6533b821fe1ef96bd127b691
RESEARCH PRODUCT
Transition paths between phases IV, III and II of ammonium nitrate predicted from X-ray powder diffractometer and differential scanning calorimeter data by partial least-squares regression
Jussi ValkonenMauno E. E. HarjuPentti Minkkinensubject
Phase transitionComponent (thermodynamics)Process Chemistry and TechnologyAmmonium nitrateAnalytical chemistryComputer Science ApplicationsAnalytical Chemistrychemistry.chemical_compoundDifferential scanning calorimetrychemistryPowder DiffractometerPartial least squares regressionPrincipal component analysisSpectroscopySoftwarePowder diffractiondescription
Abstract Ammonium nitrate solid phase transition paths between phases IV, III and II were explained and predicted on the basis of X-ray powder diffraction (XRD) and differential scanning calorimetry (DSC) data by applying partial least-squares regression (PLS) and principal component analysis (PCA). The samples were clustered according to their different transition paths with the PLS and PCA models, and the transition paths were predicted with PLS component clusters. The best PLS clusters were formed by a few first components. Prediction of the transition path with the PLS clusters made a semiquantitative prediction of the transition energy possible. In PCA, principal components 6 and 11, which best clustered the samples, explained 7% of the variation of the XRD data. The clustering showed that most of the variance is due to other factors than those affecting the transition path. The correlation between the structural data and the transition energies obtained by differential scanning calorimetry was calculated by PLS method. The correlation between the predicted and the observed energies of an independent predictor set varied from zero to over 0.7 depending on the 2θ range. The PLS analysis of the XRD and DSC measurements proved that the different phase transition paths are of structural origin.
year | journal | country | edition | language |
---|---|---|---|---|
1994-05-01 | Chemometrics and Intelligent Laboratory Systems |