Influence of Raman spectra measurement conditions on the dating results of writing compositions

When establishing the age of a document using spectral methods, the result is influenced not only by the storage conditions, but also by the methods of measuring the spectra and mathematical processing of the results. This work analyzes the influence the modes of measuring Raman spectra have on the result of measuring the writing composition spec-tra. The analysis area includes the correlation coefficients of the full spectral matrices as a criterion for the reliability of the determination date. The influence of focusing accuracy, correctness of subtraction of the baseline, sample irradiation time, and varying the number of repetitions is investigated. In the study, chemometric tools for the analysis of Raman spectra were also applied. The influence of mathematical spectra processing shows that the components PC4, P5, and PC6 are most closely related to the shelf life, although they do not make the maximum contribution to the variance in the system. Taking into account all the requirements for measuring spectra, the PCA method applied to the Raman spectra of writing compositions allows clustering samples by manufacturer’s brands and by the sample shelf life.When establishing the age of a document using spectral methods, the result is influenced not only by the storage conditions, but also by the methods of measuring the spectra and mathematical processing of the results. This work analyzes the influence the modes of measuring Raman spectra have on the result of measuring the writing composition spec-tra. The analysis area includes the correlation coefficients of the full spectral matrices as a criterion for the reliability of the determination date. The influence of focusing accuracy, correctness of subtraction of the baseline, sample irradiation time, and varying the number of repetitions is investigated. In the study, chemometric tools for the analysis of Raman spectra were also applied. The influence of mathematical spectra processing shows that the components PC4, P5, and PC6 are most closely related to the shelf life, although they do not make the maximum contribution to the variance in the system. Taking into account all the requirements for measuring spectra, the PCA method applied to the Raman spectra of writing compositions allows clustering samples by manufacturer’s brands and by the sample shelf life