raaSAFT: A framework enabling coarse-grained molecular dynamics simulations based on the SAFT- Mie force field

详细信息    查看全文

• 作者：Å ; smund Ervik^a ; ^{asmunder@pvv.org} ; Guadalupe Jimé ; nez Serratos^b ; ^{m.jimenez-serratos@imperial.ac.uk} ; Erich A. M&uuml ; ller^b ; ^{e.muller@imperial.ac.uk}
• 关键词：Molecular dynamics ; GPU ; HOOMD-blue ; Coarse-grained ; SAFT-γγ Mie
• 刊名：Computer Physics Communications
• 出版年：2017
• 出版时间：March 2017
• 年：2017
• 卷：212
• 期：Complete
• 页码：161-179
• 全文大小：4196 K
• 卷排序：212

文摘

We describe here raaSAFT, a Python code that enables the setup and running of coarse-grained molecular dynamics simulations in a systematic and efficient manner. The code is built on top of the popular HOOMD-blue code, and as such harnesses the computational power of GPUs. The methodology makes use of the SAFT-γγ Mie force field, so the resulting coarse grained pair potentials are both closely linked to and consistent with the macroscopic thermodynamic properties of the simulated fluid. In raaSAFT both homonuclear and heteronuclear models are implemented for a wide range of compounds spanning from linear alkanes, to more complicated fluids such as water and alcohols, all the way up to nonionic surfactants and models of asphaltenes and resins. Adding new compounds as well as new features is made straightforward by the modularity of the code. To demonstrate the ease-of-use of raaSAFT, we give a detailed walkthrough of how to simulate liquid–liquid equilibrium of a hydrocarbon with water. We describe in detail how both homonuclear and heteronuclear compounds are implemented. To demonstrate the performance and versatility of raaSAFT, we simulate a large polymer-solvent mixture with 300 polystyrene molecules dissolved in 42 700 molecules of heptane, reproducing the experimentally observed temperature-dependent solubility of polystyrene. For this case we obtain a speedup of more than three orders of magnitude as compared to atomistically-detailed simulations.Program summaryProgram title: raaSAFTCatalogue identifier: AFBE_v1_0Program summary URL:http://cpc.cs.qub.ac.uk/summaries/AFBE_v1_0.htmlProgram obtainable from: CPC Program Library, Queen’s University, Belfast, N. IrelandLicensing provisions: MIT LicenceNo. of lines in distributed program, including test data, etc.: 802350No. of bytes in distributed program, including test data, etc.: 30446478Distribution format: tar.gzProgramming language: Python.Computer: Any computer, optionally with Nvidia GPU(s).Operating system: Linux, Mac OSX.RAM: Depends on number of atoms and cutoff sizeClassification: 7.7, 16.13.External routines: HOOMD-blue [1][2]Nature of problem:The behaviour and properties of simple and complex fluids, including mixturesSolution method:Coarse-grained molecular dynamics using the SAFT-γγ Mie force field [3].Restrictions:Ions and ionic compounds are not supported yet. Jobscripts running with Python 2 require HOOMD-blue v1.3 or newer; for Python 3 there is no such restriction.Unusual features:Uses object-oriented programming to make reuse and sharing of models very simple. Allows the simulation to be set up and executed completely programmatically, i.e. without the use of a GUI or preprocessor. Force field parameters are available from an online database with more than 6000 molecules, http://www.bottledsaft.org [4].Additional comments:The code is hosted on http://bitbucket.org/asmunder/raasaftRunning time:On a single high-end GPU in 2015 (Nvidia K40), around 2.5 nanoseconds per hour of walltime for a million atoms (not counting hydrogens).References:[1]J.A. Anderson, C.D. Lorenz, A. Travesset, General purpose molecular dynamics simulations fully implemented on graphics processing units, Journal of Computational Physics 227 (2008) 5342–5359.[2]J. Glaser, T.D. Nguyen, J.A. Anderson, P. Lui, F. Spiga, J.A. Millan, D.C. Morse, S.C. Glotzer, Strong scaling of general-purpose molecular dynamics simulations on GPUs, Computer Physics Communications 192 (2015) 97–107.[3]E.A. Müller, G. Jackson, Force-field parameters from the SAFT-γγ equation of state for use in coarse-grained molecular simulations, Annual review of chemical and biomolecular engineering 5 (2014) 405–427.[4]Å. Ervik, A. Mejía, E. A. Müller, Bottled SAFT: A web app providing SAFT-γγ Mie force field parameters for thousands of molecular fluids, In preparation. (2016).