Unimolecular Decomposition

OBJECTIVE

Elucidate the nature of some unimolecular decomposition reaction types using 3D visualization of the electron density and related molecular properties for a system undergoing unimolecular decomposition.

METHODS

Molecular System Specification

N,N-Dimethly, N-Ethyl Amine Oxide was chosen due to the author's previous experience and familiarity with the Cope Elimination of amine oxides.

Electron Density and Molecular Properties Calculation

Gaussian 94W [1] was used to calculate the electron density, decomposition reaction path, and molecular properties for the system of interest.

There are 25 doubly occupied molecular orbitals for this system. Six of those are core orbitals and are not presented here. Orbital 26 (the Lowest Unoccupied Molecular Orbital (LUMO)) is also displayed.

Conventional Hartree Fock methods followed by a Moller-Plesset correlation energy correction (truncated at second order) were used to determine the wave function and from that, the density, reaction path and other molecular properties. A 6-31G(d) atomic orbital basis set was used with the correlation orbital basis spanning all 112 molecular orbitals (less the 6 core orbitals).

All molecular geometries (ground state and products) were optimized using the Berny algorithm using redundant internal coordinates. The transition state geometry was determined using the Synchronous Transit_Guided Quasi-Newton (STQN) method. The reaction path was calculated using the IRC method.

3D Visualization

3D Visualization was accomplished via the evaluation of the electron density, molecular orbitals, electrostatic potential, and the laplacian of the electron density over a 3D grid (cube) of points and subsequent generation of isosurfaces from each cube. The isosurfaces were computed using the Marching Cubes algorithm calculated using a modified version of the ISOSURF code of T.J. O'Donnell.

Triangle decimation was performed on all isosurfaces using Mike Garland's QSLIM to reduce the number of triangle primitives in each isosurface to a manageable size while retaining a faithful representation of the 3D geometry of each isosurface.

The POV Ray ray tracing program was used to render 3D images of the various isosurfaces. Conversion and processing of the rendered images into browser viewable gif and jpeg files was performed using Jasc's Paint Shop Pro.

VRML 97 representations of the isosurfaces were created using custom code generated by the author and are viewable with Cortona VRML Viewer (or any other VRML 97 compliant viewer). Note: all VRML entities on this site are VRML 97 compliant and require a VRML 97 capable viewer.

Isosurfaces were generated at specific values of the desired property. The values used were:

Note: values were chosen for display purposes and have no other significance.

Animation

Animated sequences derived from the rendered or VRML images at each point along the reaction path were used to visualize the dynamic change of the electron density, electrostatic potential, molecular orbitals, and laplacian of the electron density as a function of the progress of the reaction.

The rendered animations were created using a combination of ULEAD GIF Animator and Jasc's Animation Shop software from the 3D rendered images discussed above.

Animations of the VRML representations of the surfaces were created using custom code generated by the author (using suggestions from Robert Polevoi) that integrated the still VRML images into an animation sequence viewable with Cosmo Player. Animation of the VRML isosurface models allows real-time inspection of the progress of the reaction from any point of view.

Note: animation files can be quite large. Rendered image animations are usually 3 to 4 times larger in size than the corresponding VRML animation files. While rendered animations have a more 'photographic' look, VRML animations are usually more useful due to the ability of the observer to observe the animation from any point of view. In either case, I have displayed the file size for each animation file next to its hyperlink.

RESULTS

Results are found via the links in the left hand frame on this page.

CONCLUSIONS

• The methods work and provide an additional tool for chemical professionals to use in their analysis and understanding of chemical reaction phenomena.

• Additional work would be helpful. For example, volume visualization or multiple isosurface visualization would provide more insight into the totality of each property change as a function of reaction path.

• In addition, additional studies into similar chemical systems will allow possible discrimination of common reaction mechanisms or property variation as a function of reaction path which may be extensible to other systems of interest.

• Further studies will also examine the affect of a more limited MP2 molecular orbital basis set usage, the use of natural orbitals for better correlation estimates with smaller basis sets, and other new functionality in G98W that will allow extension of these methods to larger molecular systems.

 REFERENCES

[1] Gaussian 94, Revision D.3,
M. J. Frisch, G. W. Trucks, H. B. Schlegel, P. M. W. Gill, B. G. Johnson, M. A. Robb, J. R. Cheeseman, T. Keith, G. A. Petersson, J. A. Montgomery, K. Raghavachari,
M. A. Al-Laham, V. G. Zakrzewski, J. V. Ortiz, J. B. Foresman, J. Cioslowski, B. B. Stefanov, A. Nanayakkara, M. Challacombe, C. Y. Peng, P. Y. Ayala, W. Chen, M. W. Wong, J. L. Andres, E. S. Replogle, R. Gomperts, R. L. Martin, D. J. Fox, J. S. Binkley, D. J. Defrees, J. Baker, J. P. Stewart, M. Head-Gordon, C. Gonzalez, and J. A. Pople,
Gaussian, Inc., Pittsburgh PA, 1995