The U.S. Department of Energy announced on October 8 that Rakshya Khatiwada, a physicist jointly appointed at Fermi National Accelerator Laboratory and Illinois Institute of Technology, has won a DOE Early Career Award to build and operate a four-qubit entangled quantum sensing system at Fermilab and expose it to a simulated dark matter signal. The experiment will directly test whether quantum entanglement — the same phenomenon that powers leading quantum computers — provides a measurable sensing advantage over instruments that work independently, in a domain where classical physics has come up empty for decades.

Dark matter constitutes roughly five times more of the universe’s matter than all the ordinary atoms in stars, planets, and people combined, yet no dark matter particle has ever been directly detected. Every galaxy rotation curve, every gravitational lensing map, every analysis of the cosmic microwave background points to a universe whose dominant ingredient remains invisible to every instrument ever built. Khatiwada’s project is a concrete step toward changing that.

“We already see some signatures of dark matter from astrophysical observations like gravitational lensing and cosmic microwave background radiation,” Khatiwada said in a Fermilab newsroom statement. “Searching for dark matter in the laboratory requires a different approach. Scientists need highly specialized instruments to detect the faint signals that dark matter might produce through interactions with ordinary matter and electromagnetic fields.”

Why Classic Dark Matter Searches Have Hit a Wall

For the better part of four decades, most resources in particle physics dark matter searches flowed toward Weakly Interacting Massive Particles, or WIMPs — hypothetical heavy particles that the Standard Model’s preferred extensions predicted should exist in our galaxy’s dark matter halo and occasionally scatter off ordinary matter. Underground liquid-xenon detectors, built to catch those rare scatters, have steadily grown more sensitive and steadily returned nothing. The LUX-ZEPLIN experiment’s 2025 analysis set the strongest limits ever achieved for WIMPs at masses of 9 GeV or above, excluding cross-sections above roughly 2.2×10⁻⁴⁸ cm² at a WIMP mass of 40 GeV/c². XENONnT, with 3.1 tonne-years of exposure at Italy’s Gran Sasso laboratory, similarly found no significant signal, pushing exclusion limits to 1.7×10⁻⁴⁷ cm² at a WIMP mass of 30 GeV/c².

These null results have redirected significant scientific attention toward a different class of dark matter candidates: ultralight bosonic particles such as axions and dark photons. Their detection requires a fundamentally different toolkit — not the huge vats of noble liquid that WIMP searches use, but quantum sensors capable of measuring signals so faint that quantum noise itself would overwhelm any classical amplifier.

What Axions Are and Why They’re So Hard to Find

Axions occupy a peculiar double role in particle physics. They were proposed in 1977 by Roberto Peccei and Helen Quinn as a solution to an entirely separate puzzle — the “strong CP problem” in quantum chromodynamics, which asks why the strong nuclear force appears to conserve a symmetry (charge-parity symmetry) that it has no obvious reason to conserve. Frank Wilczek and Steven Weinberg independently showed that the Peccei-Quinn mechanism predicted a new particle; Wilczek named it the “axion” after a brand of laundry detergent because it “cleaned up” the problem. It later emerged that invisible axions — with masses in the range of roughly 10 microelectronvolts (μeV) — could also constitute cold dark matter, giving the particle a compelling dual theoretical motivation.

Dark photons are a separate class of hypothetical particle: analogues of the ordinary photon that could mediate forces in a hidden sector of physics, coupling weakly to electromagnetic fields as ordinary dark matter candidates. Both axions and dark photons, if they exist, would produce vanishingly small electromagnetic signals. They are the problem that quantum sensing was developed to solve.

From Squeezing to Entanglement: A Qualitative Shift in Sensing

The leading quantum sensing approach before Khatiwada’s experiment is quantum squeezing, pioneered in the HAYSTAC (Haloscope at Yale Sensitive To Axion Cold Dark Matter) experiment and in ADMX, the Axion Dark Matter eXperiment. In a squeezed-state receiver, quantum noise in a microwave cavity is redistributed — reduced in the measurement quadrature of interest at the cost of increased uncertainty in the conjugate quadrature — allowing the detector to achieve noise levels below the standard quantum limit (SQL) for a single mode. HAYSTAC’s Phase II results, published in 2024, demonstrated that this approach can scan axion parameter space faster than a quantum-limited experiment, covering 2.27 μeV of previously unexplored mass range between 16.96 and 19.46 μeV.

