Industry · Early Pilots
Quantum Computing for Energy and Utilities
From designing better battery materials to optimizing power grids, energy is an area where quantum simulation and optimization algorithms show genuine, if early-stage, promise.
Maturity
Early pilots; material simulation is the strongest long-term case
Most credible quantum workload
Simulating battery, catalyst, and solar materials
Hardware needed
Fault-tolerant machines with thousands of logical qubits
Strongest classical competition
HPC, density functional theory, mixed-integer solvers, and AI surrogates
Example of classical-first progress
AI and HPC screening of battery electrolytes, which delivers candidates today
Realistic horizon
Optimization pilots now; materials impact in the 2030s at the earliest
Overview
The power sector is being rebuilt around variable renewable generation, large-scale batteries, electrified transport, and new nuclear and fusion designs. Every part of that transition produces hard computational problems: balancing a grid in which supply and demand change by the minute, finding better battery chemistries, simulating plasmas inside fusion devices, and forecasting renewable output hours or days ahead. Utilities, grid operators, and energy majors have therefore become regular participants in quantum computing programs.
The honest picture is that most of these problems are already attacked with strong classical tools, including high-performance computing, mixed-integer solvers, and now machine learning. Quantum computing is a long-term option for the parts that classical methods struggle with, mainly the quantum-mechanical simulation of materials and some optimization and simulation tasks. This page separates what is being tried from what is likely to matter.
Why this industry is a candidate for quantum computing
Two kinds of problems in energy have a quantum flavor. The first is simulation of matter: the performance of a battery electrode, a catalyst for hydrogen production, or a solar absorber depends on electron behavior that is exponentially costly to simulate classically when correlations are strong. The second is large optimization and simulation: scheduling thousands of generators, storage units, and flexible loads, or solving nonlinear equations for plasma and fluid flow.
Energy decisions also carry long time horizons and large capital costs, which makes better design tools valuable. A modest improvement in a cathode material or in the operating schedule of a grid can be worth billions over its lifetime, so the sector is willing to fund exploratory work well before hardware matures.
Key facts and numbers
| Item | Detail |
|---|---|
| Battery electrolyte screening (2024) | Microsoft and PNNL used AI and HPC to narrow about 32 million candidates to a small set in under a week, without a quantum computer |
| Aramco and Pasqal | A 200-qubit neutral atom system deployed in Aramco's data center |
| Electrolyte molecule simulation | Resource estimates need thousands of logical qubits |
| Grid scale | National grids have thousands of generators and tens of thousands of buses, handled today by classical solvers |
| Plasma simulation | Kinetic fusion simulations use six-dimensional phase space and run on the largest supercomputers |
| Quantum speedup for Monte Carlo | Quadratic, requiring fault-tolerant hardware |
Where quantum computing could help
Power grid optimization and scheduling
The problem: Grid operators must decide which generators run, how storage charges and discharges, and how power flows through the network at every moment, subject to physical limits and reliability rules. The unit commitment and optimal power flow problems are large and combinatorial, and renewables add uncertainty.
The quantum approach: These problems are encoded as quadratic binary optimization and tried on annealers and QAOA, or split into hybrid workflows. Quantum linear algebra could in principle speed up the linear systems inside power flow calculations.
Where things stand: Studies have run small test networks on quantum hardware, and the European utility EDF has worked with Pasqal on optimizing smart charging of electric vehicles. Industrial solvers handle national-scale grids today, and no quantum method has beaten them at realistic size.
Related algorithms: Quantum Approximate Optimization Algorithm (QAOA), Quantum Annealing, HHL Algorithm
Battery, storage, and hydrogen materials
The problem: Better batteries need electrolytes, electrodes, and interfaces that are stable, conductive, and cheap, and hydrogen production needs better catalysts. Predicting these properties accurately requires solving the electronic structure of large, correlated systems.
The quantum approach: Fault-tolerant quantum phase estimation and Hamiltonian simulation could compute energies and reaction barriers beyond what classical approximations manage. Near-term variational methods are tested on small fragments.
Where things stand: Resource estimates for lithium-ion electrolyte molecules and for catalysts point to thousands of logical qubits. Meanwhile, classical AI and HPC are already delivering: in 2024 Microsoft and the Pacific Northwest National Laboratory reported screening millions of candidates down to a promising solid-state electrolyte using AI and high-performance computing, with no quantum computer involved.
Related algorithms: Quantum Phase Estimation (QPE), Variational Quantum Eigensolver (VQE), Trotter-Suzuki Decomposition
Fusion and plasma simulation
The problem: Designing fusion reactors needs models of plasmas, which are described by nonlinear kinetic and fluid equations in many dimensions. Large simulations strain even the biggest supercomputers.
The quantum approach: Researchers have proposed quantum algorithms for the Vlasov equation, for linear plasma waves, and for nonlinear differential equations, based on Hamiltonian simulation and linear algebra routines.
Where things stand: Theory shows speedups under assumptions, but nonlinearity limits them: results for dissipative nonlinear equations work only when dissipation dominates, and loading data and reading results back are costly. No practical advantage is established, and the work is exploratory.
Related algorithms: Quantum Singular Value Transformation (QSVT), HHL Algorithm, Trotter-Suzuki Decomposition
Energy trading, forecasting, and demand response
The problem: Forecasting wind, solar, and demand, and trading in volatile markets, are statistical tasks that profit from better predictions and faster scenario analysis.
