Quantum Simulation: Trotter, Qubitization and Why Chemistry Leads
Why this problem comes first
Most algorithm headlines are about beating a classical method at a puzzle. Simulation is different: the problem itself is quantum. Molecules, magnets and materials follow quantum rules, and the memory needed to track them exactly on a classical computer doubles with each added particle (see why chemistry is quantum hard). Richard Feynman proposed in 1982 that a quantum computer might be the natural solution, and Seth Lloyd studied the case of local interactions in 1996, according to the reference below. Unlike factoring, simulation does not rely on an unproven bet that no classical shortcut exists for a cleverly built puzzle. Nature is already doing the calculation, and we want a machine that can follow it.
The task: time evolution
The Hamiltonian H lists a system's energies and interactions. Hamiltonian simulation means building the operation e to the minus i H t, which tells you how the system changes over time t. Once you can do that, phase estimation (explained here) can extract energies, and energies drive questions like reaction rates and material stability.
Method 1: Trotter, the flip-book
If H is a sum of simple pieces A, B and C, you cannot apply them all at once exactly, because they do not commute. Trotter-Suzuki product formulas approximate the whole by cycling through the pieces in tiny slices: do a bit of A, a bit of B, a bit of C, repeat. Like a flip-book, more pages give a smoother movie. The reference lists a first-order gate cost that grows as t squared over epsilon, and credits Suzuki (1991) and Berry, Ahokas, Cleve and Sanders (2007). Its appeal is simplicity and the fact that it needs few extra qubits. Its weakness is that cost grows quickly as you ask for higher accuracy.
Method 2: qubitization and signal processing
Qubitization, introduced by Guang Hao Low and Isaac Chuang (arXiv, October 2016), takes a different road. It builds a quantum-walk-style operation whose spectrum encodes the Hamiltonian, and then shapes it with polynomials. Per the abstract, their algorithm uses order t plus log(1/epsilon) queries to each oracle with at most two extra ancilla qubits, which they describe as optimal across the parameters. In plain English: the cost is roughly additive in the evolution time and the log of the precision, instead of multiplying them, so high-precision simulation gets much cheaper. A follow-up by Gilyen, Su, Low and Wiebe (2018) generalized this into quantum singular value transformation, a unified framework that covers optimal Hamiltonian simulation, fixed-point amplitude amplification and more.
Trade-off summary: Trotter is simple and can be fine for smaller or noisier jobs; qubitization has the best proven scaling and is the favorite for large fault-tolerant chemistry runs.
Why chemistry and materials lead
- Natural fit. The input is a quantum system and the output is a handful of numbers (energies). That avoids the data loading and readout problems that hurt other algorithms (see data loading).
- Value. Reaction pathways, catalysts, batteries and fertilizer chemistry matter economically, and some strongly correlated molecules are hard for approximate classical methods.
- Well-studied costs. Teams have published detailed resource counts, which makes the use case measurable instead of vague.
The FeMoco benchmark
The nitrogen-fixing enzyme cofactor FeMoco is the poster child (see FeMoco explained). Reiher, Wiebe, Svore, Wecker and Troyer (arXiv 2016, PNAS 2017) argued for studying nitrogenase on quantum computers and reported resource estimates showing the work could be feasible on small fault-tolerant machines. Estimates have since fallen. Lee and coauthors (arXiv November 2020, PRX Quantum 2021) applied qubitization to a tensor hypercontraction form of the Hamiltonian and reported FeMoco needing about four million physical qubits and under four days of runtime, assuming 1 microsecond cycle times and physical gate errors no worse than 0.1 percent. Treat that as an estimate under assumptions, not a schedule: no such machine exists today.
What is not proven
Two honest caveats. First, it is not proven that classical methods will never handle these molecules well enough for industry; classical chemistry keeps improving. Second, the textbook algorithm assumes a good starting state, and finding one is itself a research question. For a candid timeline discussion see the honest guide and named pilots. Real pilots today are mostly on small molecules and hybrid methods, which prove workflows, not advantage.
Nothing here is financial advice, and the QNT memecoin is independent of Quantinuum Ltd.
Sources and further reading
- Wikipedia: Hamiltonian simulation
- Low and Chuang: Hamiltonian Simulation by Qubitization (arXiv)
- Gilyen, Su, Low, Wiebe: Quantum singular value transformation (arXiv)
- Lee et al.: Even more efficient quantum computations of chemistry through tensor hypercontraction (arXiv)
- Reiher et al.: Elucidating Reaction Mechanisms on Quantum Computers (arXiv)
Reported as of 2026-10-09. Theory results are proven only under the stated assumptions, and experimental claims and classical rebuttals keep changing, so check the primary papers. Nothing here is financial advice and nothing here predicts the price of any asset. The QNT memecoin is independent of Quantinuum Ltd, the real company, and of every lab and researcher named on this page.
Frequently asked questions
What is Hamiltonian simulation?
Building the operation that shows how a quantum system evolves in time. It is the core step in quantum chemistry and materials algorithms.
What is the difference between Trotter and qubitization?
Trotter approximates evolution by cycling through small pieces and is simple. Qubitization builds a walk-like operation and uses polynomials, reaching what its authors describe as optimal query cost.
How big a machine would FeMoco need?
A 2021 paper reported about four million physical qubits and under four days under stated assumptions. This is an estimate, and no such machine exists yet.
Keep reading
- Why Chemistry Is the Problem Quantum Computers Were Made For
A plain English look at why molecules are hard for ordinary computers, which chemistry problems are truly quantum-hard, and why most of chemistry stays classical. - FeMoco, Nitrogenase and Fertilizer: Quantum Computing's Famous Test Case
How one enzyme that makes natural fertilizer became the benchmark for quantum chemistry, how the resource estimates fell, and what is still unproven. - Quantum Chemistry Timelines: An Honest, Optimistic Guide
When might quantum computers do useful chemistry? A grounded look at published projections from the early 2030s to the 2040s and what could change them.
All Quantum computing guides | Back to top | Search the site
Main pages: Quantum computing explained | Quantum and crypto | Companies | Quantum news | Glossary