| |
| 1.1 Introduction |
| Chemistry is essentially an experimental science dealing with the structure, functions and reactivity of molecules and the methods for making them from convenient starting materials. With the advent of quantum mechanics, a better understanding of atomic and molecular structure was initiated but significant progress was delayed because of the difficulties in calculating all the necessary multicenter integrals and the necessity of diagonalizing large matrices. With the arrival of faster and even faster computers, these calculations are becoming possible and getting more accurate. A study of large molecular clusters and liquids was also initiated using interaction potentials and classical equations of motion. Now these methods are extended to surfaces, multiple phases and topics covering various aspects of molecular biology. In the present course, we shall investigate many aspects that touch upon the issues related to the above problems. All the above mentioned problems use the methods of numerical analysis, often applied iteratively and recursively to solve the problems of chemical interest. Hence we begin with an introduction to the methods of numerical analysis. Next, we shall study the ab initio and-semi empirical quantum chemical methods which form an important core of theoretical/computational chemistry. Various examples will be illustrated using the Gaussian programs available in the public domain. The later part of the course deals with computer simulations involving a large number of molecules. As fully quantum mechanical calculations are not yet possible for large systems, methods of classical mechanics are used and meaningful results can be obtained for comparison with experimental data. The final section of the course is concerned with molecules of chemical as well as biological interest. |
| Over the years, there has been a lot of interplay between laboratory experiments and computational predictions. Often, new molecules have been predicted and later on, confirmed by experiments. When there are several pathways possible for a reaction, computations of the potential energy profiles for different pathways can indicate which of the pathways corresponds to a lower energy path and further experiments can be performed to isolate “clusters” corresponding to the lower energy paths. Classical MD/MC simulations provide very detailed pictures of molecular motions which are not accessible experimentally. In problems such as protein folding in different solvent media, molecular simulations can yield comprehensive pictures of crucial solute and solvent configurations that play an important role in the dynamics of the folding process. |
|
|
|
