Classification of electronic structures and state preparation for quantum computation of reaction chemistry

Maximilian Mörchen, Guang Hao Low, Thomas Weymuth, Hongbin Liu, Matthias Troyer, Markus ReiherarXiv 2024 · DOI · arXiv

Abstract

Quantum computation for chemical problems will require the construction of guiding states with sufficient overlap with a target state. Since easily available and initializable mean-field states are characterized by an overlap that is reduced for multi-configurational electronic structures and even vanishes with growing system size, we here investigate the severity of state preparation for reaction chemistry. We emphasize weaknesses in current traditional approaches and highlight the advantage of quantum phase estimation algorithms. An important result is the introduction of a new classification scheme for electronic structures based on orbital entanglement information. We identify two categories of multi-configurational molecules. Whereas class-1 molecules are dominated by very few determinants and often found in reaction chemistry, class-2 molecules do not allow one to single out a reasonably sized number of important determinants. The latter are particularly hard for traditional approaches and an ultimate target for quantum computation. We discuss the role of the molecular orbital basis set and show that true class-2 molecules remain in this class independent of the choice of the orbital basis. Our key result is that it will always be possible to initialize a guiding state for chemical reaction chemistry in the ground state based on initial low-cost approximate electronic structure information, which is facilitated by the finite size of the atomistic structures to be considered.

The Challenge of State Preparation

To achieve high-precision energy calculations using Quantum Phase Estimation (QPE), a quantum computer requires a "guiding state"—an initial approximation with significant mathematical overlap with the target ground state. A major concern in the field is the "orthogonality catastrophe": as systems grow in size or complexity, the overlap between simple mean-field guesses (like Hartree-Fock states) and the true state can decay exponentially. If this overlap becomes too small, the quantum algorithm's success probability vanishes. This paper investigates whether this theoretical limit poses a practical barrier to simulating chemical reactions.

Quantifying Complexity via Orbital Entanglement

The authors propose a robust methodology for assessing electronic structure complexity using information theory. By analyzing single-orbital entropy, they measure the degree of electron correlation across different orbitals. This is visualized through threshold diagrams, which plot the number of selected orbitals against an entropy threshold.

To automate the selection of these critical orbitals (the "active space"), the study utilizes the AutoCAS algorithm. For generating the guiding states themselves, the researchers employ Density Matrix Renormalization Group (DMRG) to produce Matrix Product States (MPS). These classical approximations serve as high-fidelity starting points for subsequent quantum computation.

A New Taxonomy for Electronic Structures

A central contribution of this work is a classification scheme that categorizes molecules based on their inherent electronic complexity.

Threshold diagrams illustrating the classification of electronic structures into Class 0, 1, and 2.
  • Class 0 (Simple): These systems are dominated by a single electronic configuration. Water ($H_2O$) is a prototypical example. State preparation for these molecules is trivial.
  • Class 1 (Multi-configurational): These molecules are dominated by a few key configurations. They are common in transition states and reaction intermediates (e.g., cyclobutadiene, ozone). While more complex than Class 0, they remain manageable for state preparation.
  • Class 2 (Strongly Correlated): These systems lack a dominant configuration, making them the "ultimate targets" for quantum advantage. Examples include iron-sulfur clusters and the iron-molybdenum cofactor (FeMoco) of nitrogenase.

Key Insights and Results

The study reveals several critical findings that challenge the pessimistic view of state preparation:

  1. The Locality Loophole: Chemical reactions typically involve local changes within a limited number of atoms (usually 100–200). Because the physical size of the relevant system is finite, the asymptotic "orthogonality catastrophe" does not fully manifest in practical reaction chemistry.
  2. The Importance of Orbital Basis: The paper proves (Theorem 1) that "strong correlation" can often be an artifact of a poor orbital basis. By selecting appropriate localized orbitals, many seemingly complex problems simplify into manageable Class 1 systems.
  3. MPS as a Robust Bridge: Even for the most challenging Class 2 systems, approximate Matrix Product States—which can be computed with relatively low classical cost—provide sufficient overlap to serve as effective guiding states for QPE.
  4. FeMoco as a Benchmark: The analysis confirms that FeMoco is a "true" Class 2 system regardless of the orbital basis, solidifying its role as a primary target for demonstrating quantum utility.

Conclusion

The research demonstrates that initializing a guiding state for chemical reaction chemistry is feasible using low-cost approximate electronic structure information. By leveraging the finite size of atomistic structures and employing sophisticated classical approximations like MPS, the "state preparation problem" can be effectively navigated, clearing a path for meaningful quantum advantage in chemistry.

Summary updated

Abstract

Quantum computation for chemical problems will require the construction of guiding states with sufficient overlap with a target state. Since easily available and initializable mean-field states are characterized by an overlap that is reduced for multi-configurational electronic structures and even vanishes with growing system size, we here investigate the severity of state preparation for reaction chemistry. We emphasize weaknesses in current traditional approaches (even for weakly correlated molecules) and highlight the advantage of quantum phase estimation algorithms. An important result is the introduction of a new classification scheme for electronic structures based on orbital entanglement information. We identify two categories of multi-configurational molecules. Whereas class-1 molecules are dominated by very few determinants and often found in reaction chemistry, class-2 molecules do not allow one to single out a reasonably sized number of important determinants. The latter are particularly hard for traditional approaches and an ultimate target for quantum computation. Some open-shell iron-sulfur clusters belong to class 2. We discuss the role of the molecular orbital basis set and show that true class-2 molecules remain in this class independent of the choice of the orbital basis, with the iron-molybdenum cofactor of nitrogenase being a prototypical example. We stress that class-2 molecules can be build in a systematic fashion from open-shell centers or unsaturated carbon atoms. Our key result is that it will always be possible to initialize a guiding state for chemical reaction chemistry in the ground state based on initial low-cost approximate electronic structure information, which is facilitated by the finite size of the atomistic structures to be considered.
BibTeX
@misc{morchen2024classification,
  author = {Maximilian Mörchen and Guang Hao Low and Thomas Weymuth and Hongbin Liu and Matthias Troyer and Markus Reiher},
  title = {{Classification of electronic structures and state preparation for quantum computation of reaction chemistry}},
  year = {2024},
  eprint = {2409.08910},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2409.08910},
}
Figure