Biomedical Engineering Reference
In-Depth Information
24.4 ONIOM (Our Own N -layered Integrated Molecular Orbital
and Molecular Mechanics)
We note several different schemes in which a molecule is divided into two parts, with one
accurately described and the other treated at a lower level of theory. The total energy of the
real system is a combination of both parts. For molecular systems normally treated with
MM, the integrated IMOMMmethod is appropriate. For systems and reactions that require
a more accurate description of the geometries and energies, the integrated IMOMOmethod
has been shown to reproduce high-level ab initio results. The next logical step involves
combining these approaches into an onion-like multilayered scheme, with MM used to
describe steric and electrostatic effects in the exterior parts of a molecule, an intermediate
MO method describing the electronic effects of functional groups or ligands close to the
centre of action, and a highly accurate method to deal with the electron correlation on the
most important parts of the molecule. The layers are conventionally known as the low,
medium and high layers; these are treated with different model chemistries and the results
automatically combined to give the final predicted results, and such a calculation would
be denoted ONIOM3 (Svensson et al. 1996). Rather than quote the original abstract, I will
give you a more up-to-date version due to Vreven and Morokuma (2000) that contains a
wealth of useful detail.
Five years ago Morokuma and colleagues introduced the IMOMO method, which integrates
two molecular orbital methods into one calculation. Since then, the method has been expanded
in several ways; it has been generalized to consider up to three methods, and has been unified
as the ONIOM method to include both MO and MM combinations. In this review we present
the history of the method, a number of chemical problems that we have studied, how to assess
IMOMO combinations and partitionings, and our latest efforts that take the method beyond
the conventional investigation of ground state energy surfaces. In particular we emphasize
the importance of the S-value test for validation of the ONIOM method/model combinations.
The method combination depends much on the properties and accuracies required. Generally
speaking, however, if the target level is CCSD(T) or G2, the best choice of low level is MP2.
If MP2 or DFT is the target level, HF or eventually semiempirical MO methods are good
choices of low level. These methods can be further combined with an outer-most layer of the
MM level.
References
Gao, J. and Thompson, M.A. (eds) (1998) Combined Quantum Mechanical and Molecular
Mechanical Methods , ACS Symposium Series 712, American Chemical Society, Washington, DC.
Humbel, S., Sieber, S. and Morokuma, K. (1996) J. Chem. Phys. , 105 , 1959.
Maseras, F. and Morokuma, K. (1995) J. Comput. Chem. , 16 , 1170.
Svensson, M., Humbel, S., Froese, R.D.J. et al. (1996) J. Phys. Chem. , 100 , 19357.
Vreven, T. and Morokuma, K. (2000) J. Comput. Chem. , 21 , 1419.
Warshel, A. and Levitt, M. (1976) J. Mol. Biol. , 103 , 227.
Search WWH ::




Custom Search