Before joining Fermilab, Khatiwada conducted her postdoctoral research at the University of Washington working as a quantum detector lead with ADMX, the Axion Dark Matter eXperiment, developing quantum amplifiers for dark matter detection. That experience shaped her view that quantum sensing is not merely useful for detecting dark matter — it is essential.

Entanglement is a different and potentially more powerful tool than squeezing. Squeezing manages noise in a single quantum system; entanglement links multiple quantum systems into a single shared state. The distinction matters because of how each approach scales with the number of sensors.

How Entangled Qubits Could Spot Dark Matter

With N independent, non-entangled sensors, measurement precision scales as 1/√N — the standard quantum limit. Quantum theory allows a fundamentally better bound: when N sensors share an entangled quantum state, the precision can approach the Heisenberg limit of 1/N. In the context of dark matter detection, this means a signal response that scales as N² rather than N, without requiring the extraction of energy from the dark matter signal, which classical probes cannot avoid.

The physical mechanism works as follows. When a wave-like dark matter field — an axion or dark photon oscillating coherently through the detector — encounters a qubit, it alters the qubit’s quantum state very slightly. In an array of independent qubits, each qubit records that perturbation separately; the total signal accumulates linearly. But when qubits are entangled, the dark matter interaction encodes information in the relative phases between the components of the shared quantum state — a subtler, richer signature that the entangled system detects collectively. Measuring that collective state extracts more information per shot than the sum of N independent measurements.

“These entangled qubits share a single quantum state,” Khatiwada told Fermilab’s newsroom. “So, if they all encounter the same dark matter signal, their response is enhanced.”

Khatiwada’s team will build a four-qubit entangled system using superconducting transmon qubits — the same fundamental architecture that underpins the leading quantum computing platforms from IBM, Google, and others. The experiment will expose this system to a simulated dark matter signal and directly compare its response against independent, non-entangled qubits under identical conditions, producing an empirical measurement of the entanglement sensing advantage.

Engineering Challenges: Why Four Qubits Is the Right Starting Point

The theoretical entanglement advantage has a practical adversary: decoherence. Superconducting qubits lose their quantum coherence through multiple noise channels — charge noise, magnetic noise, quasiparticle generation, and phonons from the surrounding substrate — and the shared state of entangled qubits is especially fragile. Once error rates across the entangled system exceed a threshold, the N² scaling advantage collapses back toward classical behavior. A 2025 theoretical analysis found that with current state-of-the-art superconducting qubit technology, roughly 100 entangled qubits could match the sensitivity of existing power-readout cavity experiments at a dark matter mass around 20 μeV. Four qubits is not the endpoint — it is a carefully chosen early-stage test that current hardware can execute with high fidelity.

An additional engineering wrinkle: the entangled protocol suppresses uncorrelated background noise (noise that affects each qubit independently) but does NOT suppress correlated noise that acts across multiple qubits simultaneously. Khatiwada’s team will need to characterize Fermilab’s correlated noise sources — a challenge her ongoing work in the QUIET and LOUD facilities is already addressing. Her group has demonstrated very low-noise performance of superconducting qubits at those facilities, and is applying machine learning and artificial intelligence to identify subtle noise signatures that traditional methods cannot detect.

Fermilab’s Quantum Ecosystem

Khatiwada’s project is embedded in the Superconducting Quantum Materials and Systems (SQMS) Center, one of five National Quantum Information Science Research Centers funded by the DOE. SQMS is a collaboration of more than 30 partner institutions — national laboratories, universities, and industry — working to advance superconducting quantum hardware by leveraging Fermilab’s decades of expertise in building complex superconducting accelerator systems. The center’s infrastructure gives Khatiwada’s team access to the fabrication and testing capabilities required to build and operate the four-qubit entangled system.

“I am really excited about applying these ideas that I have been working on to entanglement-based sensing for dark matter,” Khatiwada said in the Fermilab award announcement. “Fermilab has excellent infrastructure, resources and technical expertise, as well as talented postdoctoral researchers. This is a great place to do research.”

She also credited her professorship at Illinois Institute of Technology, which gives her access to graduate students who are integral to the project’s execution.

