Chemistry and Mathematics in Phase Space (CHAMPS)

Lead Research Organisation: University of Bristol
Department Name: Mathematics

Abstract

The 6 year CHAMPS Programme Grant (PG) addresses the urgent need to provide a framework for understanding and exploiting the explosion in dynamical information coming out of modern experiments and simulations in chemistry and chemical biology. By developing methods in nonlinear dynamics to replace the ubiquitous configuration-space projections of these inherently multidimensional datasets with phase-space representations, it will be possible to provide comprehensible models that capture the key dynamics of complex systems. These new models will revolutionise our understanding of chemical transformation, with impacts on all industries that rely on understanding chemical change, from the pharmaceutical industries to those in the rapidly developing energy sector. By its very nature, the research will have impact on a wide variety of EPSRC research themes, including health care, energy, and environmental change. And because the work requires an unprecedented partnership between mathematicians and chemists, it will bridge several themes that EPSRC has highlighted as critical to its strategy. The young scientists trained in this programme will have unmatched interdisciplinary skills, and provide UK research leadership for decades to come. In particular, we expect that they will be creating their own new fields of study, and will thereby carry on attracting the best young scientists from around the world to come to the UK. A PG is vital to provide the critical mass, flexibility and scope needed to tackle the major scientific cross-disciplinary challenges set out in this proposal. The alternative funding model of individual grants to the PI and CoIs would leave the PDRAs localised in particular research groups and thereby maintain the traditional disciplinary separation, which we are seeking to break down.

Planned Impact

We are seeking to correct deficiencies in the very foundation of the science of chemistry. Success in this enterprise will have impact across all of chemistry and allied disciplines. Some practical consequences of this work may be forthcoming quickly - and we have highlighted one that has already emerged - but we anticipate that the full effects of this work are likely to be felt only once we have made the results known and comprehensible to the vast array of scientists for whom they are relevant. Certainly high-impact publications in journals such as Science and Nature would go some way to achieving that goal, but any time one requires people to "unlearn" things that they think they already know, there is resistance and inertia to change. For that reason, we anticipate hosting a number of conferences and workshops to discuss our work, and to receive feedback from potential stakeholders on how we might address their specific needs. These gatherings will involve key individuals from both academia and private industry. Complementing this effort will be the placement of the PDRAs coming out of our research programme into academic and industrial positions where they can influence how chemistry is understood and conducted from the inside.

Publications

10 25 50
 
Description We have examined the roaming mechanism more deeply and quantified the influence of mass and potential energy surface geometry on roaming dynamics. We have studies more deeply caldera shaped potential energy surfaces and discovered the mechanism for trajectories to transit the caldera region as well as the phase space mechanisms responsible for dynamical matching. We have shown how the method of Lagrangian descriptors can be extended and used to discover high dimensional phase space structures from low dimensional slices. This methodology has been used to develop a reactive island theory for system-bath models. We have investigated a variety of potential energy surfaces relevant to post reaction transition state bifurcations that are relevant to numerous organic reactions. We have discovered a universal phase space mechanism for selectivity, i.e., computing branching ratios in such reactions. Progress has been made on virtual reality simulations of complex molecular systems. We have shown the applicability of machine learning for deducing reduced dimensional models of reaction dynamics and for discovering the phase space structures governing reaction dynamics by using trajectory data sets. Progress has been made on going beyond the Born-Oppenheimer approximation. This has resulted in a new theory of nonadiabatic quantum dynamics that goes beyond the Born-Huang formalism. With this theory we are able to describe excited state dynamics on multiple potential energy surfaces. This has led to novel computational techniques for trajectory based simulations of non-adiabatic molecular systems.
Exploitation Route Our work on the phase space description of nonadiabatic quantum mechanics should prove to be a key enabling technology for the many "quantum start-ups" that are currently being formed, or are in development, such as Quantemol. Our work on virtual reality based quantum chemistry will provide a vehicle for STEM students to obtain an introduction to a complex, but rapidly developing and important field of research and technological advancement.
Sectors Chemicals,Digital/Communication/Information Technologies (including Software),Education,Energy

