Virtual Orbital Fragmentation Reduces Qubit Requirements by 40, 66% for Accurate Quantum Computing of Molecular Simulations

The demand for increasingly accurate molecular simulations consistently challenges the limitations of current quantum computers, which struggle with the large number of qubits required for complex calculations. Federico Zahariev and Mark S. Gordon, from Iowa State University, address this issue with a new method called Virtual Orbital Fragmentation, a technique that significantly reduces the number of qubits needed without sacrificing chemical accuracy. Their approach systematically breaks down the virtual orbital space into chemically meaningful fragments, achieving a reduction in qubit requirements of between 40 and 66 percent. Demonstrating the method’s effectiveness on several molecules, the researchers show that even a modest expansion provides highly accurate results, and when combined with existing techniques, achieves near-perfect agreement with full calculations, paving the way for practical chemical simulations on today’s quantum processors.

Fragmenting Molecules for Efficient Quantum Chemistry

Scientists are tackling a major obstacle in quantum chemistry, the limited number of qubits available on current quantum processors, by developing innovative fragmentation techniques. This research focuses on breaking down complex molecular simulations into smaller, more manageable calculations, enabling more realistic and complex simulations on near-term quantum devices. The core idea involves fragmenting molecules both spatially and electronically, combined with the Variational Quantum Eigensolver (VQE) algorithm, to reduce computational cost without sacrificing accuracy. The qubit bottleneck arises because simulating even moderately sized molecules requires a large number of qubits, exceeding the capabilities of current quantum hardware.

To address this, researchers employ Fragment Molecular Orbital (FMO) theory, a well-established classical method that divides a large molecule into smaller fragments for individual calculation. This work introduces Virtual Orbital Fragmentation (VOF), a novel approach that extends fragmentation to the electronic structure itself, breaking down the electronic orbitals into fragments and further reducing qubit requirements. Combining spatial and orbital fragmentation creates a hierarchical approach, providing multiplicative computational advantages and addressing the qubit bottleneck from multiple angles. This research demonstrates that VOF significantly reduces the number of qubits needed for VQE calculations, making it possible to simulate larger molecules on current quantum devices.

The hierarchical fragmentation approach is scalable, meaning it can be applied to even larger molecular systems while maintaining accuracy. The team leverages established methods like Boys’ perturbation theory and Pipek’s fragment-based approach as foundational to their work, integrating classical and quantum computation for optimal performance. In essence, this research presents a promising pathway to overcome the limitations of current quantum hardware and enable more realistic and complex quantum chemistry simulations. This addresses a critical limitation of current Noisy Intermediate-Scale Quantum (NISQ) devices, paving the way for more complex chemical calculations. The method achieves a 40 to 66% reduction in qubit requirements while maintaining chemical accuracy. The core of this advancement lies in partitioning the virtual orbital space into chemically intuitive fragments, analogous to established spatial fragmentation methods, but avoiding complications with chemical bonding.

Experiments demonstrate that a two-body expansion using VOF achieves errors below 3 kcal/mol, while a three-body expansion delivers sub-kcal/mol accuracy. When integrated with the Variational Eigensolver (VQE) and combined with the Effective Fragment Molecular Orbital (EFMO) method, the hierarchical approach achieves 96 to 100% accuracy relative to full calculations. This breakthrough translates directly into practical benefits for quantum computing, bringing systems that would typically require 96 to 128 qubits down to 48 to 74 qubits, well within the capabilities of current quantum processors. Researchers found that the method consistently reduces maximum qubit requirements by 40 to 66% across diverse molecular systems while maintaining chemical accuracy.

For multi-molecular systems, the hierarchical approach offers multiplicative computational advantages by fragmenting along orthogonal dimensions, enabling accurate calculations on molecular clusters that would otherwise be intractable. The team validated the method using systems including water and glycine, demonstrating its broad applicability and robustness. The team’s approach leverages the observation that virtual orbitals, comprising 70 to 90% of the total orbital space, contribute primarily to electron correlation effects, allowing for targeted reduction without sacrificing chemical accuracy. Researchers implemented the EFMO method, which incorporates the Effective Fragment Potential to describe long-range interactions, enabling accurate modeling of many-body polarization effects. By partitioning the virtual orbital space into chemically relevant fragments and applying a many-body expansion technique, the method achieves a 40 to 66% reduction in qubit requirements while maintaining chemical accuracy. Results demonstrate that a three-body expansion delivers sub-kcal/mol accuracy, comparable to full calculations. This work introduces a hierarchical approach, combining spatial and virtual orbital decomposition, which offers multiplicative computational advantages and enables calculations on larger molecular clusters than previously possible with current quantum hardware. The team successfully applied this method to molecular systems using 50 to 100 qubits, demonstrating its potential to address the limitations of Noisy Intermediate-Scale Quantum (NISQ) devices.