Heterospectral Principal Component Analysis Two-Dimensional Correlation Spectrosocpy as a Robust Method for Determining Surface Adsorption Preference

Vibrational sum-frequency spectra encode information on the orientation and population of molecules at surfaces. Although orientation information can be extracted by simultaneously analyzing spectra acquired in two different beam polarization combinations, it is more difficult to isolate information on the surface population of adsorbed species when their orientation also changes. This is critical for the widespread adoption of this technique, as many systems are known to change their adsorbed structure in response to variation of the bulk concentration. Here we demonstrate that reconstructed sum-frequency spectra using the first principal component from a principal component analysis (PCA) markedly diminishes the spectral contributions due to molecular orientation. We show that the reconstructed spectra are excellent candidates for a subsequent surface-bulk heterospectral analysis to reveal the preference for a particular species to be adsorbed from a binary mixture.