Project description
A cost-effective approach to studying the vibrational spectra of molecular crystals
The computational study of molecular condensed phases represents a notoriously difficult problem. The increased computational cost associated with modelling excited states is compounded by the need for larger system sizes in these materials. The EU-funded VibMolCryst project will use the hybrid quantum mechanics/molecular mechanics (QM/MM) approach to study the vibrational spectra and properties of molecular crystals. The method combines the accuracy and speed of the QM and MM approaches, respectively. This computationally affordable method is expected to aid the interpretation of low-frequency terahertz spectra – used, for instance, in the detection of explosives. What is more, it could increase the accuracy of calculated thermodynamical stabilities, which is critical for drug development.
Objective
By this project, the routine calculation of anharmonic vibrational spectra and properties for practically relevant molecular crystals will be enabled via the usage of a quantum-mechanical (QM:QM) embedding approach. All monomers and relevant dimers are treated with a high-level method, while the fully periodic system is considered at a lower level. Highly accurate vibrational spectra can be obtained for small molecular systems with benchmark CCSD(T) utilizing second-order vibrational perturbation theory (VPT2) only with a computational cost prohibitive for routine applications involving larger systems. Therefore, the applicant will create a diverse benchmark set of monomers and molecular dimers covering a wide range of intermolecular interactions and subsequently benchmark the performance of various dispersion-inclusive density functional approximations (DFA) against CCSD(T) for vibrational properties calculated with VPT2, independent Morse oscillators, and the harmonic approximation. Next, the QM:QM embedding approach for molecular crystals will be extended from available gradients to the calculation of harmonic vibrational spectra, which will already enable the usage of hybrid DFAs at a cost comparable to the generalized-gradient approximation. Subsequently, VPT2 calculations for monomers and dimers will be incorporated in the embedding scheme and the accuracy of the so obtained anharmonic vibrational spectra will be assessed for a variety of molecular crystals using promising DFAs identified during the first stage of the project. This methodology will be computationally affordable for practically relevant molecular crystals and is expected to aid peak assignments and interpretation of low-frequency THz spectra—used for instance for the detection of explosives. This approach is also expected to increase the accuracy of calculated thermodynamical stabilities, which is critical for drug development since existing molecular crystal polymorphs are almost degenerate.
Fields of science
Programme(s)
Funding Scheme
MSCA-IF - Marie Skłodowska-Curie Individual Fellowships (IF)Coordinator
8010 Graz
Austria