Previous Research and Teaching
My previous (before 2005) research and teaching has been at
University of Natal, South Africa(UND) Wayne State University, USA (WSU)
Goteborgs University and Chalmers (GU) University of Boras (UB)
Postdoctoral Research (WSU, 1996-1997) Molecular dynamic simulations of polyatomic systems are usually based on analytic potential energy surfaces which reproduce selected ab initio and/or experimental data. Development of these surfaces, which is often the most time consuming and problematic part of the dynamics investigation, can be avoided by calculating the internuclear forces directly from electronic structure theory at each step along the trajectory. This is known as the direct dynamics method. We have employed this technique, using both semiempirical and ab initio electronic structure theories, to study the dynamics and decay of the trimethylene biradical, cyclopropane stereomutation and decomposition of the C2H4F collision complex.
In addition to studying chemical reaction mechanisms and unimolecular decay, we developed the direct dynamics code to improve the accuracy and efficiency of the integration procedure. The major drawback of this technique (when compared to simulations based on an analytic surface) is the large amount of computer time required to calculate energies, forces and Hessians (if required). Reducing the time for each calculation or the number of calculations per trajectory, without affecting the integration accuracy, will make this method more tractable for larger molecules and longer integration times.
Another project, which was done (in collaboration with Ford Motor Company) involveed the simulation of alkanes on aluminum oxide surfaces. The Cerius II software was used to investigate possible inaccuracies that may arise when employing a united atom model to simulate chemical systems of this type.
PhD Research (GU, 1992-1995) Classical molecular dynamics techniques were employed to investigate intramolecular vibrational energy redistribution and unimolecular reaction rates. This allowed us to assess the validity of certain assumptions that underly reaction rate theories . The initial work focused on the integration methods that are used in the simulation of physical systems. We provided evidence that the Gauss-Radau technique - which has seen limited use in chemical dynamics simulations - offers improved accuracy and efficiency over most other integration techniques.
Simulations of one dimensional chain systems under pressure exemplified the important role of non-linear resonances in intramolecular energy redistribution. Results obtained for the one dimensional monatomic chain under stress indicated that the RRKM predicted rate of chain fission is greatly improved when anharmonic effects are taken into account.
Classical trajectory simulations that were based on a model trans-stilbene potential energy surface revealed quasiperiodic energy flow between the optically active vibrational mode (mode 25) and a coupled mode (mode 49). These modes are substantially decoupled from the rest of the molecule thereby retarding energy flow to the reaction coordinate. The resulting bottleneck and beating patterns are robust to an increase in molecular energy but not to collisions by a buffer gas. The presence of this bottleneck is consistent with available experimental data.
MSc Research (UND 1990-1991) It has previously been postulated that thymine dimer formation, which has been linked to the formation of skin cancer, can occur via the photosensitisation of certain sunscreen constituents such as Uvinul DS49. The photoreaction of free thymine base with Uvinul DS49 at 313nm was investigated and the results showed that the dimer was formed (i.e., sunscreens can cause skin cancer). This work was extended to include the sensitised photoreaction of DNA in vitro and in vivo using E. Coli as the host cell. Experimental work incorporated high pressure liquid chromatography, low pressure liquid chromatography, ferrioxalate actinometry, NMR, IR and GC-MS analysis in addition to various biological techniques required for culturing E. Coli and the subsequent DNA extraction and hydrolysis.
Teaching
Sweden (UB): I have taught and developed a number of courses in chemistry and chemical engineering. These include
General Chemistry,
Materials Chemistry,
Solid State Chemistry,
Physical Chemistry,
Colloid and Surface Chemistry and
Simulation in Chemical Engineering.
South Africa (UND) I taught a first year course in Physical Chemistry (1991)
South Africa (for underprivileged scholars) General science, Pre-university chemistry