Molecules are quantum mechanical objects. Their electrons do not orbit the nucleus in neat circles; they exist in superpositions of energy states, entangled with each other, obeying the Schrödinger equation. Simulating their behavior accurately requires tracking these quantum interactions, and the cost of doing so exactly on a classical computer grows exponentially with the size of the molecule.
For exact simulation, this exponential scaling is not a matter of engineering; it comes from the problem itself. Describe a molecule’s electrons using 100 orbital slots, each either filled or empty, and the full quantum state has about 2¹⁰⁰ configurations to keep track of (fewer once the number of electrons is fixed, but still astronomically many). No classical computer can store that many numbers. Computational chemists use approximations (density functional theory, coupled cluster methods) that work well for many molecules but break down for strongly correlated systems, such as transition metal catalysts, high-temperature superconductors, and enzyme active sites.
Quantum computers offer a different approach. Because qubits themselves obey quantum mechanics, they can represent quantum states natively. A quantum simulation of a molecule uses qubits to stand in for the molecule’s orbitals, and quantum gates to implement the interactions between electrons. This does not remove every approximation: the molecule must still be described with a finite set of orbitals, and the algorithm and hardware add errors of their own that must be kept under control. The advantage is that the hardest part, the tangled quantum state of the electrons, no longer has to be squeezed into a classical approximation.
Current hardware has demonstrated quantum chemistry simulations for very small molecules: hydrogen, lithium hydride, beryllium hydride. These results match what classical computers can already achieve, but they serve as proof-of-concept demonstrations for the approach. The hard problems, simulating catalysts, predicting protein-drug interactions, designing materials with specific properties, require fault-tolerant machines with thousands of logical qubits. Those are some years away, but the algorithmic foundations are being laid now.
Quantum chemistry is widely considered the most likely first domain where quantum computers will deliver scientific results that classical machines cannot match.
Subscribe on Substack at https://qubitguy.substack.com/