(IBMblog) A recent research blog written by Marc Ganzhorn, Daniel Egger, and Stefan Filipp explains how their team at IBM Research-Zurich now lays out how exchange-type two-qubit gates constitute a very promising avenue to calculate molecular properties.
Breakthroughs in materials science are among the main drivers of technological change in the modern world. Understanding the inner workings of the molecules making up those materials is key to designing better drugs, healthier foods or more energy-efficient batteries, to name a few examples. Although the quantum mechanical equations governing the behavior of molecular entities have been known for many decades, solving them still represents a serious challenge even for our best current computers.
The solution to this challenge could be within grasp. The very fact that quantum computers rely on manipulating the quantum states of their smallest units, known as qubits, makes them naturally better suited to simulate quantum mechanical systems like molecules.
In “Gate-Efficient Simulation of Molecular Eigenstates on a Quantum Computer“, published in the peer-reviewed journal Physical Review Applied, the research team proposed and experimentally demonstrated a gate-efficient method to compute the eigenstate energies of molecules using superconducting qubit hardware and Qiskit Aqua that allowed them to reduce the length of relevant algorithms by up to an order of magnitude. That improvement is roughly equivalent to increasing the coherence time of the qubits by the same factor. While exchange-type gates had been proposed before, the authors’ work represents the first time such gates are made tunable in both amplitude and phase.
The research demonstrates a gate-efficient way to simulate molecular spectra on a tailor-made superconducting qubit processor using exchange-type two-qubit gates. The team’s findings show that adapting quantum algorithms and hardware to the problem at hand is a key requirement to perform quantum simulation on a larger scale. In particular, exchange-type gates are a promising choice to compute the energy spectra of larger molecules like water on near-term quantum hardware.