we have to calculate how they are influenced by the different realizations of the surface roughness, and the corresponding differences in distance distribution of the area between the surfaces
Several studies have been reported where adhesion was measured on multiple MEMS devices simultaneously on the same wafer in a cantilever beam array (CBA) configuration
Summary
It has long been recognized that (auto-)adhesion failures, called stiction in the MEMS community, pose both a yield and a reliability hazard [1]. The key factor determining the magnitude of the adhesion of otherwise similar surfaces in a common environment is the surface roughness of the contacting bodies. That the adhesion is a distribution has been recognized by several authors [8,9,10] This notion is extremely important when one wants to do an a priori yield or reliability assessment of a certain MEMS device or MEMS technology. In this paper the question is addressed of how adhesion distributions of ‘identical’ devices with surface roughness come about, how to model them, and how to measure them. It will be shown that the model presented is useful for addressing adhesion as such, but is the missing link for an adhesion reliability predictive model This model can be used for the prediction of field stiction failures in a known environment
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
