Department of physics and technology, University of Bergen, Allegaten 55, NO-5007 Bergen, Norway.
J Hazard Mater. 2010 Mar 15;175(1-3):293-7. doi: 10.1016/j.jhazmat.2009.09.162. Epub 2009 Oct 9.
A copy of the standard ASTM spark generator for determination of MIEs of gases and vapours was built and measurements to determine MIE of propane/air at normal atmospheric conditions were performed. However, the ASTM standard does not prescribe any statistical procedure for deriving MIE values from primary test data. We therefore adopted the "highest-possible-border-line" procedure proposed by Moorhouse et al. in 1974, and obtained a MIE of 0.48 mJ, which is very close to the 0.46 mJ found by these workers, as opposed to the classical Lewis and von Elbe value of only 0.25 mJ. One possible reason for the discrepancy could be the very low ignition probability of only 1% used by Lewis and von Elbe as their MIE criterion. However, when applying both linear and logistic regression analysis to our experimental data, the spark energies yielding 1% probability of ignition were found to be 0.40+/-0.06 and 0.45+/-0.08 mJ, respectively, which are both significantly higher than 0.25 mJ. This may indicate that the classical MIE values for gases and vapours published by Lewis and von Elbe (1961) are perhaps unnecessarily conservative.
建立了一份用于确定气体和蒸气的 MIE 的标准 ASTM 火花发生器，并进行了在正常大气条件下测量丙烷/空气的 MIE 的实验。然而，ASTM 标准并没有规定任何从原始测试数据中推导出 MIE 值的统计程序。因此，我们采用了 Moorhouse 等人在 1974 年提出的“尽可能高的边界线”程序，并得到了 0.48 mJ 的 MIE 值，这非常接近这些研究人员发现的 0.46 mJ，而不是经典的 Lewis 和 von Elbe 仅为 0.25 mJ 的值。差异的一个可能原因可能是 Lewis 和 von Elbe 作为他们的 MIE 标准所使用的点火概率非常低，只有 1%。然而，当我们对实验数据应用线性和逻辑回归分析时，发现产生 1%点火概率的火花能量分别为 0.40+/-0.06 和 0.45+/-0.08 mJ，这两个值都明显高于 0.25 mJ。这可能表明 Lewis 和 von Elbe（1961）发表的气体和蒸气的经典 MIE 值可能过于保守。
