Public Abstract

DE-SC0019432: Q4Q: Quantum Computation for Quantum Prediction of Materials and Molecular Properties

Award Status: Inactive

Institution: University of Southern California, Los Angeles, CA
UEI: G88KLJR3KYT5
DUNS: 072933393

Most Recent Award Date: 08/15/2022
Number of Support Periods: 4
PM: Holder, Aaron

Current Budget Period: 09/15/2021 - 09/14/2023
Current Project Period: 09/15/2021 - 09/14/2023
PI: Di Felice, Rosa

Supplement Budget Period: N/A

Public Abstract

Q4Q: Quantum Computation for Quantum Prediction of Materials and Molecular Properties Rosa Di Felice, University of Southern California (Principal Investigator) Anna Krylov, University of Southern California (Co-Investigator) Itay Hen, University of Southern California (Co-Investigator) Amir Kalev, University of Southern California (senior personnel) Marco Fornari , Central Michigan University (Co-Investigator) Marco Buongiorno Nardelli, University of North Texas (Co-Investigator) The last 40 years have been filled with investigations seeking to demonstrate that the manipulation of quantum states of physical systems can be leveraged to carry out efficiently certain computations that standard computers are known to struggle with. Towards this effort, theorists working in quantum physics and quantum information/computation made impressive progress in developing quantum algorithms with a scaling advantage with respect to classical algorithms. While the first experimental realization of qubits and logical quantum gates appeared in 1998, for many years the availability of functional qubits was limited to research laboratories and a few computational units. This prevented the coding of practical problems, from, e.g., materials science and chemistry. Furthermore, there was no way to assess the performance of quantum algorithms and their potential advantage. The situation changed abruptly when quantum annealers were commercialized in 2010 (D-Wave Systems Inc., Canada). Then in 2016, the cloud enabled users to reach new frontiers when gate model quantum computers became available. These advances reignited the interest in quantum computing and stimulated the search for useful applications of quantum computation. Breakthrough applications were identified in the quantum simulation of physical/chemical systems. Our proposal is focused on formulating and solving electronic structure and statistical problems for available noisy intermediate-scale quantum (NISQ) devices. This is necessary to bring further technological progress that will impact society. We will use and benchmark existing algorithms and quantum software packages, as well as develop new algorithms. We propose computational research activities that will advance the application of quantum computers to selected materials/molecular properties: (1) Electronic structure of solids; (2) Molecular electronic/vibronic properties; (3) Exploiting the excited states of quantum annealing for phase transitions and machine learning in materials. We will extend the variational quantum eigensolver (VQE) hybrid algorithm to treat extended systems like solids, especially to describe the periodic nature of the system’s wave function. We will improve the VQE to solve the ground state of molecular Hamiltonians and apply techniques of complexity reduction to minimize the number of qubits needed for relevant electronic structure problems. We will develop a new quantum chemistry plugin to quantum software designed for electronic structure calculations of molecules and materials. We will exploit machine learning strategies to compute phase diagrams of materials. We note that, while quantum simulations of small molecules on quantum computers have gained popularity, quantum simulations of periodic systems are still elusive, and these deserve a significant emphasis in this proposal. This proposal is potentially transformative as it will extend the application of quantum computers to periodic solids and to molecular properties that go beyond the ground state electronic structure. With these breakthroughs, quantum computations become feasible for molecules of unprecedented complexity through the development of theoretical schemes that reduce the complexity by down-folding many body problems (e.g., via effective Hamiltonians). Finally, this work will produce new updated versions of quantum software that will be available to the scientific community. Quantum simulations have the potential to give unprecedented accuracy in the prediction of electronic and vibrational spectroscopies of molecules and solids, with enormous impact in energy science, drug design and pharmaceutical science, biology, materials manufacturing. Society will benefit from the production of new technologies and health strategies. In carrying out the proposed activities, we will train future generations of undergraduate students, graduate students and postdocs at the interface between different disciplines, supplementing their learning with other skills to thrive in the STEM pipeline.

Q4Q: Quantum Computation for Quantum Prediction of Materials and Molecular Properties

Rosa Di Felice, University of Southern California (Principal Investigator)

Anna Krylov, University of Southern California (Co-Investigator)

Itay Hen, University of Southern California (Co-Investigator)

Amir Kalev, University of Southern California (senior personnel)

Marco Fornari , Central Michigan University (Co-Investigator)

Marco Buongiorno Nardelli, University of North Texas (Co-Investigator)

The last 40 years have been filled with investigations seeking to demonstrate that the manipulation of quantum states of physical systems can be leveraged to carry out efficiently certain computations that standard computers are known to struggle with. Towards this effort, theorists working in quantum physics and quantum information/computation made impressive progress in developing quantum algorithms with a scaling advantage with respect to classical algorithms. While the first experimental realization of qubits and logical quantum gates appeared in 1998, for many years the availability of functional qubits was limited to research laboratories and a few computational units. This prevented the coding of practical problems, from, e.g., materials science and chemistry. Furthermore, there was no way to assess the performance of quantum algorithms and their potential advantage. The situation changed abruptly when quantum annealers were commercialized in 2010 (D-Wave Systems Inc., Canada). Then in 2016, the cloud enabled users to reach new frontiers when gate model quantum computers became available. These advances reignited the interest in quantum computing and stimulated the search for useful applications of quantum computation. Breakthrough applications were identified in the quantum simulation of physical/chemical systems. Our proposal is focused on formulating and solving electronic structure and statistical problems for available noisy intermediate-scale quantum (NISQ) devices. This is necessary to bring further technological progress that will impact society. We will use and benchmark existing algorithms and quantum software packages, as well as develop new algorithms.

We propose computational research activities that will advance the application of quantum computers to selected materials/molecular properties: (1) Electronic structure of solids; (2) Molecular electronic/vibronic properties; (3) Exploiting the excited states of quantum annealing for phase transitions and machine learning in materials. We will extend the variational quantum eigensolver (VQE) hybrid algorithm to treat extended systems like solids, especially to describe the periodic nature of the system’s wave function. We will improve the VQE to solve the ground state of molecular Hamiltonians and apply techniques of complexity reduction to minimize the number of qubits needed for relevant electronic structure problems. We will develop a new quantum chemistry plugin to quantum software designed for electronic structure calculations of molecules and materials. We will exploit machine learning strategies to compute phase diagrams of materials. We note that, while quantum simulations of small molecules on quantum computers have gained popularity, quantum simulations of periodic systems are still elusive, and these deserve a significant emphasis in this proposal.

This proposal is potentially transformative as it will extend the application of quantum computers to periodic solids and to molecular properties that go beyond the ground state electronic structure. With these breakthroughs, quantum computations become feasible for molecules of unprecedented complexity through the development of theoretical schemes that reduce the complexity by down-folding many body problems (e.g., via effective Hamiltonians). Finally, this work will produce new updated versions of quantum software that will be available to the scientific community. Quantum simulations have the potential to give unprecedented accuracy in the prediction of electronic and vibrational spectroscopies of molecules and solids, with enormous impact in energy science, drug design and pharmaceutical science, biology, materials manufacturing. Society will benefit from the production of new technologies and health strategies. In carrying out the proposed activities, we will train future generations of undergraduate students, graduate students and postdocs at the interface between different disciplines, supplementing their learning with other skills to thrive in the STEM pipeline.