Advanced Quantum Simulation Could Reveal Cosmic Mysteries

When theoretical physicist Roland Farrell appeared on an IBM Qiskit panel discussing "What to do with 100+ qubits?" he had a particular answer.

Farrell and his colleagues at the University of Washington's InQubator for Quantum Simulation had just used those 100+ qubits to simulate particle physics that sits beyond the reach of the world's most powerful supercomputers.

The approach required a mathematical gamble.

The Washington team recently pulled off what amounts to a mathematical high-wire act. They used classical computers to design quantum circuits that classical computers fundamentally cannot execute.

Then they scaled those designs from 28 qubits to 112 qubits by trusting that certain mathematical patterns would hold across a hundred-fold expansion.

The bet isn't blind. Quantum systems have a property: the further apart two regions are, the less they influence each other. This decay is exponential - fast and reliable. If the pattern holds at small scales, physics says it should hold at large ones too. The team couldn't verify the answer, but they could trust the principle.

The gamble worked.

They've now simulated particle physics on IBM's quantum computers that sits beyond the reach of the world's most powerful supercomputers. The work, published in Physical Review D, demonstrates a path toward understanding some of the universe's deepest mysteries, including why there's more matter than antimatter.

The irony runs deeper.

Error mitigation in quantum computing often means making quantum processors behave more classically. This undoes some of the very quantumness that makes them powerful. The Washington team's breakthrough involves this paradox at every step.

Testing Small, Trusting Large

The algorithm at the heart of this work exploits a fundamental property of quantum systems. The team calls it Scalable Circuits ADAPT-VQE. Correlations between distant parts of the ground state decay exponentially with distance. This means the team could test their approach on modest lattices up to L=14, using just 28 qubits on classical simulators.

Then they extrapolated those same circuit structures to L=50 systems running on 100 qubits.

The approach requires trusting mathematical patterns across a hundred-fold expansion.

Marc Illa, Farrell's co-author, earned his PhD from the University of Barcelona in 2021 as a member of the NPLQCD collaboration. He writes technical explainers for the Wolfram community, translating the team's esoteric quantum field theory work for public audiences. His Barcelona background connects him to lattice quantum chromodynamics methods that the team is now adapting for quantum hardware.

Together with Anthony Ciavarella and Martin Savage, they've constructed what amounts to a recipe. Ciavarella splits his time between the InQubator and Lawrence Berkeley National Laboratory. Savage is the InQubator's director. Determine the quantum circuit ingredients on systems small enough to verify classically. Then scale up by orders of magnitude, trusting the mathematical structure to hold.

Savage, a professor of physics on leave from the Institute for Nuclear Theory, founded the NPLQCD lattice quantum chromodynamics collaboration in 2004.

The team first prepared the vacuum state of the lattice Schwinger model on up to 100 qubits. They used IBM's Eagle-processor quantum computers, specifically the machines designated ibm_brisbane and ibm_cusco. The Schwinger model represents quantum electrodynamics in one spatial dimension plus time.

It's a simplified cousin of the full three-dimensional theory but complex enough to demonstrate the approach's potential.

From Vacuum to Hadrons

Having established that they could prepare quantum vacuum states at scale, the team pushed further. Using IBM's newer 133-qubit Heron processor, housed in a machine called ibm_torino, they simulated hadron dynamics on 112 qubits. This required not just preparing the vacuum state. They had to initialize hadron wavepackets on top of it and then watch those packets evolve in time.

The technique they developed for error mitigation achieved percent-level accuracy. They named it Operator Decoherence Renormalization. They compared their results against classical Matrix Product State simulations.

That precision matters.

Without it, the noise inherent in current quantum processors would swamp any meaningful signal.

IBM's quantum computers have become something of a shared facility for physics groups worldwide. The same Eagle and Heron processors the Washington team accessed via the cloud are used by researchers at the University of Tokyo, Argonne National Laboratory, and UC Berkeley.

Dario Gil, IBM's Senior Vice President and Director of Research, characterizes this ecosystem as a "working agreement between IBM and the global quantum research community," where IBM provides the hardware and researchers discover the algorithms that might eventually achieve quantum advantage.

The arrangement hints at how quantum computing progress might unfold. Not through isolated laboratory breakthroughs but through a distributed network of researchers testing different approaches on shared machines.

Each group's work informs the next.

Current simulations represent early steps toward that goal rather than its achievement.

Simulating Extreme Matter

The physics the Washington team is simulating matters for more than methodological elegance.

The techniques could eventually illuminate why the universe contains more matter than antimatter. That's a fundamental asymmetry without which we wouldn't exist.

They might also help explain how supernovae produce heavy elements like gold and platinum.

And they could reveal the behavior of matter at ultra-high densities, the kind found inside neutron stars or in the early universe.

These aren't abstract puzzles.

As Savage noted in discussing related work on neutrinoless double-beta decay, the timescales involved reach down to yocto-seconds. That's 10^-24 seconds. The distinction between theoretical prediction and experimental verification becomes exceptionally difficult at that scale without quantum simulation.

The Schwinger model the team used serves as a testbed precisely because it's tractable. Full quantum chromodynamics in three spatial dimensions plus time involves quarks, gluons, and a complexity that scales catastrophically.

Current simulations represent early steps toward that goal rather than its achievement.

The scalable circuits approach suggests a pathway, though a long one.

