EMC: Enhanced Monte Carlo A Structure Generator

Pieter J. in ’t Veld, GMC/M: Materials Modeling

Choice Application Areas for Quantum Mechanics Overview Builder Attributes or Particle Dynamics Advantages and Capabilities

 Freeware under GPL v3.1  Structure builds in  Script-driven command line execution  Source code in C  Gas phase  Flexible input  Complex chemistry, including  Compiles under MacOS or  Condensed phase branched structures through SMILES with Intel and GNU compilers  Use of multiple force fields  Multiple force fields, e.g.  Command line with scripting interface  Atomistic typing by EMC – COMPASS  Advantage over commercial  Ports to PDB, LAMMPS, XYZ – OPLSUA solutions  Applications – DPD – Colloidal  Flexible scripting  Atomistic and coarse-grained  Algorithms  Can build and graft surfaces in structures condensed media  Formulation problems  Inverse Monte Carlo (CCB growth)  Mechanical, rheological and  Local spatial incremental relaxation interfacial properties  Composite core exclusion

Molecular Structure Investigations Complex Structure Builds Branched Semicrystalline Polyethylene Coupling 2D to 3D From this SMILES EMC builds Fullerene Mechanical Properties of Interphase

2D 3D

2D 3D

 Structure input through SMILES  Example: tryptophan (build based on COMPASS force field)  Draw structure with Accelrys Draw (or JChemPaint (Freeware)) 1D c12c3c4c5c1c6c7c8c2c9c1c3c2c3c4c4c%10c5c5c6c6c7c7c  Obtain SMILES (both Accelrys Draw and JChemPaint convert) %11c8c9c8c9c1c2c1c2c3c4c3c4c%10c5c5c6c6c7c7c In collaboration with G.C. Rutledge,  Build with EMC (resulting structure dependent on random seed) %11c8c8c9c1c1c2c3c2c4c5c6c3c7c8c1c23 MIT

Validation Validation Builder Validation FENE Model Description Comparison between Literature and Builder

 Describes polymeric behavior well  Ongoing research  Building polyethylene-like chains Auhl et al.

 Allows for comparison with previous  Comparison between kθ = 0  100, 200, 500 and 5000 repeat work by Auhl et al. equilibration simulations units

 Identical model parameters  10 million steps for building method  Variation of angle bend constant kθ after building  Rouse-like behavior for shorter  Identical integration conditions chains  Shape of the curve correct  Short cutoff ranges result in a  Follow exact trend as described in computationally inexpensive  Difference of 8% in final curve Building method Auhl et al. model between built and equilibrated  Directly after building  Building procedure consisted of  Can be fine-tuned with parameters equilibration  Structures already near 1000 chains of 500 length  Building method fast in equilibration  Systems equilibrated for 35 million constructing  Chain physics follow long time steps  Linear algorithm equilibration runs as performed by Auhl et al. K. Kremer and G. S. Grest, J. Chem. Phys. 92, 5057, 1990. R. Auhl et al., J. Chem. Phys. 119, 12718, 2003.

Curved Surfaces Curved Surfaces Simulations vs. Experiments DPD Building Concept Final DPD Structure

 EU Project NanoModel  Structure consists of 3.3 million sites (10K graft on 16 nm SiO2 in 10K matrix)  Grown with full force field  Both polymer graft and matrix are grown concurrently  Flexible choice of growth methods  Pure overlaps with grace  Energetic considerations with grace  Local relaxation at insertion point Crystal Carve Etch Graft Sphere Sphere Sphere  Final structure energetically close to equilibrium (5.11 vs eq 5.21 kT)  Build of 0.2 million sites ~30 minutes on laptop (5.5 million sites ~2 hours on single processor large machine) Experiments: W. Pyckhout, M. Meyer, Forschungszentrum Jülich FTI-MC: D. Theodorou, E. Voyiatzis, NTU Athens

External