# Flexural fracture strength, fracture locations, and Monte Carlo predictions for a silicon nitride by ten U.S. laboratories

## Abstract

The work reported was conducted to provide a basis for a number of structural ceramic mechanical property standardization activities in the US, Germany, Japan, and Sweden. A comparison of key property values of a commercial silicon nitride determined in a number of laboratories was a major objective. The work reported was conducted by 10 US laboratories on GN-10 silicon nitride, and represented the US work within an International Energy Agency program including the US, Germany, Japan, and Sweden. Flexural strengths were measured at 25 and 1250 C in air and were analyzed using the two-parameter Weibull model in terms of m and {sigma}{sub {Theta}} using both linear regression (LR) and maximum likelihood (ML) methods. Under the measurement conditions for the 10 room-temperature strength sets, the value of the ML estimator for m varied by as much as 36%, while the value for the {sigma}{sub {Theta}} parameter estimator varied only 3.3%. The LR estimator for m varied by about 54%. For the high-temperature specimens, the NL estimator for m varied by 48% while the LR estimator varied by 38%. Monte Carlo predictions showed that for room-temperature strength, the maximum likelihood estimator m for all 10 laboratories fit within the 10% andmore »

- Authors:

- Publication Date:

- Research Org.:
- Oak Ridge National Lab., TN (US)

- OSTI Identifier:
- 20075975

- Resource Type:
- Journal Article

- Journal Name:
- Journal of the American Ceramic Society

- Additional Journal Information:
- Journal Volume: 83; Journal Issue: 5; Other Information: PBD: May 2000; Journal ID: ISSN 0002-7820

- Country of Publication:
- United States

- Language:
- English

- Subject:
- 36 MATERIALS SCIENCE; FRACTURE PROPERTIES; SILICON NITRIDES; DATA COMPILATION; PREDICTION EQUATIONS; MONTE CARLO METHOD; STATISTICAL DATA

### Citation Formats

```
Tennery, V.J., Breder, K., Ferber, M.K., and Jenkins, M.G.
```*Flexural fracture strength, fracture locations, and Monte Carlo predictions for a silicon nitride by ten U.S. laboratories*. United States: N. p., 2000.
Web. doi:10.1111/j.1151-2916.2000.tb01351.x.

```
Tennery, V.J., Breder, K., Ferber, M.K., & Jenkins, M.G.
```*Flexural fracture strength, fracture locations, and Monte Carlo predictions for a silicon nitride by ten U.S. laboratories*. United States. doi:10.1111/j.1151-2916.2000.tb01351.x.

```
Tennery, V.J., Breder, K., Ferber, M.K., and Jenkins, M.G. Mon .
"Flexural fracture strength, fracture locations, and Monte Carlo predictions for a silicon nitride by ten U.S. laboratories". United States. doi:10.1111/j.1151-2916.2000.tb01351.x.
```

```
@article{osti_20075975,
```

title = {Flexural fracture strength, fracture locations, and Monte Carlo predictions for a silicon nitride by ten U.S. laboratories},

author = {Tennery, V.J. and Breder, K. and Ferber, M.K. and Jenkins, M.G.},

abstractNote = {The work reported was conducted to provide a basis for a number of structural ceramic mechanical property standardization activities in the US, Germany, Japan, and Sweden. A comparison of key property values of a commercial silicon nitride determined in a number of laboratories was a major objective. The work reported was conducted by 10 US laboratories on GN-10 silicon nitride, and represented the US work within an International Energy Agency program including the US, Germany, Japan, and Sweden. Flexural strengths were measured at 25 and 1250 C in air and were analyzed using the two-parameter Weibull model in terms of m and {sigma}{sub {Theta}} using both linear regression (LR) and maximum likelihood (ML) methods. Under the measurement conditions for the 10 room-temperature strength sets, the value of the ML estimator for m varied by as much as 36%, while the value for the {sigma}{sub {Theta}} parameter estimator varied only 3.3%. The LR estimator for m varied by about 54%. For the high-temperature specimens, the NL estimator for m varied by 48% while the LR estimator varied by 38%. Monte Carlo predictions showed that for room-temperature strength, the maximum likelihood estimator m for all 10 laboratories fit within the 10% and 90% confidence bounds for 30 specimen sets. The dispersion of the high-temperature data was such that the m estimator satisfied the model only at the 1% and 99% confidence levels for the 15 specimen sets. ANOVA results showed that for the room-temperature flexural strength, data from all 10 laboratories were not distinguishable for this evaluator at the 95% confidence level and that scatter within individual data sets was a larger effect than was the variation between the data sets. For the high-temperature data, the results from one laboratory were clearly outside the allowable range at this confidence level.},

doi = {10.1111/j.1151-2916.2000.tb01351.x},

journal = {Journal of the American Ceramic Society},

issn = {0002-7820},

number = 5,

volume = 83,

place = {United States},

year = {2000},

month = {5}

}