Quantum chemistry calculations have long been touted as a key application for quantum computers, with the potential to revolutionize drug development and agriculture. However, a recent analysis suggests that this may not be the case after all.
The rapid progress in building quantum computers in recent years has raised hopes for solving complex problems in quantum chemistry, such as calculating the energy levels of molecules. This task involves considering the behavior of multiple quantum particles simultaneously, making it a seemingly ideal match for quantum computers.
Xavier Waintal and his team at CEA Grenoble in France have conducted a thorough analysis of two leading quantum computing algorithms for molecular energy calculations. Their findings cast doubt on the feasibility of using quantum computers for this purpose.
The team’s analysis focused on two scenarios: one involving existing quantum computers, which are prone to errors, and another involving future fault-tolerant quantum computers. When using error-prone quantum computers, the variational quantum eigensolver (VQE) algorithm can be used to calculate molecular energy levels. However, the accuracy of the results is heavily dependent on the level of noise in the quantum computers.
The study revealed that in order for VQE to match the accuracy of chemistry algorithms running on conventional computers, the noise in quantum computers would need to be significantly reduced, approaching fault-tolerant levels. However, practical fault-tolerant quantum computers have not yet been developed.
While some quantum computing companies aim to build fault-tolerant quantum computers within the next five years, capable of running the quantum phase estimation (QPE) algorithm, the study highlights a significant challenge known as the “orthogonality catastrophe.” This phenomenon suggests that as the size of molecules increases, the ability of QPE to calculate their lowest energy levels decreases exponentially.
Thibaud Louvet of Quobly, a French quantum computing company, emphasizes that even with advanced quantum computers, QPE may only be suitable for a limited number of cases. He views the ability to run this algorithm as a benchmark of quantum computers’ maturity rather than a widespread solution for working chemists.
George Booth of King’s College London, who was not involved in the study, acknowledges the challenges highlighted by the research. While quantum computers may not revolutionize quantum chemistry as quickly as anticipated, there are still potential applications in simulating chemical systems’ responses to external stimuli, such as laser light.
In conclusion, while quantum chemistry remains a promising field for quantum computers, the road ahead is fraught with challenges. The study serves as a reminder that the integration of quantum computers into chemistry workflows may require more time and development than initially anticipated.