A Culture of Quantum Collaboration

The InQubator for Quantum Simulation, where Farrell completed his PhD, operates with a distinctive culture. One workshop participant described it as arriving "as fermions and leaving as bosons." Individual researchers resist being in the same state, like fermions. Through collaboration they transform into something more collective, like bosons happily occupying the same state.

Farrell's trajectory illustrates this transformation. His earlier work on neutron superfluidity used traditional theoretical methods. The quantum computing work came later, enabled by hardware improvements and algorithmic innovations happening in parallel. He's now at Caltech as a DuBridge Fellow, continuing quantum field theory simulations.

The Hardware Progression

IBM's quantum processor evolution enabled this work. The Eagle processors, with 127 qubits, demonstrated in 2023 what IBM calls "quantum utility"–accurate calculations beyond classical brute-force methods. The Heron processors, introduced late 2023, delivered a five-fold improvement in error rates while maintaining 133 qubits. Lower error rates mean longer circuits can run before noise overwhelms the signal.

This hardware serves a global community. The same machines Farrell's team accessed are used by researchers at the University of Tokyo, Argonne National Laboratory, and UC Berkeley–a distributed network testing different approaches on shared quantum systems.

The Long View

The team's systematic progression from small-scale theoretical work through 100-qubit vacuum simulations to 112-qubit dynamical simulations suggests they're building toward something larger. Other groups are pursuing parallel paths. A collaboration between IonQ and University of Washington researchers recently simulated neutrinoless double-beta decay on trapped-ion quantum computers, exploring the matter-antimatter asymmetry through different hardware.

The questions the field is pursuing require simulation capabilities that may still be years away. Why the universe exists in its current form rather than annihilating itself in a burst of matter-antimatter symmetry. How stars forge the heavy elements that make planets and life possible. What happens to matter under conditions we cannot reproduce in laboratories.

But the pathway Farrell and his colleagues have demonstrated suggests those capabilities are achievable through incremental hardware improvements and algorithmic innovations.

Rather than waiting for a hypothetical quantum computing revolution.

This marks a shift in how physics can progress. For centuries, science has relied on verification: check your predictions against reality, or against calculations you can confirm. Quantum simulation is entering territory where neither is possible. The only check is internal consistency and trust in underlying principles. It's not reckless. But it is new.

For now, the physicists have demonstrated that quantum computers can simulate physics beyond classical reach by trusting mathematical patterns to scale from the tractable to the impossible. The simulations remain simplified models, the Schwinger model a stand-in for full quantum chromodynamics. But the approach works, and that matters considerably more than whether the first demonstrations tackle the hardest problems.

Those can wait for the next generation of machines, the next cohort of researchers learning what to do with hardware that finally performs as quantum theory predicts it should.

Sources

• Primary Sources:
  • Farrell, R. C., Illa, M., Ciavarella, A. N., & Savage, M. J. (2024). Scalable Circuits for Preparing Ground States on Digital Quantum Computers: The Schwinger Model Vacuum on 100 Qubits. PRX Quantum, 5(2). https://journals.aps.org/prxquantum/abstract/10.1103/PRXQuantum.5.020315
  • Farrell, R. C., Illa, M., Ciavarella, A. N., & Savage, M. J. (2024). Quantum simulations of hadron dynamics in the Schwinger model using 112 qubits. Physical Review D, 109(11). https://journals.aps.org/prd/abstract/10.1103/PhysRevD.109.114510
  • U.S. Department of Energy Office of Science (2025, November 18). Nuclear Physicists Create Scalable Quantum Circuits to Simulate Fundamental Forces. ScienceDaily. https://www.sciencedaily.com/releases/2025/11/251118220104.htm
• Context Sources
  • University of Washington InQubator for Quantum Simulation. Faculty page: Martin J. Savage. https://iqus.uw.edu/person/martin-j-savage/
  • U.S. Department of Energy Office of Nuclear Physics (2024). Nuclear Physicists Create Scalable Quantum Circuits to Simulate Fundamental Forces of Nature. https://www.energy.gov/science/np/articles/nuclear-physicists-create-scalable-quantum-circuits-simulate-fundamental
  • IonQ (2024). IonQ and the University of Washington Simulate Process Linked to the Origin of Matter. https://ionq.com/news/ionq-and-the-university-of-washington-simulate-process-linked-to-the
  • IBM Newsroom (2023, December 4). IBM Debuts Next-Generation Quantum Processor & IBM Quantum System Two, Extends Roadmap to Advance Era of Quantum Utility. https://newsroom.ibm.com/2023-12-04-IBM-Debuts-Next-Generation-Quantum-Processor-IBM-Quantum-System-Two,-Extends-Roadmap-to-Advance-Era-of-Quantum-Utility
  • University of Washington InQubator for Quantum Simulation. About Us. https://iqus.uw.edu/about-us/
  • University of Washington InQubator for Quantum Simulation. IBM Qiskit Panel: Roland Farrell. https://iqus.uw.edu/events/ibm-qiskit-panel-roland-farrell/
• ArXiv Preprints (Technical Details)
  • Farrell, R. C., Illa, M., Ciavarella, A. N., & Savage, M. J. (2023). Scalable Circuits for Preparing Ground States on Digital Quantum Computers: The Schwinger Model Vacuum on 100 Qubits. arXiv:2308.04481. https://arxiv.org/abs/2308.04481
  • Farrell, R. C., Illa, M., Ciavarella, A. N., & Savage, M. J. (2024). Quantum simulations of hadron dynamics in the Schwinger model using 112 qubits. arXiv:2401.08044. https://arxiv.org/abs/2401.08044