Reader

MA MSci PhD MInstP FHEA

I am a Reader in the School of Engineering at the University of Warwick, where I am also associated with the EPSRC Centre for Doctoral Training in Modelling of Heterogeneous Systems (HetSys; Co-director) and the Warwick Centre for Predictive Modelling (WCPM; Co-director).

June 2019-present Reader, School of Engineering, University of Warwick
2016-2019 Associate Professor, School of Engineering, University of Warwick
2014-2016 Assistant Professor, School of Engineering, University of Warwick 
2009-2014 Postdoc in the Department of Physics at King's College London
2007-2008 Postdoc in the Department of Engineering at the University of Cambridge
2004-2007 PhD in the TCM Group at the Cavendish Laboratory, University of Cambridge

Teaching

In 2019/20 I am teaching ES386 - Dynamics of Vibrating Systems (with Dr Peter Brommer) and PX914 - Predictive Modelling and Uncertainty Quantification in the HetSys CDT.

Previously I was module leader for IL027 Interdisciplinary Computer Modelling (with Prof. Christoph Ortner).

My office hours are Fridays 3-4pm during term.

Vacancies

• Multiple PhD projects available through the HetSys Centre for Doctoral Training for an October 2020 start:
  
  • Adaptive probabilistic meshless methods for evolutionary systems with Tim Sullivan

News

• May 2020 Interviewed for the Multiscale Musings podcast series by HetSys CDT PhD students
• Feb 2020 Article on quantum/classical modelling of dislocations in tungsten out in Phys. Rev. Materials
• Feb 2019 HetSys Centre for Doctoral Training launched at Warwick, with several fully funded PhD positions open, including one to work with me and Christoph Ortner on advanced boundary conditions for defect simulations
• Dec 2018 Article on machine learning a general-purpose interatomic potential for silicon published in Physical Review X

Research Interests

I develop multiscale materials modelling algorithms and the software that implements them. My recent work applies this parameter-free modelling to make quantitative predictions of "chemomechanical" materials failure processes where stress and chemistry are tightly coupled, e.g. near the tip of a propagating crack (left), where local bond-breaking chemistry is driven by long-range stress fields. Recents projects include:

Slow Crack Growth in Brittle Crystals. Everyday experience suggests that when brittle materials like glass or silicon wafers break they crack very fast, usually at a large fraction of the speed of sound in the material. In research published in Physical Review Letters, we presented QM-based atomistic simulations together with new experimental results from collaborators at the Technion in Israel that overturn this intuition, showing how cracks in silicon can propagate very slowly via intrinsically 3D kink mechanisms. Molecular Dynamics with On-the-fly Machine Learning of Quantum Mechanical Forces. In an article published in Physical Review Letters, we report a new molecular dynamics scheme which combines first-principles molecular dynamics and machine-learning (ML) techniques in a single information-efficient approach. The scheme works by going “shopping” in a database of reference configurations whenever QM forces are required. If there’s something similar enough, it interpolates using Bayesian inference to predict forces. If not, a new QM calculation is carried out on the fly to extend the database (Read more). Scattering of cracks by individual atomic-scale impurities. In an article published in Nature Communications, we showed that a single atomic defect (e.g. a boron dopant, coloured orange in movie, above right) can be enough to deflect a crack as it travels through a crystal, leading to macroscopically observable surface features (centre right). Stress corrosion crack propagation in a chemically aggressive environment. Our joint experimental/theoretical work published in Physical Review Letters showed that oxygen molecules (coloured red, below right) arriving at a crack tip dissociate spontaneously, releasing enough energy to "burn" individual silicon bonds. This work benefited from access to the Argonne Leadership Computing Facility. Low speed fracture instabilities. In an earlier work published in Nature, we discovered that the (111) cleavage plane in silicon becomes unstable at speeds below about 1000 m/s. The instability arises from an atomic scale reconstruction of the crack tip which can only be accurately modelled with quantum mechanical precision.

