Skip to content
BY-NC-ND 3.0 license Open Access Published by De Gruyter Open Access January 23, 2013

Selection of mineralised methods to analyse different types of matrices. Applying the Box-Cox transformation to chemometrics study the coexistence of heavy metals in natural samples

Marcin Sajdak and Celina Pieszko
From the journal Open Chemistry

Abstract

Chemometric methods are mostly used to optimise technological processes and analytical procedures. Applying chemometric methods in environmental tests may reveal relationships among chemical elements in biomes. Cluster analysis and principal component analysis (PCA) are very helpful for detecting relationships among studied parameters. However, large amounts of data may have a negative effect on this analysis and can lead to misinterpretation of the results. This situation was observed when the samples, taken from several places in the Silesian Province, were used to test the relationship between heavy metals contained in various environmental matrices. Samples were collected from a small area and were characterised by a single biome (pine forest) because direct interpretation of PCA and CA was insufficient to correctly describe such data. The solution to this problem was the use of the Box-Cox transformation, which is a rapid method to normalise input data. The application of chemometric tools enabled the relationships between sampling sites (industrialised and non-industrialised) to be examined and was very helpful in illustrating the relationship between the methodologies of plant preparation samples. Furthermore, the results may indicate the need for further data analysis. The tools described in this paper can be useful for choosing the optimal mineralisation method according to the type of test matrix.

[1] M. Żelazny, A. Astel, A. Wolanina, S. Małek, Environmental Pollution 159, 1048 (2011) http://dx.doi.org/10.1016/j.envpol.2010.11.02110.1016/j.envpol.2010.11.021Search in Google Scholar PubMed

[2] T. Togkalidou, H. Tung, Y. Sun, A. Andrews, R.D. Braatz, Organic Process Research & Development 6, 317 (2002) http://dx.doi.org/10.1021/op015516x10.1021/op015516xSearch in Google Scholar

[3] G.E.P. Box, D.R. Cox, J. Roy. Statist. Soc. B 26, 211 (1964) 10.1111/j.2517-6161.1964.tb00553.xSearch in Google Scholar

[4] STATISTICA — data analysis software system, version 9 (StatSoft, Inc. STATISTICA — data analysis software system, version 9 (StatSoft, Inc., USA, 2009) Search in Google Scholar

[5] M. Sajdak, C. Pieszko, Eur. J. Chem. 10(5), 1696 (2012) 10.2478/s11532-012-0096-0Search in Google Scholar

[6] M. Sajdak, Eur. J. Chem. 11(2), 151 (2013) 10.2478/s11532-012-0145-8Search in Google Scholar

[7] M. Sajdak, O. Piotrowski, Eur. J. Chem. 11(2), 259 (2013) 10.2478/s11532-012-0147-6Search in Google Scholar

[8] M. Tobiszewski, J. Namieśnik, Anal. Bioanal. Chem. 399, 3565 (2011) http://dx.doi.org/10.1007/s00216-011-4676-110.1007/s00216-011-4676-1Search in Google Scholar PubMed PubMed Central

[9] P. McCullagh, J.A Nelder, Generalized Linear Models, 2nd edition (Chapman & Hall, London, 1989) 10.1007/978-1-4899-3242-6Search in Google Scholar

[10] G. Romanik, E. Gilgenast, A. Przyjazny, M. Kamiński, Dear Sirce 70, 253 (2007) 10.1016/j.jbbm.2006.09.012Search in Google Scholar PubMed

[11] P. Konieczka, J. Namieśnik, Chem. Anal. (Warsaw) 53, 785 (2008) Search in Google Scholar

[12] C.H. Lochmuller (Ed.), Chemical Laboratory — a Practical Approach (CRC/Taylor & Francis, Boca Raton, USA, 2009) 131–216 Search in Google Scholar

Published Online: 2013-1-23
Published in Print: 2013-4-1

© 2013 Versita Warsaw

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.

Scroll Up Arrow