Recently, Google and the Columbia University team and others have adopted a new quantum classical hybrid algorithm to achieve 16 qubits of chemical computation, which is the largest chemical computation that a quantum computer can currently complete.
Google Sycamore quantum computer, image from Google
In theory, a quantum computer could achieve quantum superiority and find answers to specific problems that classical computers cannot solve. The more qubits a quantum computer has, the more computational power it has.
Quantum computers can be applied in chemical fields, such as simulating molecular reactions, and will help develop new batteries or new drugs. As molecules get larger, the complexity and difficulty of chemical simulations will grow exponentially, and quantum computers may overcome these challenges.
This time, the Google Quantum Artificial Intelligence Project team, researchers at Columbia University and the University of California, Berkeley, used the Monte Carlo algorithm to use 16 qubits to calculate the ground state energy of molecules (that is, the lowest energy ground state of molecules) on Google's Sycamore quantum computer, achieving the largest chemical quantum computation to date. The results were recently published in the journal Nature.
Google's previous 12 qubit experiments (left) and this 16 qubit experiments (right), picture from the paper
In the study, the aforementioned team proposed and experimentally verified a new method of combining classical computing and quantum computing for chemical research. This is a method of combining the constrained Fermi Quantum Monte Carlo algorithm (QMC) with quantum computing. The Fermi Quantum Monte Carlo Algorithm (QMC) is a Monte Carlo algorithm designed for a quantum physics model of a fermion, a quantum particle containing electrons.
In general, running the Fermi quantum Monte Carlo algorithm on a classical computer does not simulate larger molecules well. So the team took a hybrid approach of classical computing and quantum computing to overcome this difficulty. To evaluate the performance of this quantum-classical hybrid algorithm, the researchers used 16 qubits to calculate the energy of two carbon atoms in a diamond crystal.
The experiment produced 4 qubits more than Google's previous chemical calculations on the Planewood quantum computer, and achieved more accurate experimental results, achieving the largest chemical quantum computation to date.