Physicists spent eight years arguing whether a purple crystal contained exotic Majorana particles useful for quantum computers. Cornell found neither particles nor dirt — just soundwaves spiraling like corkscrews. Neither camp predicted it.
Physicists have spent eight years arguing about what happens when heat flows sideways through a purple crystal. The answer, according to Cornell, published in Nature this week, is soundwaves taking a corkscrew path — and that restructures a debate nobody predicted would end this way.
Ruthenium trichloride, a material that looks unremarkable on a lab bench, has been the center of a running argument since 2018, according to Cornell. That year, researchers in Japan reported finding evidence of Majorana fermions — exotic particles that could serve as building blocks for a topological quantum computer — inside a quantum spin liquid made of the compound. If true, it meant ruthenium trichloride was not merely interesting physics but a candidate material for the hardware that might eventually outperform conventional quantum computers. A 2021 paper in Science appeared to confirm the signature, reporting a half-integer quantized thermal Hall effect consistent with Majorana physics.
Then the counterargument formed: the signal was not from particles at all. It was dirt. Magnetic impurities or defects in the sample, the skeptics said, not an intrinsic quantum phenomenon. Both camps had data. Both were confident.
The Cornell team, led by associate professor Brad Ramshaw and doctoral student Avi Shragai, did not take sides. They changed the question. Rather than debating what was in the material, they directly measured how heat moved through it: by following the phonons. Phonons are quantum packets of lattice vibration — essentially soundwaves at the atomic scale. Insulators carry heat through phonons because they have no free electrons to move charge, and normally phonons do not care about magnetic fields. But in ruthenium trichloride, spin orbit coupling gives the phonons a left-right awareness they should not have. Applying a magnetic field and tracking phonon movement with ultrasonic measurements, the researchers found the phonons had twisted corkscrew trajectories. Ramshaw called this the acoustic Faraday effect, after the phenomenon that lets light know left from right in certain materials. It demonstrated that the sample had Hall viscosity — a property that Cornell describes as theorized but never demonstrated as a working probe of quantum states.
The corkscrew paths are not what either camp predicted. The Majorana interpretation required quantized thermal Hall conductance — a specific fingerprint — and the Cornell measurements showed something different. The impurity explanation required the signal to vary with sample quality, and the Cornell team found an intrinsic effect that did not depend on how clean the material was. Neither saw this coming.
This matters for quantum computing because Majorana fermions are not just a theoretical curiosity. They are one of the few particle species that might naturally host topological protection — a property that could make qubits resistant to the noise that currently limits quantum computers. For years, Microsoft has bet its quantum computing strategy on Majorana-based hardware. If ruthenium trichloride's thermal Hall signal is not a Majorana signature at all, it does not kill the search for Majoranas everywhere, but it removes one data point that seemed to confirm they could exist in real materials. Meanwhile, the corkscrew phonons are their own phenomenon — a newly characterized intrinsic effect that had not been observed before. Ramshaw was direct: it is not the magic Majorana material, and it is not fancy dirt. It is something new.
The finding restructures rather than resolves the broader search. Several groups have reported possible Majorana signatures in different platforms — semiconductor-superconductor nanowires, vortex cores in iron-based superconductors — as documented in the Nature paper — and each has faced scrutiny over whether the signal really means what the researchers claim. Ruthenium trichloride is not unique in being contested. The pattern of contested Majorana claims is itself a data point about how hard these signatures are to interpret and how cautious the field needs to be before declaring that the particles have been found. What the Cornell result does is narrow the space of what ruthenium trichloride is — and by doing so, clarifies which questions about it remain open.
Hall viscosity as a measurement tool may prove more durable than the ruthenium trichloride debate. The Cornell team demonstrated it works as a probe of quantum states in a real material for the first time. If the technique generalizes, other groups could use ultrasonic measurements to characterize states of matter that have been difficult to examine directly. That is a narrower contribution than a confirmed new particle, but it is also a tool other researchers can use — and unlike a contested Majorana signal, it does not require accepting a particular theoretical framework to be useful.
The result does not immediately affect any quantum computing roadmaps. Ruthenium trichloride is a research material, not a device. But for the people tracking which physical systems are viable candidates for robust quantum bits, the paper closes one door and opens a different question: what does a corkscrew phonon system look like when someone tries to build something with it?
Story entered the newsroom
Research completed — 5 sources registered. Three facts the wire missed: (1) The measurement technique is the real story — ultrasonic tracking of phonon paths, not just heat flow. (2) Both camps
Draft (8 words)
Reporter revised draft (580 words)
Reporter revised draft (818 words)
Reporter revised draft (580 words)
Reporter revised draft (820 words)
Published (820 words)
@Rachel — kill story_11437. IBM Q1 beats on EPS and revenue, keeps 2026 guidance, but shares drop 6% after hours. Wire headline overstates Iran angle; the real headline is IBM shares turn lower as company beats but opts to maintain guidance. Solid numbers — 9% YoY revenue, software +11%, Z mainframe +51% — but they don't clear type0 without a sharper hook. We've already covered IBM earnings recently (Anthropic/COBOL threat Feb, Confluent B deal), and there's no new capability, funding signal, or policy angle. Reject — reporter can pitch a stronger angle if they find one.
Get the best frontier systems analysis delivered weekly. No spam, no fluff.
Quantum Computing · 2h 42m ago · 3 min read
Quantum Computing · 1d ago · 2 min read
Quantum Computing · 1d ago · 2 min read
