Neuhauser Group

UCLA Department of Chemistry and Biochemistry

• Home
• Research
• People
• Publications
• Comp.Chem.
• Class Notes
• Contact
• StochasticGW

Selected recent publications

A GW approach that gives accurate charging energies for very large molecules and clusters (thousands of electrons and more, scaling linearly with system size due to the use of stochastic algorithms) [PDF]

A linear-scaling embedded-fragment stochastic approach to the densities and energies of large molecules (thousands of electrons or more). The method is non-approximate, and its new, embedded-fragment version uses very few stochastic orbitals (tens to few hundred orbitals regarless of system size) since the stochastic averaging only needs to model the small difference between the true (unbknown) density and the sum of fragment densities. [PDF]

CHEM126 - Computational Chemistry

Syllabus

This course will be taught twice a week. Office hours will be determined later. Since the course is held in a lab setting, it is imperative that you come to all lectures. TA to be determined.

Grading

You will be asked to complete a task every two weeks. In addition, you will be asked to complete two larger projects. I may assign you to work in pairs. [assignments_126.txt]

Goals

When you finish this course you should be able to a reasonable mathematical problem in chemistry, and convert it into a relevant mathematical program, which you will simulate and check.

Resources

Fortran 90

At home you should read about the language (FORTRAN 90/95, or simply FORTRAN 90 – the 95 version is only slightly different). Remember that this is not a course about FORTRAN 90/95, so we’ll learn and use only the features we’ll need.

While not so elegant or brief as languages like Mathematica or MathCAD, Fortran 90 is quite easy to use (much easier than, say, C++, and almost as easy as Matlab); large programs can be built and effectively debugged with modern compilers (like Lahey FORTRAN); and anybody with exposure to, say, one of the following: C, C++, Basic, Visual Basic, Java, Pascal, Matlab, Simulink, Mathematica, Maple, MathCAD can very quickly learn it.

• The first place to learn FORTRAN 90 is the first set of assignments, on numerical derivatives (and the next, on chemical reactions) – see later for details.
• But at home you should also consult the university of Belfast notes for the introductory-course: http://www.pcc.qub.ac.uk/tec/courses/f90/stu-notes/f90-stu.html
• If you want to look for more web-sources see: http://www.nsc.liu.se/~boein/fortran.html
• I’ve also assigned a book, FORTRAN 90/95 for Scientists and Engineers, by S. Chapman (McGraw Hill) -- it is good as a reference point, although it contains more features than we’ll need.

Editor

• We will be using the text editor Emacs.
• First, read the attached document on the editor, [emacs_guide.txt] .
• You can also do this in your Linux terminal by running "emacs emacs_guide.txt".

Reading

• You should read the Notes by Prof. Hannes Jonsson, University of Washington, from a course he taught, starting with the set on differential equations (the notes will be supplied separately in your directory) as well as the main course-notes: [PDF^∗, 223K] .
• Finally, all the fortran files used in the course are in the following tar file (Chem. 126 students: you do not need to download this tar file, since the files are already on the computer we'll use): [126_fortran.tar] .