Combined kinetics and iterative target transformation factor analysis for spectroscopic monitoring of reactions.

Obtaining rate constants and concentration profiles from spectroscopy is important in reaction monitoring. In this paper, we combined kinetic equations and Iterative Target Transformation Factor Analysis (ITTFA) to resolve spectroscopic data acquired during the course of a reaction. This approach is based on the fact that ITTFA needs a first guess (test vectors) of the parameters that will be estimated (target vectors). Three methods are compared. In the first, originally proposed by Furusjö and Danielsson, kinetic modelling is only used to provide the initial test vectors for ITTFA. In the second the rate constant used to provide the test vectors is optimised until a best fit is reached. In the third, a guess of the rate constant is used to provide the test vectors to ITTFA. The outcome of ITTFA is then used to fit the kinetic model and obtain a new guess of the rate constant. With this constant new concentration profiles are generated and provided to the ITTFA algorithm as new test vectors, in an iterative manner, minimising the residuals of the predicted dataset, until convergence. The second and third methods are new implementations of ITTFA and are compared to the first, established, method. First order (both one and two step) and second order reactions were simulated and instrumental noise was introduced. An experimental second order reaction was also employed to test the methods.

[1]  Edmund R. Malinowski,et al.  Theory of error for target factor analysis with applications to mass spectrometry and nuclear magnetic resonance spectrometry , 1978 .

[2]  Marcel Maeder,et al.  Target transform fitting: a new method for the non‐linear fitting of multivariate data with separable parameters , 2001 .

[3]  R. Brereton,et al.  Estimation of second order rate constants using chemometric methods with kinetic constraints. , 2002, The Analyst.

[4]  Miquel Esteban,et al.  Multivariate curve resolution with alternating least squares optimisation: a soft-modelling approach to metal complexation studies by voltammetric techniques , 2000 .

[5]  R. Bro,et al.  A fast non‐negativity‐constrained least squares algorithm , 1997 .

[6]  R. Brereton,et al.  Influence of Different Sources of Error on Estimated Kinetics Parameters for a Second Order Reaction , 2004 .

[7]  R. Brereton,et al.  Comparison of PLS and kinetic models for a second-order reaction as monitored using ultraviolet visible and mid-infrared spectroscopy. , 2006, Talanta.

[8]  H. Haario,et al.  COMBINING SOFT AND HARD MODELLING IN CHEMICAL KINETIC MODELS , 1998 .

[9]  Romà Tauler,et al.  Application of a novel resolution approach combining soft- and hard-modelling features to investigate temperature-dependent kinetic processes , 2001 .

[10]  R. Brereton,et al.  Monitoring of a second‐order reaction by electronic absorption spectroscopy using combined chemometric and kinetic models , 2003 .

[11]  Erik Furusjö,et al.  Target testing procedure for determining chemical kinetics from spectroscopic data with absorption shifts and baseline drift , 2000 .

[12]  Edmund R. Malinowski,et al.  Factor Analysis in Chemistry , 1980 .

[13]  Svante Wold,et al.  Modelling and diagnostics of batch processes and analogous kinetic experiments , 1998 .

[14]  Dominique Bonvin,et al.  On the Rank Deficiency and Rank Augmentation of the Spectral Measurement Matrix , 1996 .

[15]  Paul J. Gemperline,et al.  A priori estimates of the elution profiles of the pure components in overlapped liquid chromatography peaks using target factor analysis , 1984, J. Chem. Inf. Comput. Sci..

[16]  F. Wilkinson,et al.  Chemical kinetics and reaction mechanisms , 1980 .

[17]  Yi-Zeng Liang,et al.  Principles and methodologies in self-modeling curve resolution , 2004 .

[18]  Erik Furusjö,et al.  A method for the determination of reaction mechanisms and rate constants from two-way spectroscopic data , 1998 .

[19]  Zhongliang Zhu,et al.  Iterative target transformation factor analysis for the resolution of kinetic–spectral data with an unknown kinetic model , 2002 .

[20]  P. Gemperline,et al.  Advantages of soft versus hard constraints in self-modeling curve resolution problems. Alternating least squares with penalty functions. , 2003, Analytical chemistry.

[21]  Olof Svensson,et al.  Estimation of kinetic parameters from non-isothermal batch experiments monitored by in situ vibrational spectroscopy , 2003 .

[22]  G. Kateman,et al.  Multicomponent self-modelling curve resolution in high-performance liquid chromatography by iterative target transformation analysis , 1985 .

[23]  Richard G. Brereton,et al.  Chemometrics: Data Analysis for the Laboratory and Chemical Plant , 2003 .

[24]  E. Furusjö,et al.  Uncertainty in rate constants estimated from spectral data with baseline drift , 2000 .

[25]  Paul J. Gemperline,et al.  Evolving factor analysis applied to flow injection analysis data , 1989 .