Analysis of variance in forensic glass analysis by ICP-MS: variance within the method

Douglas C. Duckworth; Charles K. Bayne; Shelby J. Morton; José Almirall

doi:10.1039/A908813J

Douglas C. Duckworth,*^a Charles K. Bayne,^b Shelby J. Morton^a and José Almirall^c

* Corresponding authors

^a Chemical and Analytical Sciences Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA
E-mail: duckworthdc@ornl.gov

^b Computer Science and Mathematics Division, Oak Ridge National Laboratory, Oak Ridge, TN, USA

^c International Forensic Research Institute, Florida International University, Department of Chemistry, University Park, Miami, FL, USA

Abstract

An analysis of variance (ANOVA) was performed on inductively coupled plasma mass spectrometric analyses of glass fragments to identify the source and magnitude of variation in elemental concentration determinations. Accuracy in the multielement analysis was also evaluated. ANOVA was performed on 68 isotopes representing 57 elements in NIST SRM 612 Trace Elements in Glass (nominal concentration 50 mg L⁻¹). The analysis apportioned variance into components of time (day-to-day), dissolution (sample-to-sample), calibration and instrumental (internal) variance. Elements were classified into four categories defined by the combination of precision and accuracy. Those not falling into the classification of both good precision (RSD < 10% or s ≤ 3 µg g⁻¹) and accuracy (|bias| < 5 µg g⁻¹ and |%bias| ≤ 10%) were critically examined before inclusion in a list of potentially discriminating elements to be used for the development of an elemental concentration database for float glasses. The final list consists of 59 isotopes, representing 46 elements.

Article information

https://doi.org/10.1039/A908813J

Article type

Paper

Submitted

04 Nov 1999

Accepted

18 Apr 2000

First published

09 Jun 2000

Download Citation

J. Anal. At. Spectrom., 2000,15, 821-828

Permissions

Request permissions