The quantum approach: Quantum machine learning models and quantum amplitude estimation for scenario analysis have been proposed, parallel to finance applications.
Where things stand: Experiments use small datasets and show performance comparable to classical models. As in finance, the speedup for Monte Carlo is quadratic and needs large fault-tolerant machines.
Related algorithms: Quantum Amplitude Estimation, Quantum Kernel Estimation, Quantum Monte Carlo Integration
Who is working on it
- EDF, which has collaborated with Pasqal and Quandela on energy applications such as smart charging and grid optimization.
- TotalEnergies, which has worked with Quantinuum on quantum chemistry for carbon capture materials and with other partners on optimization.
- Saudi Aramco, which installed a 200-qubit neutral atom machine from Pasqal in its data center and explores simulation and optimization.
- ExxonMobil and other energy majors, which have participated in IBM's quantum network and published studies on shipping and logistics optimization.
- U.S. national laboratories including Pacific Northwest, Oak Ridge, and Lawrence Berkeley, which combine AI, HPC, and quantum research for energy materials.
Realistic timeline
| Horizon | What to expect |
|---|---|
| Now to 2028 | Hybrid optimization pilots, benchmark studies, and quantum-inspired solvers. Classical AI and HPC continue to deliver most gains in materials screening. |
| 2028 to 2032 | Early fault-tolerant machines might compute small but chemically relevant fragments of battery and catalyst systems more accurately than classical methods. |
| 2032 and later | If large fault-tolerant machines arrive, accurate simulation of electrode interfaces and catalysts could feed into design. Grid optimization advantage is uncertain. |
How to judge quantum claims in this sector
Energy is a sector where classical computing keeps moving the goalposts. Machine learning potentials now approximate quantum-accurate energies for many materials at a tiny fraction of the cost, and exascale computers run molecular and plasma simulations that were out of reach a decade ago. A quantum claim should therefore specify which accuracy gap it fills, such as strongly correlated transition-metal chemistry, and show that classical approximations fail there.
For grid and trading applications, the baseline is a commercial solver running on a cluster, and the key metric is solution quality within the operator's time limit, which is often seconds to minutes. Pilots that solve a toy network or that take hours on a quantum device do not address that requirement. Look for results on realistic network sizes with realistic constraints and a clear account of how much of the calculation ran classically.
What stands in the way
- Scale: chemistry of real battery systems needs thousands of logical qubits.
- Moving target: AI and HPC keep narrowing the problems that would need quantum hardware.
- Integration: utilities have legacy IT and strict reliability rules that slow adoption of new tools.
- Data loading: grid and plasma problems involve large datasets that must enter a quantum state.
What organizations in this sector should do now
- Track which of your problems are limited by quantum-mechanical accuracy and which by data or compute, since only the former point to quantum computing.
- Join consortia or national lab programs that offer access to hardware and expertise at low cost.
- Use quantum-inspired and classical AI tools now for optimization and materials screening.
- Build in-house literacy so you can judge vendor claims and resource estimates.
Myths versus reality
- Myth: quantum computers will optimize the grid in real time. Reality: grid operators already use powerful solvers, and there is no quantum advantage shown.
- Myth: quantum computing will discover the next battery soon. Reality: classical AI is currently delivering candidates, and quantum simulation needs hardware that does not yet exist.
- Myth: fusion needs quantum computers. Reality: fusion progress depends mostly on engineering and classical simulation.
The bottom line
The energy transition needs better materials and smarter operation, and quantum computing might contribute to the first in the long run. In the near term, the biggest gains come from classical AI, HPC, and better engineering. Utilities and energy firms should invest modestly in understanding quantum methods while delivering results with proven tools.
Frequently asked questions
How could quantum computing help the energy transition?
Mainly by simulating materials for batteries, catalysts, and solar cells with higher accuracy, and possibly by accelerating some optimization and simulation tasks, once fault-tolerant machines exist.
Is quantum computing used to run power grids?
No. Pilots test small networks, but grid operators rely on classical optimization software.
What are the best-known energy pilots?
Examples include EDF with Pasqal on smart charging, TotalEnergies with Quantinuum on carbon capture chemistry, and Aramco's installation of a Pasqal neutral atom machine.
Will quantum computers help fusion?
Algorithms for plasma equations exist on paper, but practical advantage is unproven and limited by nonlinearity and data loading.
When might the energy sector benefit?
Optimization pilots are under way now. Materials simulation benefits are likely in the 2030s at the earliest.
Keep exploring
- Algorithm: Quantum Phase Estimation (QPE)
- Algorithm: Quantum Approximate Optimization Algorithm (QAOA)
- Algorithm: Variational Quantum Eigensolver (VQE)
- Algorithm: Quantum Annealing
- Hardware: Pasqal Fresnel (Orion generation)
- Hardware: IBM Heron (r1, r2, and r3)
- Company: Pasqal
- Company: Quantinuum
- Company: IBM
- Research: Higher-Accuracy Molecular Energy Estimation Using Adaptive VQE Ansätze
- Research: Automated Active Space Selection Improves Practical VQE Chemistry Simulations
- Industry: Quantum Computing for Materials Science
- Industry: Quantum Computing for Oil, Gas, and Mining
- Industry: Quantum Computing for Climate and the Environment
- Industry: Quantum Computing for the Automotive Industry
This page is written for educational purposes and summarizes public announcements, published research, and published resource estimates (last reviewed September 30, 2026). Vendor and customer claims have not all been independently verified, timelines are estimates, and nothing here is investment or business advice.