What Success Would Mean — and What Remains Open

If the four-qubit experiment demonstrates a measurable entanglement advantage, the implications extend well beyond axions and dark photons. An entanglement-enhanced sensor could, in principle, scan dark matter mass space more efficiently than any resonant cavity experiment — which must be laboriously tuned one narrow band at a time, because each setting is sensitive to a tiny range of potential axion or dark photon masses. An entangled sensor that covers multiple masses simultaneously, with N² signal scaling, would be a qualitative acceleration.

The project also fits a broader pattern of DOE investment in quantum-enhanced dark matter searches at Fermilab. In April 2026, Fermilab scientist Yao Lu received his own Early Career Award, dark photon cavity research, to build a scalable superconducting cavity array aimed at detecting dark photons through quantum-enhanced searches. And in June 2026, Fermilab postdoctoral researcher Christina Wang received the APS Mitsuyoshi Tanaka Award in Experimental Particle Physics for pioneering detector work that opens new pathways to identifying dark matter candidates. The pattern reflects a field-wide consensus: after decades of null results from WIMP searches, quantum sensing technology has opened genuinely new experimental territory.

“Entangled qubits could not only help researchers detect fainter signals,” Khatiwada explained in Fermilab’s newsroom, “but also search more efficiently across a range of possible dark matter masses.”

Whether four entangled qubits at Fermilab can demonstrate that advantage in a controlled setting will be answered in the coming years of research. If they can, the result would validate a detection strategy that scales naturally as quantum computing hardware improves — giving dark matter searches a roadmap toward the sensitivity that the universe’s invisible majority may require.

Frequently Asked Questions

How does quantum entanglement improve dark matter detection compared to existing quantum sensing methods?

Previous quantum sensing approaches for dark matter — such as squeezed-state receivers used in the HAYSTAC experiment — manage noise within a single quantum system, reducing uncertainty in one measured quantity below the standard quantum limit. Entanglement connects multiple qubits into a shared quantum state. When a dark matter signal touches all of them simultaneously, information is encoded in the collective correlations between those qubits rather than in each qubit separately. Theory predicts the signal response can scale as the square of the number of qubits (N²) rather than linearly — a potentially enormous advantage as qubit counts grow. Khatiwada’s four-qubit Fermilab experiment will test whether that theoretical advantage is real and measurable in practice.

Why does dark matter produce signals too faint for classical instruments to catch?

Dark matter candidates such as axions and dark photons are hypothesized to interact with ordinary matter only through extraordinarily weak forces — in the case of axions, through a coupling to the electromagnetic field that converts them into microwave photons in a strong magnetic field at a rate set by their mass. The expected signal power is on the order of 10⁻²⁴ watts: many orders of magnitude below anything a conventional amplifier can detect without adding more noise than signal. Quantum sensors exploit the laws of quantum mechanics to make measurements that classical amplifiers cannot, and entanglement pushes that capability further still.

What are axions, and why do physicists think dark matter could be made of them?

Axions were proposed in 1977 by Roberto Peccei and Helen Quinn — not to explain dark matter, but to solve a different puzzle in nuclear physics called the strong CP problem, which asks why the strong nuclear force conserves charge-parity symmetry despite no known reason to do so. Frank Wilczek and Steven Weinberg independently showed the Peccei-Quinn mechanism predicts a new particle; Wilczek named it the axion. It later emerged that invisible axions in a specific mass range could also constitute the cold dark matter inferred from galactic rotation curves and gravitational lensing — giving the particle dual theoretical motivation that makes it one of the field’s most compelling search targets.

What is the engineering limitation preventing immediate scale-up to many entangled qubits?

Decoherence. Superconducting qubits lose their quantum coherence through noise channels including charge fluctuations, magnetic noise, and quasiparticle excitations. In an entangled system, the shared quantum state is especially fragile, and the sensing advantage disappears once error rates across the array exceed a threshold. A 2025 theoretical study found that with current hardware, approximately 100 well-controlled entangled qubits could match the sensitivity of existing cavity-based dark matter experiments at a dark matter mass around 20 μeV. Four qubits is a tractable starting point that current technology can execute with high fidelity, providing an empirical baseline for what improvements in qubit coherence will unlock.