Iterative reverse Monte Carlo and molecular statics for improved atomic structure modeling: a case study of zinc oxide grown by atomic layer deposition.
Ryan C GettlerHenry D KoenigMatthias J YoungPublished in: Physical chemistry chemical physics : PCCP (2021)
Reverse Monte Carlo (RMC) modeling is a common method to derive atomic structure models of materials from experimental diffraction data. However, conventional RMC modeling does not impose energetic constraints and can produce non-physical local structures within the simulation volume. Although previous strategies have introduced energetic constraints during RMC modeling, these approaches have limitations in computational cost and physical accuracy. In this work, we periodically introduce molecular statics (MS) energy minimizations during RMC modeling in an iterative RMC-MS approach. We test this iterative RMC-MS approach using diffraction data collected by in operando high energy X-ray diffraction during atomic layer deposition of ZnO as a sample case. For MS relaxations we employ ReaxFF pair potentials previously established for ZnO. We find that RMC-MS and RMC provide equivalent agreement with experimental data, but RMC-MS structures are on average 0.6 eV per atom lower in energy and are more consistent with known ZnO atomic structure features. The iterative RMC-MS approach we report can accommodate large systems with minimal additional computational burden beyond a standard RMC simulation and can leverage established pair potentials for immediate application to study a wide range of materials.
Keyphrases
- mass spectrometry
- multiple sclerosis
- ms ms
- monte carlo
- electron microscopy
- high resolution
- image quality
- physical activity
- electronic health record
- mental health
- big data
- dual energy
- computed tomography
- risk factors
- magnetic resonance imaging
- magnetic resonance
- molecular dynamics simulations
- deep learning
- data analysis
- contrast enhanced