Modelling Vibrational Sum Frequency Generation Spectra of Interfacial Proteins

Research output: Book/anthology/dissertation/reportPh.D. thesis

8 Downloads (Pure)


Interfaces are of great importance to the function of chemical and biological systems.
Proteins are a highly diverse group of molecules that are found throughout nature to control biological processes.
Proteins show a high degree of structural flexibility, and unique interactions
in an interfacial environment. They are able to alter their structure thereby enabling or disabling the functions of proteins.
However, the flexibility that enables proteins to suit all the various needs of biological systems also makes them incredibly complex with many degrees of freedom.
The interfacial protein structures of interest are found in the thin monomolecular layers and the interfacial signals are often obscured by the surrounding bulk media.
Therefore, studying interfacial biological processes requires highly surface specific techniques, which can probe the structural information of proteins in their native interfacial environment.

Sum frequency generation spectroscopy is a nonlinear spectroscopic technique with very high surface specificity. The technique allows measurement of vibrational spectra from interfacial proteins. These spectra contain information about the molecular structures and their orientation at the interface, but it is complicated to interpret the experimental spectra and extract the molecular information.
In this thesis, I present how sum frequency generation spectra can be interpreted via molecular dynamic simulations.
The interpretation relies on calculation of the spectra, which is possible with two fundamentally different approaches, called ``the frequency mapping model'' and ``the time correlation model'', respectively.
These models can be used complementarily due to their various advantages and disadvantages.
The key difference between the two approaches is that the frequency mapping model uses parameters derived from static protein structures to calculate the vibrational frequencies while the time correlation model relies on sampling of the time dependent dynamic motion in molecular simulations, calculate the appropriate time correlation function and Fourier transform it into the frequency spectra.

The focus of this thesis is to describe the theoretical background for the models and show how they allow the study of protein structure at interfaces. Applications to several relevant proteins at various water based interfaces are presented and discussed. Furthermore, the technique is used to study the local water environment around an interfacial protein.
Original languageDanish
PublisherAarhus University
Number of pages199
Publication statusPublished - 13 Feb 2024

Cite this