Unsupervised Screening of Vibrational Spectra by Principal Component Analysis for Identifying Molecular Clusters

Abstract

Vibrational spectra are commonly used to study molecular interactions in solutions. However, the data analysis is often demanding and requires significant experience in order to obtain meaningful results. This study demonstrates that principal component analysis (PCA) can serve as an unsupervised tool for initial screening of non-ideal mixture systems. Taking the aqueous solutions of dimethyl sulfoxide (DMSO) as an example, PCA reveals-easily and fast-the two prominent stoichiometries at 1:2 and 1:1 molar DMSO:water ratio and significantly outperforms elaborate spectral profile analysis or common algorithms as indirect hard modeling (IHM) or multivariate curve resolution (MCR). The corresponding molecular 1:1 and 1:2 clusters are known to be dominating configurations in the solutions.

Ion Pairs of 1-Butyl-3-Methylimidazolium Triflate Do Not Dissociate in Propan-1-ol: A Vibrational Spectroscopic Viewpoint

“…It shows a clear minimum around an alcohol mole fraction of x(PrOH) = 0.66. In the previous study [27], it was shown that such extreme values are likely pointing towards stoichiometries, at which interesting phenomena can be expected. In the present case, the mole fraction of x(PrOH) = 0.66 translates to a situation in which the solution contains two alcohol molecules per ion pair.…”

“…In the second step, the data were screened in order to identify interesting compositional regimes, for instance where ion pair dissociation likely takes place. For this purpose, principal component analysis (PCA) was applied to the data sets as described in reference [27]. Figure 2a shows the cumulative variance of the first five principal components (PCs) of the FTIR data.…”

Quantitative vibrational spectroscopy on liquid mixtures: concentration units matter

“…Similar to these results, spectroscopic non-ideality is the rule rather than the exception for reported vibrational studies on liquid mixtures, even in seemingly ideal systems such as methanol-ethanol 18,19 and binary aliphatic or aromatic hydrocarbon systems (e.g., benzene-toluene and n-hexane-nheptane), 20 and PCA has recently been proposed as a powerful tool for unsupervised screening of non-ideal mixtures. 21 Spectral non-ideality results are generally interpreted as providing evidence for the existence of microheterogeneity (i.e., homo-and heteroclusters) at the molecular level in mixtures. [18][19][20] Excess spectroscopy 13 is another powerful tool to reveal non-ideal features in spectroscopic data on liquid mixtures.…”

“…20 The ER values correlate fairly well with the variance explained by PC2 (R 2 = 0.55) as obtained from the PCA results, confirming the report by Kiefer and Eisen that analysis of higher PCs can be used as screening tool for non-ideal mixture systems. 21 The intensity of excess spectra, from which ER values are calculated, depends strongly on the magnitude of peak shifts as a function of composition. For example, the strong anti-symmetric CCl 3 stretching band at 744 cm −1 in the MIR spectrum of chloroform shifts to 758 cm −1 upon mixing with toluene (ESI Fig.…”

Vibrational Spectroscopy as a Promising Toolbox for Analyzing Functionalized Ceramic Membranes

“…Principal component analysis is an unsupervised chemometric tool 37,38 that is commonly applied to spectroscopic data. 39 Principal component analysis builds a model for a data matrix X aiming at the most meaningful representation of the data set. This is obtained by reducing the dimension of the data matrix X and thereby extracting only the relevant information.…”