Project

Back to overview

Sampling chemical space with alchemical perturbation theory

English title Sampling chemical space with alchemical perturbation theory
Applicant von Lilienfeld-Toal Otto Anatole
Number 175747
Funding scheme Project funding
Research institution Physikalische Chemie Departement Chemie Universität Basel
Institution of higher education University of Basel - BS
Main discipline Physical Chemistry
Start/End 01.10.2017 - 31.05.2021
Approved amount 600'000.00
Show all

All Disciplines (9)

Discipline
Physical Chemistry
Molecular Biology
Chemical Engineering
Biochemistry
Information Technology
Condensed Matter Physics
Theoretical Physics
Mathematics
Biophysics

Keywords (4)

chemical space; density functional theory; perturbation theory; computational alchemy

Lay Summary (German)

Lead
Computational alchemy for chemical space exploration
Lay summary
Viel Forschung dreht sich um das theoretische Studium des Zusammenhanges zwischen Materialien und Molekuelen und ihrer chemischen Zusammensetzung und Struktur. Basierend auf Quantenmechanik, Statistischer Mechanik, und Computerwissenschaften ermoeglichen es heutzutage computergestuetzte Methoden in der Chemie die Routinevorhersage vieler relevanter Eigenschaften fuer neue Kombinationen von atomaren Konfigurationen und Zusammensetzung. In diesem Projekt geht es um den methodischen Fortschritt, der noetig ist, um die effizientere virtuelle Exploration und Erprobung des chemischen Raumes mit Hilfe des Konzeptes der "alchemischen" Interpolation zwischen verschiedenen chemischen Verbindungen zu ermoeglichen. Ein erfolgreiches Projekt wird es ermoeglichen, dass man Eigenschaften von tausenden, wenn nicht sogar millionen, von chemischen Verbindungen mit vernachlaessigbarem Rechenaufwand und mit hoher Genauigkeit voraussagen kann. So eine Methode kann hilfreich fuer die schnelle Entdeckung und Identifizierung von Molekuelen und Materialien mit interessanten Eigenschaften sein.
Direct link to Lay Summary Last update: 29.09.2017

Responsible applicant and co-applicants

Employees

Publications

Publication
Simplifying inverse materials design problems for fixed lattices with alchemical chirality
von Rudorff Guido Falk, von Lilienfeld O. Anatole (2021), Simplifying inverse materials design problems for fixed lattices with alchemical chirality, in Science Advances, 7(21), eabf1173-eabf1173.
Thousands of reactants and transition states for competing E2 and S$_\mathrm{N}$2 reactions
von Rudorff Guido Falk, Heinen Stefan N, Bragato Marco, von Lilienfeld O Anatole (2020), Thousands of reactants and transition states for competing E2 and S$_\mathrm{N}$2 reactions, in Machine Learning: Science and Technology, 1(4), 045026-045026.
Data enhanced Hammett-equation: reaction barriers in chemical space
Bragato Marco, von Rudorff Guido Falk, von Lilienfeld O. Anatole (2020), Data enhanced Hammett-equation: reaction barriers in chemical space, in Chemical Science, 11(43), 11859-11868.
Machine Learning Models of Vibrating H 2 CO: Comparing Reproducing Kernels, FCHL, and PhysNet
Käser Silvan, Koner Debasish, Christensen Anders S., von Lilienfeld O. Anatole, Meuwly Markus (2020), Machine Learning Models of Vibrating H 2 CO: Comparing Reproducing Kernels, FCHL, and PhysNet, in The Journal of Physical Chemistry A, 124(42), 8853-8865.
Effects of perturbation order and basis set on alchemical predictions
Domenichini Giorgio, von Rudorff Guido Falk, von Lilienfeld O. Anatole (2020), Effects of perturbation order and basis set on alchemical predictions, in The Journal of Chemical Physics, 153(14), 144118-144118.
Machine learning the computational cost of quantum chemistry
Heinen Stefan, Schwilk Max, von Rudorff Guido Falk, von Lilienfeld O Anatole (2020), Machine learning the computational cost of quantum chemistry, in Machine Learning: Science and Technology, 1(2), 025002-025002.
Alchemical perturbation density functional theory
von Rudorff Guido Falk, von Lilienfeld O. Anatole (2020), Alchemical perturbation density functional theory, in Physical Review Research, 2(2), 023220-023220.
Rapid and accurate molecular deprotonation energies from quantum alchemy
von Rudorff Guido Falk, von Lilienfeld O. Anatole (2020), Rapid and accurate molecular deprotonation energies from quantum alchemy, in Physical Chemistry Chemical Physics, 22(19), 10519-10525.
Atoms in Molecules from Alchemical Perturbation Density Functional Theory
von Rudorff Guido Falk, von Lilienfeld O. Anatole (2019), Atoms in Molecules from Alchemical Perturbation Density Functional Theory, in The Journal of Physical Chemistry B, 123(47), 10073-10082.
Boosting Quantum Machine Learning Models with a Multilevel Combination Technique: Pople Diagrams Revisited
Zaspel Peter, Huang Bing, Harbrecht Helmut, von Lilienfeld O. Anatole (2019), Boosting Quantum Machine Learning Models with a Multilevel Combination Technique: Pople Diagrams Revisited, in Journal of Chemical Theory and Computation, 15(3), 1546-1559.
Alchemical Normal Modes Unify Chemical Space
Fias Stijn, Chang K. Y. Samuel, von Lilienfeld O. Anatole (2018), Alchemical Normal Modes Unify Chemical Space, in The Journal of Physical Chemistry Letters, 10(1), 30-39.
AlxGa1−xAs crystals with direct 2 eV band gaps from computational alchemy
Chang K. Y. Samuel, von Lilienfeld O. Anatole (2018), AlxGa1−xAs crystals with direct 2 eV band gaps from computational alchemy, in Physical Review Materials, 2(7), 073802-073802.
Quantum Machine Learning in Chemical Compound Space
von Lilienfeld O. Anatole (2018), Quantum Machine Learning in Chemical Compound Space, 57(16), 4164-4169, Wiley, Germany 57(16), 4164-4169.

