Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Quantum Simulations of Chemical Reactions: Achieving Accuracy with NISQ Devices

Maitreyee Sarkar, Lisa Roy, Akash Gutal, Atul Kumar, Manikandan Paranjothy

TL;DR

This work tackles the challenge of obtaining chemical-accuracy reaction energies on NISQ devices using VQE by introducing a symmetry-informed active-space framework. A group-theory based symmetry-matched fraction (SMF) guides the selection of a single optimal active-space for each reactant and product, reducing combinatorial search while preserving accuracy. Across five industrially relevant reactions, SMF-guided VQE energies differ from CCSD references by less than 1 kcal/mol, with some cases reaching precision on the order of 10^-5 to 10^-3 kcal/mol relative to CCSD. The approach extends symmetry usage beyond circuit optimization, offering a scalable, generalizable path toward quantum-enabled reaction modeling on near-term hardware, and points to future work on larger basis sets and noise-mitigation techniques.

Abstract

Quantum computing is viewed as a promising technology because of its potential for polynomial growth in complexity, in contrast to the exponential growth observed in its classical counterparts. In the current Noisy Intermediate-Scale Quantum (NISQ) era, the Variational Quantum Eigensolver (VQE), a hybrid variational algorithm, is utilized to simulate molecules using qubits and calculate molecular properties. However, simulating a chemical reaction to compute the reaction energy using VQE algorithm has not yet reached chemical accuracy relative to the benchmark computational chemistry methods due to limitations such as the number of qubits, circuit depth, and noise introduced within the model. To address this issue, we propose the definition of different active spaces for studying chemical reactions, incorporating irreducible representations of both the ground and excited states of the molecules. Our results demonstrate that this approach achieves chemical accuracy in predicting the reaction energy for various reactions. For all reactions studied, the difference in reaction energies between conventional computational chemistry methods and the quantum-classical hybrid VQE algorithm is less than 1 kcal/mol. Furthermore, our analysis simplifies the process of selecting active spaces and electrons for each reaction, reducing it to a single optimal combination that ensures the chemical accuracy for each reaction.

Quantum Simulations of Chemical Reactions: Achieving Accuracy with NISQ Devices

TL;DR

This work tackles the challenge of obtaining chemical-accuracy reaction energies on NISQ devices using VQE by introducing a symmetry-informed active-space framework. A group-theory based symmetry-matched fraction (SMF) guides the selection of a single optimal active-space for each reactant and product, reducing combinatorial search while preserving accuracy. Across five industrially relevant reactions, SMF-guided VQE energies differ from CCSD references by less than 1 kcal/mol, with some cases reaching precision on the order of 10^-5 to 10^-3 kcal/mol relative to CCSD. The approach extends symmetry usage beyond circuit optimization, offering a scalable, generalizable path toward quantum-enabled reaction modeling on near-term hardware, and points to future work on larger basis sets and noise-mitigation techniques.

Abstract

Quantum computing is viewed as a promising technology because of its potential for polynomial growth in complexity, in contrast to the exponential growth observed in its classical counterparts. In the current Noisy Intermediate-Scale Quantum (NISQ) era, the Variational Quantum Eigensolver (VQE), a hybrid variational algorithm, is utilized to simulate molecules using qubits and calculate molecular properties. However, simulating a chemical reaction to compute the reaction energy using VQE algorithm has not yet reached chemical accuracy relative to the benchmark computational chemistry methods due to limitations such as the number of qubits, circuit depth, and noise introduced within the model. To address this issue, we propose the definition of different active spaces for studying chemical reactions, incorporating irreducible representations of both the ground and excited states of the molecules. Our results demonstrate that this approach achieves chemical accuracy in predicting the reaction energy for various reactions. For all reactions studied, the difference in reaction energies between conventional computational chemistry methods and the quantum-classical hybrid VQE algorithm is less than 1 kcal/mol. Furthermore, our analysis simplifies the process of selecting active spaces and electrons for each reaction, reducing it to a single optimal combination that ensures the chemical accuracy for each reaction.

Paper Structure

This paper contains 20 sections, 8 equations, 11 figures, 1 table.

Table of Contents

  1. Introduction
  2. Methods
  3. VQE Algorithm
  4. Hamiltonian
  5. Ansatz
  6. Classical optimization
  7. Group Theory
  8. Determination of Reaction Energies
  9. Computational Details
  10. Computational Chemistry Methods
  11. Quantum Computation Methods
  12. Mathematical Details of Group Theory
  13. Results
  14. H2 + F2 $\rightarrow$ 2HF
  15. H2 + Cl2 $\rightarrow$ 2HCl
  16. ...and 5 more sections

Figures (11)

  • Figure 1: Example representation of a random single term ($e^{i\frac{\theta}{2}(Z_1Z_2X_3Z_4)}$) of UCCSD ansatz in as a circuit fragment
  • Figure 2: Different descriptions of H$_2$ molecule.
  • Figure 3: Energy and SMF values for reactants, products and reaction energies of Reaction 1.
  • Figure 4: MO of HF molecule (image is not upto scale)
  • Figure 5: SMF values for reactants and products of Reaction 2.
  • ...and 6 more figures