Selected Publications

1. P. Grigorev, T. Swinburne and J. R. Kermode, QM/MM study of hydrogen-decorated screw dislocations in tungsten: ultrafast pipe diffusion, core reconstruction and effects on glide mechanism,
   Phys. Rev. Materials 4 023601 (2020) [Open access]
2. S. Makri, C. Ortner and J. R. Kermode, A preconditioning scheme for Minimum Energy Path finding methods, J. Chem. Phys 150, 094109 (2019) [arXiv:1810.02705
3. A. P. Bartók, J. R. Kermode, N. Bernstein, and G. Csányi, Machine Learning a General-Purpose Interatomic Potential for Silicon, Phys. Rev. X 8, 041048 (2018) [arXiv] [Open Access]
4. A. P. Bartok, S. De, C. Poelking, N. Bernstein, J. R. Kermode, G. Csányi and M. Ceriotti, Machine learning unifies the modeling of materials and molecules. Science Advances 3, e1701816 (2017). [arXiv] [Open Access]
5. T. D. Swinburne and J. R. Kermode, Computing energy barriers for rare events from hybrid quantum/classical simulations through the virtual work principle, Phys. Rev. B 96, 144102 (2017). [arXiv] [Open Access]
6. G. Sernicola, T. Giovannini, P. Patel, J. R. Kermode, D. S. Balint, T. Ben Britton, and F. Giuliani, In situ stable crack growth at the micron scale, Nat. Commun. 8, 108 (2017). [Open Access]
7. F. Bianchini, J.R. Kermode, and A. De Vita, Modelling defects in Ni–Al with EAM and DFT calculations, Modell. Simul. Mater. Sci. Eng. 24, 045012 (2016) [Open Access]
8. D. Packwood, J. R. Kermode, L. Mones, N. Bernstein, J. Woolley, N. Gould, C. Ortner, and G. Csányi, A universal preconditioner for simulating condensed phase materials, J. Chem. Phys. 144, 164109 (2016).
   [arXiv] [Open Access]
9. M. Aldegunde, J. R. Kermode, and N. Zabaras, Development of an exchange–correlation functional with uncertainty quantification capabilities for density functional theory, J. Comput. Phys. 311, 173 (2016).
10. J. R. Kermode, A. Gleizer, G. Kovel, L. Pastewka, G. Csányi, D. Sherman, and A. De Vita, Low Speed Crack Propagation via Kink Formation and Advance on the Silicon (110) Cleavage Plane, Phys. Rev. Lett. 115, 135501 (2015) [Open Access]
11. Z. Li, J. R. Kermode, and A. De Vita, Molecular Dynamics with On-the-Fly Machine Learning of Quantum-Mechanical Forces, Phys. Rev. Lett. 114, 096405 (2015) [Open Access]
12. E. Bitzek, J. R. Kermode and P. Gumbsch, Atomistic aspects of fracture, Int. J. Fract. 191, 13-30 (2015)
    [Open Access]
13. A. Gleizer, G. Peralta, J. R. Kermode, A. De Vita and D. Sherman, Dissociative Chemisorption of O2 Inducing Stress Corrosion Cracking in Silicon Crystals. Phys. Rev. Lett. 112, 115501 (2014).
14. J.R. Kermode, L. Ben-Bashat, F. Atrash, J.J. Cilliers, D. Sherman and A. De Vita, Macroscopic scattering of cracks initiated at single impurity atoms. Nat. Commun. 4, 2441 (2013).
15. N. Bernstein, J. R. Kermode and G. Csányi, Hybrid atomistic simulation methods for materials systems. Rep. Prog. Phys. 72, 026501 (2009).
16. J. R. Kermode, T. Albaret, D. Sherman, N. Bernstein, P. Gumbsch, M. C. Payne, G. Csányi and A. De Vita, Low speed fracture instabilities in a brittle crystal, Nature 455, 1224-1227 (2008).

See also my full Publications page, my Talks page, and my profiles on ORCID, Google Scholar and the Warwick Research Archive Portal. My PhD Thesis is available from the University of Cambridge's repository.

