The aim of this project is to use FTIR spectroscopy, Quantum Mechanical (QM) calculations and Machine Learning (ML) methods to understand the changing geometry of molecules in mixtures due to specific interactions. A simple starting system of water and acetone causes a shift of the carbonyl peak at around 1700cm-1 with increasing water content and hydrogen bond strength.
In the case of the water acetone system, QM methods are used to sample geometries of water at distances of between 2-4 angstroms from acetone. This provides features including frequency data, dipole moment, reduced mass, intensity, energy, geometry, coulomb matrices to be used with ML models. With the inclusion of experimental FTIR data of the acetone mixtures as the target, the geometry of the acetone water complex can be reverse engineered.
FTIR blue and red shifts,
Dilution of acetone with either water or carbon tetrachloride shifts the carbonyl band red or blue, respectively
https://assets.thermofisher.com/TFS-Assets/CAD/Application-Notes/AN50733_E.pdf
"The strong bathochromic shifts observed on methanol OH and acetone CO stretch IR bands are related to hydrogen bonds between these groups. Factor analysis separates the spectra into four acetone and four methanol principal factors." https://doi.org/10.1063/1.1790431
"Analysis of IR spectra of ethylene glycol shows that thereare only a few contributing bands with solidly fixed vibrationalfrequencies, which only change in relative intensity whentemperature is changed. It did not show any clear evidence ofan intrinsic frequency shift indicating the gradual weakening ofhydrogen bonding interaction. Only the relative population ofspecies, e.g., strongly bonded and dissociated or much moreweakly bonded groups, seems to be changing. IR spectra ofacetone in a mixed solvent of CHCl3/CCl4with varyingcomposition also show that intrinsic IR frequency does notshift appreciably. Instead, only the relative contributions ofhighly overlapped adjacent bands are changing." https://doi.org/10.1366/000370210792434396
https://cs229.stanford.edu/proj2017/final-reports/5244394.pdf