Title of article :
Soft and harder multivariate modelling in developing the properties of titanium dioxide pigments
Author/Authors :
Taavitsainen، نويسنده , , Veli-Matti and Jalava، نويسنده , , Juho-Pertti، نويسنده ,
Issue Information :
دوفصلنامه با شماره پیاپی سال 1995
Abstract :
Two models have been developed for the determination of the titanium dioxide pigments properties. It is shown that it is possible to make a satisfactory empirical (partial least squares) model, the soft model, for determining the crystal size distribution by measuring the turbidity spectrum of a suitably prepared process sample. The method is quite insensitive to the degree of grinding in the sample preparation. The theoretically orientated or semi-empirical model, the harder model, gives more insight in understanding how the different size fractions affect the light scattering properties of the product. This model is not independent of the degree of grinding of the sample. Both models are useful, the soft model gives a robust method for crystal size distribution determination and the harder model provides understanding about the optical properties of TiO2 pigments.
Keywords :
Non linearity , Mechanistic modelling , Soft modelling , Regression , partial least squares , titanium dioxide pigments
Journal title :
Chemometrics and Intelligent Laboratory Systems
Journal title :
Chemometrics and Intelligent Laboratory Systems