
[Illinois] CSE Seminar Series: Advances in Firstprinciples Computational Materials Science
20 Nov 2012  Online Presentations  Contributor(s): Elif Ertekin
Title: Advances in firstprinciples computational materials science
Subtitle: Things we can calculate now, that we couldn't when I was in grad school.
The capability to rationally design new materials with tailored properties and functionality on a computer remains a grand challenge whose success would have tremendous impact on several globallyrelevant issues. Guided materials design in the form of Integrated Computational Materials Engineering (ICME) is now taking its foundational steps …

Computational Nanoscience, Lecture 20: Quantum Monte Carlo, part I
15 May 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
This lecture provides and introduction to Quantum Monte Carlo methods. We review the concept of electron correlation and introduce Variational Monte Carlo methods as an approach to going beyond the mean field approximation. We describe briefly the SlaterJastrow expansion of the wavefunction, and show how we can recover the some of the correlation energy using a variational approach to optimizing this form of the wavefunction.Lucas K. Wagner
University of California, Berkeley

Computational Nanoscience, Lecture 21: Quantum Monte Carlo, part II
15 May 2008  Teaching Materials  Contributor(s): Jeffrey C Grossman, Elif Ertekin
This is our second lecture in a series on Quantum Monte Carlo methods. We describe the Diffusion Monte Carlo approach here, in which the approximation to the solution is not restricted by choice of a functional form for the wavefunction. The DMC approach is explained, and the fixed node approximation is described as well. We conclude with a few examples demonstrating the application of VMC and DMC to methane and ethane.Lucas K. Wagner
University of California, Berkeley

Computational Nanoscience, PopQuiz
15 May 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
This quiz summarizes the most important concepts which have covered in class so far related to Molecular Dynamics, Classical Monte Carlo Methods, and Quantum Mechanical Methods.University of California, Berkeley

Computational Nanoscience, PopQuiz Solutions
15 May 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
The solutions to the popquiz are given in this handout.University of California, Berkeley

Computational Nanoscience, Lecture 23: Modeling Morphological Evolution
15 May 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
In this lecture, we present an introduction to modeling the morphological evolution of materials systems. We introduce concepts of coarsening, particlesize distributions, the LifshitzSlyozovWagner model, thin film growth modes (LayerbyLayer, Island growth, and StranskiKrastanov), and morphological instabilities. An introduction to phase field modeling, diffuse interface models, and CahnHilliard and CahnAllen analysis is presented. We conclude with some examples of phase field methods ...

Computational Nanoscience, Lecture 27: Simulating Water and Examples in Computational Biology
16 May 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
In this lecture, we describe the challenges in simulating water and introduce both explicit and implicit approaches. We also briefly describe protein structure, the Levinthal paradox, and simulations of proteins and protein structure using First Principles approaches and Monte Carlo sampling.University of California, Berkeley

Computational Nanoscience, Lecture 28: WishList, Reactions, and XRays.
16 May 2008  Teaching Materials  Contributor(s): Jeffrey C Grossman, Elif Ertekin
After a brief interlude for class feedback on the course content and suggestions for next semester, we turn to modeling chemical reactions. We describe chainofstate methods such as the Nudged Elastic Band for determining energy barriers. The use of empirical, QM/MM methods are described. We give some examples of Hartree Fock and DFT methods applied to determining dissociation energies, and show where these methods can fail.University of California, Berkeley

Computational Nanoscience, Lecture 29: Verification, Validation, and Some Examples
16 May 2008  Teaching Materials  Contributor(s): Jeffrey C Grossman, Elif Ertekin
We conclude our course with a lecture of verification, and validation. We describe what each of these terms means, and provide a few recent examples of nanoscale simulation in terms of these concepts.University of California, Berkeley

Computational Nanoscience, Lecture 19: Band Structure and Some InClass Simulation: DFT for Solids
30 Apr 2008  Teaching Materials  Contributor(s): Jeffrey C Grossman, Elif Ertekin
In this class we briefly review band structures and then spend most of our class on inclass simulations. Here we use the DFT for molecules and solids (Siesta) course toolkit. We cover a variety of solids, optimizing structures, testing kpoint convergence, computing cohesive energies, and computing band structures and density of states.

Computational Nanoscience, Lecture 18.5: A Little More, and Lots of Repetition, on Solids
30 Apr 2008  Teaching Materials  Contributor(s): Jeffrey C Grossman, Elif Ertekin
Here we go over again some of the basics that one needs to know and understand in order to carry out electronic structure, atomicscale calculations of solids.

Computational Nanoscience, Lecture 16: More and Less than HartreeFock
30 Apr 2008  Teaching Materials  Contributor(s): Jeffrey C Grossman, Elif Ertekin
In the lecture we discuss both techniques for going "beyond" HartreeFock in order to include correlation energy as well as techniques for capturing electronic structure effects while not having to solve the full HartreeFock equations (ie, semiempirical methods). We also very briefly touch upon the pseudopotential approximation.

Computational Nanoscience, Lecture 15: InClass Simulations: HartreeFock
30 Apr 2008  Teaching Materials  Contributor(s): Jeffrey C Grossman, Elif Ertekin
Using a range of examples, we study the effect of basis set on convergence, the HartreeFock accuracy compared to experiment, and explore a little bit of molecular chemistry.

