Aplicação de Análise Multivariada de dados a Espectroscopia de Alta Resolução em Sólidos por Ressonância Magnética Nuclear. / Multivariate data analysis applied to high resolution spectroscopy in solids by nuclear magnetic resonance.

AUTOR(ES)
DATA DE PUBLICAÇÃO

1997

RESUMO

High Resolution Solid-State 13C-NMR Spectroscopy techniques Cross Polarization, Decoupling and Magic Angle Spinning, were employed in this work for the study of the chemical composition of seeds and food. A Multivariate Analysis procedure was also employed in the development of a calibration and prediction method for the determination of the components content based on a matrix multiplication. Singular Value Decomposition was carried on the 13C-NMRspectra matrix followed by Multiple Linear Regression on the components content matrix with the purpose of producing a model that relates the spectra to the sample components content determined by referee methods. When the resulting model is then applied to a new sample, assuming that the correlation found between the calibrations set matrices also exists in this sample, it gives the components content values. The protein and starch content were analyzed.

ASSUNTO(S)

multivariate analysis análise multivariada nuclear magnetic resonance sólidos ressonância magnética nuclear solids

ACESSO AO ARTIGO

Documentos Relacionados