Based on the original position statistic distribution analysis technique, the characterization method of segregation for large-size metal materials gives significant guidance to the research of material properties and production. However, random errors are inevitably brought into the calculation of segregation degree for materials characterization by the Spark Mapping Analysis for Large Sample (SMALS) technique, resulting in a misguide of the segregation degree. In this paper, we present the lower limit of segregation degree (Ds(L)) method to distinguish the random error from metal material segregation for large-size samples over the SMALS method. The random error of standard material in the 95% confidence interval was utilized as Ds(L) and the method has been applied for macro-segregation quantitative analysis. The precision correlation between Spark Atomic Emission Spectrometry (Spark-AES) and SMALS was established. Furthermore, the functional relationship between the Ds(L) and element content C can be obtained in the SMALS method. The Ds(L) method as the criterion can be used to not only characterize the minimum limit of the segregation degree but also the segregation existence for large-size samples. Applying to low-alloy steel can illustrate the effective performance of the Ds(L) method. Results on both spark mapping analysis and Spark-AES verify the substantial consistency.
Read full abstract