Masters/PhD Course: Computational Chemistry BB2300

Course description

This course is intended in the first place to the students of the International Masters Programme in Scientific Computing who wish to learn about computational chemistry and the way it uses numerical analysis and numerical methods.

Computational chemistry has made in recent years big advances in understanding of chemical processes. The field has become very broad, with a variety of methods with varying level of generality, depending on the investigated systems and processes. The purpose of this course is two-fold. First, an overview of available computational methods: ranging from molecular dynamics to ab-initio methods. More attention will be devoted to the very popular Hartree-Fock (HF) approximation and widely used Kohn-Sham formulation of density functional theory. Even semi-empirical methods will be covered as a special case of HF. Second part of the course will give a working knowledge of mathematical and computational methods used in HF and Kohn-Sham programs. This includes representing the wave function representing the molecule with help of assumptions about the form of the solution and a basis set, efficient ways of computing resulting matrix elements, and solving relevant equations. This theoretical part is illustrated with examples based on simple systems.

Course Objectives

After a successful course completion, the student will be able to:

• Solve the equations resulting from the HF model. Ability to point out the time- and memory bottlenecks as well as to propose appropriate solutions will be graded higher.
• Produce a simple self-consistent field cycle based HF implementation for a small molecule. Ability to apply it for eg. Prediction of interatomic distances, and analysis of the outcome will give a higher grade.
• Explain the limitations of HF and DFT, as well as impact of basis set and DFT functional choice on the calculation trustworthiness; be able to propose adequate basis sets for selected calculation types.
• Judge applicability of various chemical modeling methods for chemical applications, as constrained by the approximations made.

Language

The course is given in English. Reports can be written in English and Swedish.

Examination

A set of 4-5 assignments is handed out during the course. The assignments involve solving an analytical or numerical problem and writing a short report (most often 1-2 pages; final report up to 5 pages).

• Problems can be solved in small groups (two members)
• Reports are individual – please indicate with whom you solved the problem.

• Part of the course is learning of the peer-review process. This means that each participant will review and suggest grade for 1-2 other participants. Reviewing instructions will be given.

  Reports are to describe the solution so that it can be understood and reproduced. Some questions to be answered in the report are formulated in an open way and may have more than one solution: it is then necessary to elaborate on what options were considered. If the solution is numerical, the source code is to be sent to the course teacher in the electronic form.

• Handing in the last report will be connected with a short interview that gives the student to elaborate on the report's content.

Literature

The first item listed below is the primary one (F. Jensen). Other items serve only complementary purpose:

1. 1.F. Jensen: Introduction to Computational Chemistry.
2. P. W. Atkins and R.S Friedman: Molecular Quantum Mechanics.
3. C.J. Cramer: Essentials of Computational Chemistry - theories and models.
4. T. Helgaker, P. Jørgensen, J. Olsen: Molecular Electronic-Structure Theory.

Related links

• Course's entry in KTH Student's handbook
• Läsårindelning

Theoretical chemistry > Courses

webmaster-at-theochem.kth.se