Computational Nanoscience, Lecture 14: HartreeFock Calculations
30 Apr 2008  Teaching Materials  Contributor(s): Jeffrey C Grossman, Elif Ertekin
A description of the HartreeFock method and practical overview of its application. This lecture is to be used in conjunction with the course toolkit, with the HartreeFock simulation module.

Computational Nanoscience, Lecture 13: Introduction to Computational Quantum Mechanics
30 Apr 2008  Teaching Materials  Contributor(s): Jeffrey C Grossman, Elif Ertekin
In this lecture we introduce the basic concepts that will be needed as we explore simulation approaches that describe the electronic structure of a system.

Computational Nanoscience, Lecture 17: TightBinding, and Moving Towards Density Functional Theory
21 Mar 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
The purpose of this lecture is to illustrate the application of the TightBinding method to a simple system and then to introduce the concept of Density Functional Theory. The motivation to mapping from a wavefunction to a densitybased description of atomic systems is provided, and the necessary ingredients to do so (two HohenbergKohn Theorems and the KohnSham formalism) is presented.University of California, Berkeley

Computational Nanoscience, Lecture 18: Density Functional Theory and some Solid Modeling
21 Mar 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
We continue our discussion of Density Functional Theory, and describe the mostoften used approaches to describing the exchangecorrelation in the system (LDA, GGA, and hybrid functionals). We discuss as well the strengths and weaknesses of the LDA and present some examples of its use. Finally, a short introduction to modeling band structures in solids is presented.University of California, Berkeley

Computational Nanoscience, Lecture 12: InClass Simulation of Ising Model
28 Feb 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
This is a two part lecture in which we discuss the spinspin correlation function for the the Ising model, correlation lengths, and critical slowing down. An inclass simulation of the 2D Ising Model is performed using the tool "Berkeley Computational Nanoscience Class Tools". We look at domain wall formation at low temperature, and the phase transition for the antiferromagnetic and ferromagnetic system.
University of California, Berkeley

Computational Nanoscience, Homework Assignment 4: HardSphere Monte Carlo and Ising Model
05 Mar 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
In this assignment, you will explore the use of Monte Carlo techniques to look at (1) hardsphere systems and (2) Ising model of the ferromagneticparamagnetic phase transition in twodimensions. This assignment is to be completed following lecture 12 and using the "Hard Sphere Monte Carlo" and the "Ising Model" program in the Berkeley Computational Nanoscience Toolkit.University of California, Berkeley

Computational Nanoscience, Lecture 10: Brief Review, Kinetic Monte Carlo, and Random Numbers
25 Feb 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
We conclude our discussion of Monte Carlo methods with a brief review of the concepts covered in the three previous lectures. Then, the Kinetic Monte Carlo method is introduced, including discussions of Transition State Theory and basic KMC algorithms. A simulation of vacancymediated diffusion is provided as an example of KMC. Finally, a brief primer on random number generation is presented.University of California, Berkeley

Computational Nanoscience, Lecture 11: Phase Transitions and the Ising Model
27 Feb 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
In this lecture, we present an introduction to simulations of phase transitions in materials. The use of Monte Carlo methods to model phase transitions is described, and the Ising Model is given as an example for modeling the ferromagneticparamagnetic transition. Some of the subtleties of simulating phase transitions are also discussed, including finite size effects and critical slowing down. The concept of linear response is introduced as well.University of California, Berkeley

Computational Nanoscience, Lecture 9: HardSphere Monte Carlo InClass Simulation
19 Feb 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
In this lecture we carry out simulations inclass, with guidance from the instructors. We use the HSMC tool (within the nanoHUB simulation toolkit for this course). The hard sphere system is one of the simplest systems which exhibits an orderdisorder phase transition, which we will explore with Monte Carlo simulations.Nanoscale Science and Engineering C242/Physics C203 University of California, Berkeley

Computational Nanoscience, Lecture 7: Monte Carlo Simulation Part I
15 Feb 2008  Teaching Materials  Contributor(s): Jeffrey C Grossman, Elif Ertekin
The purpose of this lecture is to introduce Monte Carlo methods as a form of stochastic simulation. Some introductory examples of Monte Carlo methods are given, and a basic introduction to relevant concepts in statistical mechanics is presented. Students will be introduced to the Metropolis approach to Monte Carlo simulation. Using Metropolis as an example, these lectures also introduce the comcepts of balance and detailed balance, and what "efficient sampling" means.

Computational Nanoscience, Lecture 8: Monte Carlo Simulation Part II
14 Feb 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
In this lecture, we continue our discussion of Monte Carlo simulation. Examples from Hard Sphere Monte Carlo simulations based on the Metropolis algorithm and from Grand Canonical Monte Carlo simulations of fullerene growth on spherical surfaces are presented. A discussion of meaningful statistics, result interpretation, and error analysis is presented as well.University of California, Berkeley.

Computational Nanoscience, Homework Assignment 3: Molecular Dynamics Simulation of Carbon Nanotubes
14 Feb 2008  Teaching Materials  Contributor(s): Elif Ertekin, Jeffrey C Grossman
The purpose of this assignment is to perform molecular dynamics simulations to calculate various properties of carbon nanotubes using LAMMPS and Tersoff potentials.
This assignment is to be completed following lectures 5 and 6 using the "LAMMPS" program in the Berkeley Computational Nanoscience Toolkit.University of California, Berkeley