Calculating the chemical properties of molecules can be a job for quantum computers
ETH Zurich
Quantum chemistry calculations that could advance drug development or agriculture have recently emerged as a promising “killer application” of quantum computers, but a new analysis suggests this is unlikely to be the case.
Advances in building quantum computers have accelerated greatly in recent years, but it remains an open question which applications are most likely to justify the ongoing investment in this technology. A popular contender is solving problems in quantum chemistry, such as calculating the energy levels of molecules relevant to biomedicine or industry. This requires taking into account the behavior of many quantum particles – electrons in the molecule – at the same time, so it seems a good match for computers made of many quantum parts.
However, Xavier Waintal at CEA Grenoble in France and his colleagues have now shown that two leading quantum computing algorithms for this task may actually be of limited use at best.
“My personal thinking is that it’s probably doomed, not proven doomed, but probably doomed,” he says of the use of quantum computers for molecular energy calculations.
The researchers divide their mathematical analysis into two parts, one that concerns existing quantum computers, all of which are prone to errors, and one that concerns future quantum computers that would be “fault-tolerant,” or completely error-proof.
When using error-prone or noisy quantum computers, molecular energy levels can be calculated with the variational quantum eigensolver (VQE) algorithm, but the accuracy of the results depends on the severity of this noise.
The team’s analysis found that for VQE to compete in accuracy with chemistry algorithms that can be run on conventional computers, quantum computers’ noise must be suppressed so severely that they must be effectively fault-tolerant. In particular, a practical fault-tolerant quantum computer has not yet been created.
Several quantum computing companies aim to build fault-tolerant quantum computing within five years, and these devices can calculate the molecules’ energies with another algorithm called quantum phase estimation (QPE). Here, the problem of error is almost eliminated, but the study highlights a problem that goes by the ominous name of “orthogonality disaster”.
Simply put, this means that as the size of the molecules increases, the probability that the QPE can calculate its lowest energy level decreases exponentially. As a result, team member Thibaud Louvet of the French quantum computing company Quobly says that even with great quantum computers, there will only be a small number of cases where using them to run QPE would be the most practical and best choice. In his view, being able to run this algorithm should be seen more as a measure of the maturity of quantum computers than something that could become a mainstay for working chemists.
“It is easy to overhype the prospects of quantum computers in this domain, with many believing that the advent of quantum computers will immediately render any classical approach to quantum chemistry obsolete,” says George Booth of King’s College London, who was not involved in the work. “This study is clear to point to significant challenges for accurate molecular simulation, which will remain even in the ‘fault-tolerant era’, casting doubt on whether quantum chemistry is really such a quick victory for quantum computers.”
But he says there are still other ways quantum computers could be used in chemistry. For example, they can simulate how chemical systems change after being disturbed, such as being hit with laser light.
Topics:
