Hello, A new project on the LCRC cluster has been requested. Please forward the information on to the LCRC Allocation sub-committee. Applicant's name: Dongwei Xu Applicant's institution: ANL Applicant's division: MSD Project Name: CrysGrowKMC Project title: Kinetic Monte Carlo simulation on the crystal thin film growth Associated funding: DOE Other Systems: Science: Understanding growth of films on surfaces of various crystal orientations is essential to technologies such as GaN-based light-emitting diodes and high-power electronics. Kinetics Monte Carlo (KMC) simulation is a powerful method to simulate the film growth. There are several fundamental questions are worthy to explore, i.e., the temperature effect in the competition of atom diffusion and deposition during the growth, what affects the island shape, how to connect the step edge barrier to the behavior of the steps and what is the correlation between the island during the growth. The KMC simulation will help to understand these questions. By scaling the simulation parameters according to the experiments, we can give a connection between the experiment and KMC simulation. Our simulation results will help to design and guide experimental studies. Project description: The project started from the beginning of 2015 and we have performed calculations using Kinetic Monte Carlo simulation on the thin film growth on c- and m-planes of GaN. Next, we will study the time correlation of the surface step evolutions at equilibrium and during the growth. We will also study the situation when evaporation and condensation are both included in the model. The Stochastic Parallel PARticle Kinetic Simulator (SPPARKS) will be used to do the simulation. SPPARKS is an open source code developed at Sandia National Lab. It is installed at LCRC and we can modify the code to install in our own directory as needed. The code can only ran in one single core when the atom deposition is included. However, we have written a parallel script to run 16 (or any other) single jobs at the same time using the same node. The computation time is size and parameters dependent. The typical running time is 15 hours for one simulation condition and there are about 64 different situations for both c- and m-planes. Matlab, which is also installed at LCRC, is used to do the analysis. We can only use single core right now and it is very memory consuming (sometimes, we need to use BIGgpu). We plan to include the evaporation in our simulation, which is more time consuming. For each configuration, we need about 25 hours of run time when using 16 cores. Therefore, expected time estimate for simulation with evaporation is (16 cores/node * 25 hours of run time *64*2) 512,00 core-hours. We would also study the steps and the time correlation for both c- and m-planes, which needs about 102,400 core-hours. The time needed for the matlab program is about (16 cores/node * 3 hours of run time *64*3*2) 18,432 core-hours. We also want to do some tests on larger system sizes, which adds about 20,000 core-hours. The total time requested is about 200,000 core-hours. There will be two to three members in the project. Industry partnership: Project URL: Requested allocation: 200000 Q1: 40000 Q2: 60000 Q3: 60000 Q4: 40000 Justification: Storage requirements: 20TB. Every single output is a very large data file (about 1.8GB) that will be used to analysis later. We need at least 16 runs for one condition and there are 64*2*3=384 situations. The total will be about 11TB. We will also calculate some larger system which will have lager output files. The requester has used undetermined amount hours of their initial startup project. In addition to approving an initial amount, please specify a Category and Subcategory for this project. For a list of the current selection of approved categories, please see: https://wiki.lcrc.anl.gov/wiki/Processes/Categories Once the Allocation committee has approved the project, please go to the Project Management page to create it: https://accounts.lcrc.anl.gov/projects.php Thank You, The LCRC Accounts System