University of Kentucky
Chemical and Materials Engineering
Department
CME
599 / MSE 620: Computational Materials Science & Engineering
Fall
2008
Instructor: Dr.
Stephen Rankin Office
Hour:
159
FPAT (Anderson Tower) W
10:00—11:00 am
257-9799 or
by appointment
srankin@engr.uky.edu
Textbook: D.
Frenkel and B. Smit, Understanding Molecular Simulation, 2nd ed.
Academic Press (2002)
Optional supplement: T. L. Hill, An
Introduction to Statistical Thermodynamics, Dover, NY: 1986.
Time and Place: Tuesday and Thursday, 2:00-3:15
pm, 269 FPAT
Course Relevance
Molecular simulations are to statistical
mechanics as numerical methods such as the finite element method are to
continuum mechanics. They are computational techniques that allow one to derive
macroscopic, observable properties from atomic or molecular interactions. Just
as there are only some cases where analytical solutions can be found in continuum
mechanics (for instance, diffusion into a 1-D semi-infinite slab), in
statistical mechanics, only certain simplified cases (such as ideal gases or
monatomic crystals) can be solved exactly using analytical techniques.
Molecular simulations expand the range of problems that can be solved by using
computational methods to derive equilibrium and kinetic properties from the
interactions among collections of particles. As available computational power
expands, these methods provide an important complement to other experimental
and theoretical research methods. They allow one to conduct “numerical
experiments” that provide molecular understanding of macroscopic properties, or
can be used for screening and discovery of specialty chemicals, materials, or
pharmaceuticals.
Topics to Be Covered
• Brief overview of quantum mechanical methods
• Interatomic potentials: Hard spheres,
Lennard-Jones fluids, long-range interactions
• Why molecular simulations (vs. quantum or
fluid mechanics)?
• Statistical thermodynamics, probability and
statistics background
• Monte Carlo simulations: integration and
sampling
• Molecular dynamics: equilibrium and
non-equilibrium methods
• Ensembles and when to use them
• Free energies and phase equilibria
• Intramolecular interactions and complex
fluids
• Barrier crossing, Monte Carlo dynamics, and
Markov chains
• Accelerating
simulations, coarse-graining, and bridging length scales
Expected Outcomes
At the completion of this course,
students should:
1. Have been provided with the background needed to understand molecular simulations
2. Understand basic Monte Carlo and molecular dynamics simulation methods
3. Know what interaction potentials are available
and how they are parameterized
4. Distinguish between different ensembles, and
know when each is appropriate
5. Have
gained experience writing and using computer programs for molecular simulation
6. Understand the strengths and limitations of
molecular simulation methods
7. Be able to select and apply molecular
simulations to new problems
8. Write about
and critically read others’ reports about molecular simulations.
Additional Resources Available from UK Libraries
• M. R. Allen and D. J. Tildesley Computer
Simulations of Liquids, Oxford: 1987.
• J. M. Haile Molecular dynamics simulations, Wiley: 1992.
·
D.P. Landau and K. Binder, A Guide to Monte Carlo
Simulations in Statistical Physics, Cambridge University Press, 2005, http://www.netlibrary.com/AccessProduct.aspx?ProductId=139749
• D. A. McQuarrie Statistical Mechanics, Harper
Collins: 1976.
Important Dates
August 28 First class session.
October 20 Midterm of the semester.
November 27 Thanksgiving Holiday - No Classes.
December 11 Last day of class.
December 18 Final report on project due by 5:00 pm.
Exams and Project
Two exams will be scheduled during
regular class periods. These exams will be based on the subject matter of the
course, but will not require computer simulation for their solution. Dates for exams will be set at appropriate
points in the course, and will be announced at least one week in advance. Undergraduate and graduate students will take
the same exam, but grades will be assigned separately.
In lieu of a final exam, graduate students
will be required to complete a research project on a topic of their own
choosing. For graduate students, this project would ideally be related to their
dissertation. Any molecular simulation method can be used for this project,
with the consent of the instructor, even if it is not covered in depth in this
course. The project can be based on work in research literature, but must make
some original contribution (either simulations, formulation of a new simulation,
or critical review).
One-page biweekly progress reports on
the project will be required from graduate students, and weekly reports for
undergraduate students taking the class. The first should describe the project to
be undertaken and will be due on September 25th. You are encouraged
to meet with the instructor to help to select the topic. The final report on
this project should be modeled on peer-reviewed research papers. It should an appropriate introduction to the
method(s) employed, a survey of related work in print, results, discussion, and
conclusions. It should not be a general term paper or summary of the course.
Graduate students will also be required to give a mini-lecture on a
topic relevant to the course at some point in the semester. The mini-lectures
will be in the second half of the semester, and can be based on the literature
review (although they can also be on a more traditional topic instead). They
will be evaluated by the instructor, primarily for technical content.
Avoiding Plagiarism
Students should be careful to avoid
plagiarism, even if unintentional. Plagiarism includes not only verbatim
copying of whole sentences or paragraphs, but also “borrowing” someone else’s
text and making minor changes (substituting words or rearranging phrases). Any
text or sequence of ideas taken from another source must be clearly and
specifically cited. It will be far preferable to submit a report with imperfect
grammar than to risk receiving an “E” in the course due to plagiarism.
Homework Assignments
The homework assignments are an
opportunity for the student to test her or his understanding of basic concepts
and to develop problem solving skills at relatively little risk to their final
grade. They also provide the opportunity for the instructor to monitor
understanding of the material and to adjust the pace of the course. Group
discussion of problems aids learning, for everyone involved. Debate of the
approach to homework problems is encouraged, but students are required to
independently write their solutions. When computer programming is needed,
students must each write their own program.
Grading
The worst possible grading scale will
be:
90%—100% = A; 80%-89% = B ; 70%—79% = C ; 60%—69% = D
; <60% = E
The instructor may choose to adjust this
scale in favor of higher grades. The weighting of course components will be:
Homework 10% (20% for undergraduate
students)
Mini-lecture 10% (n/a for undergraduate
students)
Exams (2) 20% each
Research Project 40%
Feedback
Every effort will be made to provide
timely, helpful feedback to students regarding their progress and their class
standing. Feedback from students about the course is also encouraged.
You will have the opportunity to evaluate the course at the end, but that is
too late to make improvements. Constructive criticism and suggestions to
improve your learning of the material are always welcome.
CORRECTIONS, CHANGES, AND ADDITIONAL INFORMATION ARE
AVAILABLE ON LINE AT:
http://www.engr.uky.edu/~srankin/CME599.htm