Sample Size for Steel Analysis

In low-alloy steel analysis by ICP, Ernie wanted to know what would be considered an adequate sample size in grams. He noted that the samples would be dissolved in acid under pressure in a microwave oven.

	RE: Sample Size for Steel Analysis Dear Ernie,      Steels would be termed 'Compact Solids' and the mode of sampling is the first issue to be addressed, making the sample size issue more apparent. For compact solids, random sampling is suggested (random drilling where any portion of the bulk material has the same probability of being included in the sample - all of the drillings are collected). Typically it is difficult to keep the sample size smaller than 1 gram. If the sample size collected is more on the order of 50 to 500 grams (100 - 500 grams is a better situation for steels), the drillings can be mixed, pilled, and separated into quadrants and opposite quadrants taken (sample size reduction) and repeated until sample size reduction to a workable size (i.e., ~ 10 grams is achieved). Multiple 1 gram samples can then be taken (sub-sampling) from the reduced sample size for analysis.      Therefore, the first and biggest issue is whether a representative sample was taken? Secondly, is the acceptable reduction of the sample down to the lowest workable representative 'sub-sample' size. With respect to sub-sampling errors, working with small (~0.1 gram) sample sizes for alloys such as steel is much easier for microwave digestion but has an increased risk of providing results that do not represent the sample as a whole. The above link can help you make that determination. Please feel free to call with any questions     Paul R. Gaines, Ph.D. Serving you in chemistry