Probabilistic strength of {1 1 1} n-type silicon

A. A. Wereszczak, A. S. Barnes, K. Breder, Sukhminder Binapal

Research output: Contribution to journalArticlepeer-review

20 Scopus citations

Abstract

The two-parameter Weibull strength distribution of {1 1 1} n-type silicon prismatic bars was determined in four-point bending and analyzed as a function of specimen size (width), loading rate, two different crystallographic orientations, and specimen orientation (polished or etched surface in tension). 100% fractography was performed to classify strength-limiting flaw types and to censor the strength data. All flaw types were extrinsic. Machining or cutting damage in the form of chipped edges (an edge-type flaw) was the dominant strength-limiting flaw when the polished surface was subjected to tensile stress, while a flat-bottomed etch pit (a surface-type flaw) was the dominant strength-limiting flaw when the etched surface was subjected to tensile stress. The censored Weibull strength distribution was independent of specimen width, loading rate (indicative of slow crack growth insusceptibility), and the two crystallographic orientations; however, it was dependent on specimen orientation. Pooling of the strength data was employed to tighten the confidence intervals about the censored parameters. The results from this study indicate that different extrinsic strength-limiting flaws and strength distributions will be operative depending on the manner in which a silicon component is stressed.

Original languageEnglish
Pages (from-to)291-303
Number of pages13
JournalJournal of Materials Science: Materials in Electronics
Volume11
Issue number4
DOIs
StatePublished - Jun 2000

Funding

Research sponsored by the US Department of Energy, Assistant Secretary for Energy Efficiency and Renewable Energy, Office of Transportation Technologies, as part of the Advanced Automotive Materials Program, under contract DE-AC05-00OR72564 with UT-Battelle, LLC.

Fingerprint

Dive into the research topics of 'Probabilistic strength of {1 1 1} n-type silicon'. Together they form a unique fingerprint.

Cite this