URL https://champsproject.com/
 
Description The methods developed in CHAMPS are finding their way to industry. Currently we are working with the company Quantemol providing them with the data on dissociation channels of molecules after electron impact. This data is needed for modelling of plasma etching, a technological process in semiconductor industry, where chemical composition of plasma is of great interest. An application for Impact Acceleration Account has been submitted in Leeds to fund this work and joint application for UKRI Knowledge Transfer Partnership is in preparation. Virtual Reality with the NARUPA software developed by one of the CHAMPS participants (D. Glowacki) is currently extensively used in Leeds for outreach events, for example chemistry open days. The software allows children to submerge into molecular world and observe atomistic motion. They can also experience interaction with atoms and molecules in a way similar to a VR computer game. This stimulates their interest to science and hopefully increases the intake of mathematics and chemistry students in Leeds. CHAMPS funding has been used to build the VR kit in Leeds.
First Year Of Impact 2022
Sector Chemicals,Creative Economy,Digital/Communication/Information Technologies (including Software),Education
Impact Types Societal,Economic,Policy & public services

 
Title Computational data for the simulation of cases of the Spin Boson Model using the Multiconfigurational Ehrenfest method - dataset 
Description This data gives input and output files for simulations carried out using the Multiconfigurational Ehrenfest method to show the importance of sampling methods such as cloning and trains, and how the use of these techniques in the "Multiple Cloning" scheme can allow the MCEv2 method to generate results which are converged to benchmark calculations where previously this was not possible. Included also is the source code of the program used to generate these data files. 
Type Of Material Database/Collection of data 
Year Produced 2019 
Provided To Others? Yes  
 
Title Data associated with 'Floquet Hamiltonian for incorporating electronic excitation by a laser pulse into simulations of non-adiabatic dynamics' 
Description The dataset contains the spreadsheets, hi-resolution images and raw research data associated with our paper: 'Floquet Hamiltonian for incorporating electronic excitation by a laser pulse into simulations of non-adiabatic dynamics', Chem. Phys., 515 (2018) 46-51, DOI: https://doi.org/10.1016/j.chemphys.2018.07.048 
Type Of Material Database/Collection of data 
Year Produced 2019 
Provided To Others? Yes  
 
Title Data associated with 'Ultrafast Photodissociation Dynamics of 2-Ethylpyrrole: Adding Insight to Experiment With Ab Initio Multiple Cloning.' 
Description Ab initio multiple cloning calculated and experimental total kinetic energy release spectra, dissociation times, velocity map images and electronic state populations for the ultrafast photodissociation of 2-ethylpyrrole. 
Type Of Material Database/Collection of data 
Year Produced 2019 
Provided To Others? Yes  
 
Title Data from CHAMPS (02-2018) 
Description The data in this deposit were used in the paper "Empirical Classification of Trajectory Data: An Opportunity for the Use of Machine Learning in Molecular Dynamics." Carpenter, B. K.; Ezra, G. S.; Farantos, S. C.; Kramer, Z. C.; Wiggins, S. J. Phys. Chem. B, 2017, 121, in press, doi: 10.1021/acs.jpcb.7b08707. 
Type Of Material Database/Collection of data 
Year Produced 2018 
Provided To Others? Yes  
 
Title Data to support 'Simulation of protein pulling dynamics on second time scale with boxed molecular dynamics' 
Description This data set contains the all the figure and supporting data relating to the paper 'Simulation of protein pulling dynamics on second time scale with boxed molecular dynamics' (doi: 10.1063/5.0059321). Each data set, titled by the PMF used to generate it, is split into subdirectories for the velocities the code was run at (in Angstrom/ns) each of which contain the population, free energy and force outputs for the simulation. The simulations can be created again using the files in the code folder and each arrays2.cm contained in each PFM folder in which it is contained. Compile with ifort compiler for fortran 77. 
Type Of Material Database/Collection of data 
Year Produced 2021 
Provided To Others? Yes  
URL https://archive.researchdata.leeds.ac.uk/894/
 
Title LDDS: Python package for computing and visualizing Lagrangian Descriptors for Dynamical Systems 
Description The LDDS software is a Python-based module that provides the user with the capability of analyzing the phase space structures of both continuous and discrete nonlinear dynamical systems in the deterministic and stochastic settings using Lagrangian descriptors (LDs). 
Type Of Technology Software 
Year Produced 2021 
Open Source License? Yes  
Impact None yet to report. 
URL https://joss.theoj.org/papers/10.21105/joss.03482.pdf
 
Title UPOsHam: A Python package for computing unstable periodic orbits in two-degree-of-freedom Hamiltonian systems 
Description A Python package for computing unstable periodic orbits in two-degree-of-freedom Hamiltonian systems. 
Type Of Technology Software 
Year Produced 2020 
Open Source License? Yes  
Impact None yet to report. 
URL https://joss.theoj.org/papers/10.21105/joss.01684.pdf