Many materials are challenging to simulate on ordinary computers due to their complex, quantum nature. This is where quantum computers can help but, until now, most of the research has focussed on the simulation of molecules, not materials. This is because materials have additional structure such as translational symmetry or periodicity.
But could we modify an existing molecular algorithm to take advantage of this structure?
In a recent paper, published today in Physical Review Research, we figured out how to do just this. Working with researchers from Johnson-Matthey Technology Centre, we developed a new quantum algorithm to simulate the catalysts used in many industrial chemical processes.
The paper demonstrates how an error-corrected quantum computer can simulate nickel oxide and palladium oxide, which are important materials in heterogeneous catalysis, a process used to create a broad range of chemicals and fuels.
This work paves the way towards future practical simulations of materials on error-corrected quantum computers.
Our algorithm enables the quantum simulation of large solid-state systems with runtimes often associated with much smaller molecular systems. This reduces the quantum resource requirements to run such algorithms, including the size of quantum computer and the number of operations it needs to carry out.
In other words, future quantum computers will require far fewer qubits and a reduced circuit depth, compared to when quantum algorithms are used without any modification.
When we submitted the paper, we were the first to modify the qubitization algorithm to account explicitly for translational symmetry present in materials leveraging concepts used in classical computation. Now, other researchers are undertaking similar approaches citing our work.
The only caveat here is that we will have to wait until someone builds an error-corrected quantum computer.
Building an error-corrected (aka fault tolerant) quantum computer is incredibly hard. Millions of qubits must be controlled and calibrated (for some qubit types at millikelvin temperatures). The decoding cycle and control loops must be highly scale and fast, allowing terabytes of data to be processed every second. This is a massive real-time information processing problem that we can solve.
To reach error-correction sooner, Riverlane is building the operating system for error-corrected quantum computers, which includes a control system (Deltaflow.Control to control and calibrate the millions of qubits required) and fast decoders (Deltaflow.Decode to stop errors propagating and rendering calculations useless).
We call the solution the Operating System (OS) for quantum computing: Deltaflow.OS. But we also need a series of fault-tolerant algorithms, which are ready to run on error-corrected quantum computers.
That’s why it’s important to not wait and develop these algorithms now, while identifying useful industrial applications for quantum computers. Just as scaling up the quantum hardware is a complex undertaking, so is scaling up quantum algorithms for the fault-tolerant regime.
Because when error-corrected quantum computers are ready, we also need fault-tolerant quantum algorithms to unlock the true potential of quantum computers. This work pushes us closer to realising that ambition and takes us one step closer to useful quantum computing.