General Procedure for Characterizing a Reaction

Each semiempirical computational chemistry program has specific functions rather than use any specific program, we will consider a hypothetical program, although the bias will be toward the MOPAC series of programs. 

Each method used corresponds to a different reality the value of any given method lies in how closely that reality matches the real universe, and how expensive the simulations are to run. Methods commonly used are 

MINDO/3, MNDO, MNDOC, SINDOl, CNDO/ ZINDO, AMI, and recently PM3, as mentioned before. Newcomers to die field should be aware that the program and the method it runs may have, but do not need to have, the same name. MOPAC, for example, will allow MlNDO/3, MNDO, AMI, and PM3 calculations to be run. 

A general reaaion could be written as A + B TS C + D, where TS is a transition state. Not all species need be present in some reactions either B or D or both would be missing. For the purposes of a theoretical investigation A, B, C, and D are stable species, and TS is the saddle point on the reaction coordinate connecting reactants to products. 

Before commencing the investigation proper, the theoretician, like the experimentalist, must prepare starting materials. For the theoretician, these consist of fully optimized geometries of all stable species. Normally no constraints are applied at this point. Individual molecules involved in a reaction may have high symmetry, but once any reaction gets under way, almost invariably any symmetry present is destroyed. 

---