Source Attribution of Cyanides using Anionic Impurity Profiling, Stable Isotope Ratios, Trace Elemental Analysis and Chemometrics
Abstract
Chemical attribution signatures (CAS) for chemical threat agents (CTAs) are being investigated to provide an evidentiary link between CTAs and specific sources to support criminal investigations and prosecutions. In a previous study, anionic impurity profiles developed using high performance ion chromatography (HPIC) were demonstrated as CAS for matching samples from eight potassium cyanide (KCN) stocks to their reported countries of origin. Herein, a larger number of solid KCN stocks (n = 13) and, for the first time, solid sodium cyanide (NaCN) stocks (n = 15) were examined to determine what additional sourcing information can be obtained through anion, carbon stable isotope, and elemental analyses of cyanide stocks by HPIC, isotope ratio mass spectrometry (IRMS), and inductively coupled plasma optical emission spectroscopy (ICP-OES), respectively. The HPIC anion data was evaluated using the variable selection methods of Fisher-ratio (F-ratio), interval partial least squares (iPLS), and genetic algorithm-based partial least squares (GAPLS) and the classification methods of partial least squares discriminate analysis (PLSDA), K nearest neighbors (KNN), and support vector machines discriminate analysis (SVMDA). In summary, hierarchical cluster analysis (HCA) of anion impurity profiles from multiple cyanide stocks from six reported country of origins resulted in cyanide samples clustering into three groups: Czechmore »
- Authors:
- Publication Date:
- Research Org.:
- Pacific Northwest National Lab. (PNNL), Richland, WA (United States)
- Sponsoring Org.:
- USDOE
- OSTI Identifier:
- 1239476
- Report Number(s):
- PNNL-SA-113038
Journal ID: ISSN 0003-2700; 400904120
- DOE Contract Number:
- AC05-76RL01830
- Resource Type:
- Journal Article
- Journal Name:
- Analytical Chemistry
- Additional Journal Information:
- Journal Volume: 88; Journal Issue: 3; Journal ID: ISSN 0003-2700
- Publisher:
- American Chemical Society (ACS)
- Country of Publication:
- United States
- Language:
- English
- Subject:
- chemical forensics
Citation Formats
Mirjankar, Nikhil S., Fraga, Carlos G., Carman, April J., and Moran, James J. Source Attribution of Cyanides using Anionic Impurity Profiling, Stable Isotope Ratios, Trace Elemental Analysis and Chemometrics. United States: N. p., 2016.
Web. doi:10.1021/acs.analchem.5b04126.
Mirjankar, Nikhil S., Fraga, Carlos G., Carman, April J., & Moran, James J. Source Attribution of Cyanides using Anionic Impurity Profiling, Stable Isotope Ratios, Trace Elemental Analysis and Chemometrics. United States. https://doi.org/10.1021/acs.analchem.5b04126
Mirjankar, Nikhil S., Fraga, Carlos G., Carman, April J., and Moran, James J. 2016.
"Source Attribution of Cyanides using Anionic Impurity Profiling, Stable Isotope Ratios, Trace Elemental Analysis and Chemometrics". United States. https://doi.org/10.1021/acs.analchem.5b04126.
@article{osti_1239476,
title = {Source Attribution of Cyanides using Anionic Impurity Profiling, Stable Isotope Ratios, Trace Elemental Analysis and Chemometrics},
author = {Mirjankar, Nikhil S. and Fraga, Carlos G. and Carman, April J. and Moran, James J.},
abstractNote = {Chemical attribution signatures (CAS) for chemical threat agents (CTAs) are being investigated to provide an evidentiary link between CTAs and specific sources to support criminal investigations and prosecutions. In a previous study, anionic impurity profiles developed using high performance ion chromatography (HPIC) were demonstrated as CAS for matching samples from eight potassium cyanide (KCN) stocks to their reported countries of origin. Herein, a larger number of solid KCN stocks (n = 13) and, for the first time, solid sodium cyanide (NaCN) stocks (n = 15) were examined to determine what additional sourcing information can be obtained through anion, carbon stable isotope, and elemental analyses of cyanide stocks by HPIC, isotope ratio mass spectrometry (IRMS), and inductively coupled plasma optical emission spectroscopy (ICP-OES), respectively. The HPIC anion data was evaluated using the variable selection methods of Fisher-ratio (F-ratio), interval partial least squares (iPLS), and genetic algorithm-based partial least squares (GAPLS) and the classification methods of partial least squares discriminate analysis (PLSDA), K nearest neighbors (KNN), and support vector machines discriminate analysis (SVMDA). In summary, hierarchical cluster analysis (HCA) of anion impurity profiles from multiple cyanide stocks from six reported country of origins resulted in cyanide samples clustering into three groups: Czech Republic, Germany, and United States, independent of the associated alkali metal (K or Na). The three country groups were independently corroborated by HCA of cyanide elemental profiles and corresponded to countries with known solid cyanide factories. Both the anion and elemental CAS are believed to originate from the aqueous alkali hydroxides used in cyanide manufacture. Carbon stable isotope measurements resulted in two clusters: Germany and United States (the single Czech stock grouped with United States stocks). The carbon isotope CAS is believed to originate from the carbon source and process used to make the HCN utilized in cyanide synthesis. Classification errors for two validation studies using anion impurity profiles collected over five years on different instruments were as low as zero for KNN and SVMDA, demonstrating the excellent reliability (so far) of using anion impurities for matching a cyanide sample to its country of manufacture (i.e., factory). Variable selection reduced errors for those classification methods having errors greater than zero with iPLS-forward selection, and F-ratio typically providing the lowest errors. Finally, using anion profiles to match cyanides to a specific stock or stock group resulted in cross-validation errors ranging from zero to 5.3%.},
doi = {10.1021/acs.analchem.5b04126},
url = {https://www.osti.gov/biblio/1239476},
journal = {Analytical Chemistry},
issn = {0003-2700},
number = 3,
volume = 88,
place = {United States},
year = {Fri Jan 08 00:00:00 EST 2016},
month = {Fri Jan 08 00:00:00 EST 2016}
}