Current Research Projects

• EPSRC grant Boundary Conditions for Atomistic Simulation of Material Defects with C. Ortner (PI, Warwick), R. Catlow (Co-I, UCL) and Researcher Co-Is J. Braun (Warwick) and A. Sokol (UCL).
• Leverhulme Research Project Grant proposal The Nature of Interatomic Forces in Metallic Systems with Christoph Ortner.
• EPSRC grant Reducing Risk through Uncertainty Quantification for Past, Present and Future Generations of Nuclear Power Plant in collaboration with Manchester and Bristol
• I am a member of the UK Car Parinello Consortium (UKCP), which has EPSRC funding (2017-2021) to provide access to the UK national supercomputer ARCHER.

Previous Research Projects

• Hydrogen Embrittlement of Steels - I was an external associate of this EPSRC Programme Grant involving Oxford, Cambridge, Imperial, King's and Sheffield.
• My EPSRC First Grant Predictive Modelling of the Fundamentals of Failure in Metals (Aug 2016 - July 2018) had the overarching aim of developing new models to enable continuum-scale modelling of failure processes, in particular crack growth, by incorporating pre-computed first-principles information. This funding supported postdoctoral researcher Dr Petr Grigorev.
• The Novel Materials Discovery (NoMaD) laboratory, funded under the Horizon 2020 Centre of Excellence initiative, enables access to the huge amount of data routinely produced by computational materials science calculations through the online NoMaD repository. I was a senior researcher within the project, which is led by the Fritz Haber Institute and involves a consortium of 11 organisations throughout Europe. This project supported postdoctoral reseaarcher Dr Berk Onat.
• I held a Royal Society Research Project Grant (April 2017-Mar 2018) Bridging from Quantum Mechanical to Continuum Mechanical Models of Fracture, which provided funding for local computational resources.
• ARCHER eCSE project 11-7 (Sept 2017 - Feb 2018) supported postdoc Dr Letif Mones to implement preconditioned geometry optimisers in the CASTEP and ONETEP codes. Co-investigators were Dr Nick Hine and Prof. Christoph Ortner (Warwick) and Prof. Matt Probert (York).
• SiO₂ Fracture: Chemomechanics with a Machine-Learning Hybrid QM/MM Scheme - project allocated 126 million core hours in 2016 on the Mira machine at the Argonne Leadership Computer Facility under the US DoE's INCITE programme to investigate stress corrosion cracking in silica. Co-investigators were Anatole von Lilienfeld and Alessandro De Vita.
• Multiscale Atomistic Simulation of the Mechanical Behaviour of Nickel-based Superalloys. Project allocated 20 million core hours at the Cineca and Jülich supercomputer centres. Other project members were Federico Bianchini, Alessio Comisso and Alessandro De Vita (KCL).

Software

• I’m one of the authors of the libAtoms/QUIP molecular dynamics framework, available from our public github repository, and licensed under the GNU GPLv2.
• matscipy, a Python software library for computational materials science developed with Lars Pastewka.
• f90wrap, a utility for wrapping Fortran 95 code to make it accessible from Python, including support for derived types.
• Other software is available from my GitHub page, including an enhanced version of the AtomEye atomistic visualisation software which can read Extended XYZ (.xyz) and NetCDF (.nc) files (Note: the extended XYZ format is now also supported directly by OVITO and ASE).

External Collaborators

• Noam Bernstein (Naval Research Laboratory, Washington DC, USA)
• Gábor Csányi (Cambridge, UK)
• Peter Gumbsch (Fraunhofer IWM, Germany)
• Gianpietro Moras (Fraunhofer IWM, Germany)
• Lars Pastewka (Freiburg, Germany)
• Mike Payne (Cambridge, UK)
• Dov Sherman (Tel Aviv, Israel)

Warwick Centre
for Predictive Modelling
School of Engineering
University of Warwick
Coventry
CV4 7AL
United Kingdom

Office: D210
Phone: +44 (0) 24 765 28614
E-mail: j.r.kermode@warwick.ac.uk