The Chemical Educator
ISSN: 1430-4171 (electronic version)
Abstract Volume 8
Issue 3 (2003) pp 187-191
Spectrophotometric Analysis of Mixtures by Classical Least-Squares Calibration: An Advanced Experiment Introducing MATLAB
David González Gómez,† Arsenio Muñoz de la Peña,† Anunciación Espinosa Mansilla† and Alejandro César Olivieri*,‡
Departamento de Química Analítica, Universidad de Extremadura,
06071, Badajoz, Spain;† Departamento de Química Analítica,
Facultad de Ciencias Bioquímicas y Farmacéuticas, Universidad
Nacional de Rosario, Suipacha 531, Rosario (S2002LRK), Argentina,‡ email@example.com
Published online: 20 May 2003
Abstract. An advanced experiment in analytical spectroscopy and chemometrics is described. It involves calibration and prediction of the concentration of two colorants (cresol red and bromophenol blue in aqueous binary mixtures) as a model system for classical least-squares regression analysis. Precision estimates in the calculated concentrations are assessed using (1) spectral replicate measurements and (2) theoretical error-propagation equations. The experiment allows the introduction of the most simple of the multivariate calibration techniques to advanced chemistry students. The use of spectrophotometric data simplifies the experiments because students should be very familiar with spectroscopic data acquisition from previous laboratory courses. The power of calibration using a full-spectrum method is demonstrated in this particular case, in which a simple binary mixture of components is assayed, and it makes the potential of using chemometrics to resolve common analytical problems evident to the students. All calculations are performed with suitable routines written for the matrix environment MATLAB. The use of MATLAB allows the calculation and resolution of complex matrix equations with very simple programming commands This is valuable for the popularization of using advanced chemometric techniques in laboratory practices.
Key Words: Laboratories and Demonstrations; analytic chemistry; instrumental analysis
(*) Corresponding author. (E-mail: firstname.lastname@example.org)
Supporting Materials:A listing of the MATLAB code (cls.m) for CLS calibration and prediction is available in a Zip file (62 KB) 10.1333/s00897000690a.
Issue date: June