Scientific events

Active participation

Title Type of contribution Title of article or contribution Date Place Persons involved
MRS Fall Meeting Talk given at a conference Alchemical Perturbation Density Functional Theory (APDFT) 02.12.2019 Boston, United States of America von Rudorff Guido Falk;
Machine Learning for Physics and the Physics of Learning Talk given at a conference Chemical Space Quantum Machine Learning 09.09.2019 UCLA, Los Angeles, United States of America von Lilienfeld-Toal Otto Anatole;
SCS Fall Meeting Talk given at a conference Alchemical Perturbation Density Functional Theory (APDFT) 06.09.2019 Zurich, Switzerland von Rudorff Guido Falk;
18th International Conference on Density-Functional Theory and its Applications Talk given at a conference Alchemical perturbation density functional theory 26.07.2019 Alicante, Spain von Rudorff Guido Falk;
18th International Conference on Density-Functional Theory and its Applications Talk given at a conference Chemical space from alchemical perturbation DFT and quantum machine learning 24.07.2019 Alicante, Spain von Rudorff Guido Falk; von Lilienfeld-Toal Otto Anatole;
Thomas Young Center Seminar Individual talk Alchemical Perturbation Density Functional Theory (APDFT) 14.12.2018 London, Great Britain and Northern Ireland von Rudorff Guido Falk;
Seminar Departement of Chemistry Individual talk Quantum Machine Learning 08.11.2017 Department of Chemistry, University of Cambrigde, Great Britain and Northern Ireland von Lilienfeld-Toal Otto Anatole;


Awards

Title Year
Foresight Institute Feynman prize in Nanotechnology Theory (https://foresight.org/about/2018Feynman.html) 2018

Associated projects

Number Title Start Funding scheme
160067 Multiscale dynamics of dog rabies elimination 01.08.2015 Project funding

Abstract

Much research is devoted to the study of the relationship between the properties of materials and their chemical composition and structure. Rigorously rooted in quantum mechanics (QM), statistical mechanics, and heavy computing, employed methods nowadays permit to routinely predict relevant properties for novel combinations of atomic configurations and composition, effectively sampling chemical compound space (CCS). The task of using QM for the discovery of novel compounds with improved properties is therefore equivalent to a combinatorial optimization problem in CCS. Albeit straight-forward, tackling this task through QM based high-throughput screening is the least efficient way. This proposal is about the advancement of methods which enable the exploration and sampling of CCS in dramatically more efficient ways, without loss of accuracy. In particular, we plan to use ``alchemical'' perturbation theory which leverages differential, response-like, QM calculations of reference molecules and materials. Results for these perturbation theoretical calculations can subsequently be usedto accurately estimate properties of thousands, if not millions, new molecules and materials ``close by'' in CCS with negligible, or constant, computational overhead. Building on recent contributions, we propose three sub projects: (1) to investigate alchemical second and higher order perturbations to afford potential energy estimates of molecules with unprecedented speed and accuracy. (2) to use alchemical estimates for the tailoring of electronic properties in novel materials (important for renewable energy applications),and (3) to make accurate alchemical estimates of chemical reaction energy profiles (important for improved reaction conditions or catalysts).
-