Quantum Simulation: Modeling Complex Systems

Quantum simulation sits at an unusual intersection: it uses the strangeness of quantum mechanics to understand more strangeness of quantum mechanics — and a great deal of classical physics too. This page covers what quantum simulation is, how quantum simulators are built and operated, where they are being applied, and how researchers decide when a quantum simulator is the right tool rather than a conventional computer.

Definition and scope

Classical computers, even extraordinarily powerful ones, hit a hard wall when modeling quantum systems. The wall is exponential: describing the full quantum state of a system of n particles requires storing roughly 2ⁿ complex numbers. At around 50 interacting quantum particles, that number exceeds the storage capacity of every computer on Earth combined. Richard Feynman noticed this problem in a 1982 lecture at MIT and proposed the solution: build a controllable quantum system and use it to mimic the behavior of another quantum system (Feynman, "Simulating Physics with Computers," International Journal of Theoretical Physics, 1982).

That is the essential definition of quantum simulation — using one well-controlled quantum system to model the behavior of another that is harder to study directly. The scope spans chemistry, materials science, condensed matter physics, high-energy physics, and drug discovery. Quantum simulation is distinct from quantum computing basics in a subtle but important way: a quantum simulator does not have to be universal. It does not need to run arbitrary algorithms. It only needs to reproduce the specific physics of the system being modeled.

How it works

Quantum simulators come in two broad architectures that researchers distinguish sharply:

Analog quantum simulators physically map one quantum system onto another. Cold atoms trapped in optical lattices, for instance, can be tuned to replicate the Hubbard model — a foundational model of electron interactions in solid-state materials. The simulator's Hamiltonian is engineered to match the target system's Hamiltonian. No gate operations, no circuit — just direct physical correspondence.

Digital quantum simulators use sequences of quantum logic gates on a programmable quantum processor to approximate the time evolution of a target system. This approach is more flexible but requires error correction overhead. A 2022 experiment by Google's quantum team demonstrated simulation of a 1+1 dimensional lattice gauge theory using 20 qubits on the Sycamore processor (Google Quantum AI, Nature, 2022).

The core mechanism in both cases draws on quantum superposition and quantum entanglement — properties that allow a quantum system of modest size to carry exponentially more information than a classical register of equivalent hardware. The Schrödinger equation governs time evolution, and the simulator's job is to enact that evolution on tunable hardware rather than solving the equation numerically.

The full physics of the site is laid out at the main reference index, which connects simulation to related areas including quantum field theory and condensed matter.

Common scenarios

Quantum simulation has demonstrated practical traction in four domains:

  1. Quantum chemistry: Calculating ground-state energies of molecules such as FeMoco (the iron-molybdenum cofactor in nitrogenase) is intractable classically for molecules beyond roughly 50 active electrons. Quantum simulations using variational quantum eigensolvers (VQE) have already matched classical results on small molecules like hydrogen and lithium hydride, providing a benchmark for scaling up.

  2. Condensed matter physics: The Hubbard model, used to study high-temperature superconductivity, has resisted exact classical solution for decades despite its apparent simplicity. Cold-atom simulators have probed regimes of the model's phase diagram that remain analytically inaccessible.

  3. High-energy physics and lattice gauge theories: Understanding quantum chromodynamics — the theory governing quarks and gluons — requires lattice QCD calculations that consume months of supercomputer time. Quantum simulators offer a potential shortcut for specific observables.

  4. Drug discovery and protein folding: Modeling the quantum mechanical interactions between drug candidates and biological targets, particularly metalloenzymes, is a near-term application that pharmaceutical researchers at institutions including the National Institute of Standards and Technology (NIST) are tracking closely.

Decision boundaries

Quantum simulation is not always the right choice. The decision to deploy a quantum simulator — rather than a classical algorithm or a high-performance computing cluster — depends on identifiable criteria:

The contrast between analog and digital approaches mirrors a broader tension in quantum computing basics: specialization versus universality, robustness versus programmability. Neither architecture dominates for all use cases, and hybrid approaches — using classical pre-processing to configure analog hardware — are an active area at institutions including Caltech, MIT, and the Max Planck Institute for Quantum Optics.

References

Read Next

Quantum Simulation: Modeling Complex Systems with Quantum Hardware This page covers what quantum simulation is, how quantum simulators are built and operated, where they are being applied, and... Quantum Computing Basics: Qubits, Gates, and Quantum Algorithms A qubit breaks that rule in a way that still unsettles physicists who've spent careers thinking about it. Quantum Superposition: How Particles Exist in Multiple States This page covers the formal definition of superposition, how states combine and collapse through measurement, the causal...