Arctangent normalization and principal-component analyses merge method to classify characteristics utilizing time-dependent material data.

Abstract

We propose a technique for classifying paints with time-dependent properties using a new method of merging principal-component analyses (the "PCA-merge" method) that utilizes shifting of the barycenter of the PCA score plot. To understand the molecular structure, elemental concentrations, and the concentrations in the evolved gaseous component of various paints, we performed comprehensive characterizations using Fourier transform infrared spectroscopy, inductively coupled plasma mass spectrometry, and head-space-gas chromatograph/mass spectrometry while drying the paint films for 1-48 h. As various detected intensity- and time-axis variables have different dimensions that cannot be handled equally, we normalized those data as an angle parameter (θ) using arctangent to reduce the influence of high/low intensity data and the various analytical instrument. We could classify the paints into suitable categories by applying multivariate analysis to this arctangent-normalized data set. In addition, we developed a new PCA-merge method to analyze data groups that include different time components. This method merges the PCA data groups of each time-component axis into that of specific-component axes and distinguishes each sample by utilizing the shift in the barycenter of the PCA score plot. The proposed method enables the simultaneous utilization of various data groups that contain information about static and dynamic properties. This provides further insight into the characteristics of the paint materials via shifts in the barycenter of the PCA scores without requiring numerous peak identifications.