Analysis: Science & Technology — 26 October 2026
RIKEN runs five silicon spin qubits at once with ~99.9% fidelity
Researchers at RIKEN’s Center for Emergent Matter Science report in Nature Physics that five silicon spin qubits on one shared microwave line can be driven together while single-qubit gate fidelity stays near 99.9 percent. Tested one at a time, each qubit exceeded 99.99 percent, with the best approaching 99.999 percent; three qubits in parallel remained above 99.99 percent. Silicon qubits can be fabricated with existing semiconductor tools, but simultaneous control has usually cost accuracy. Pairwise calibration of the AC Stark shift, plus pulse shaping, is what held the error rate down.
Silicon spin qubits store information in the spin of single electrons confined in silicon. Microwave pulses rotate those spins, yet a shared drive line also shifts energy levels and adds unwanted phase—the AC Stark effect. Independent wiring for every qubit does not scale toward the large arrays needed for fault-tolerant machines. A calibration that uses pairwise measurements rather than every combination is a more practical control scheme.
The leftover error is drive-induced decoherence from the extra microwave power required for parallel gates; the team did not eliminate it. Five qubits is still a small device, and two-qubit entangling gates are not the focus of this paper. Whether pairwise calibration still works when arrays grow by orders of magnitude remains untested.
JWST finds three feeding black holes in galaxy J0148-4214 at z=5.02
An MPE-led team using JWST/NIRSpec integral-field spectroscopy reports three actively accreting supermassive black holes in J0148-4214, a galaxy at redshift 5.02, seen about 1.2 billion years after the Big Bang. Two sit near the center, 620 light-years apart in projection; a third lies about 5,500 light-years out. Masses are estimated at roughly 80 million, 600,000 and 2 million solar masses. The 600,000-solar-mass object appears to accrete above the Eddington limit.
Hydrogen-line kinematics, not direct imaging, revealed the pair: spectro-astrometry split a complex central emission pattern into two sources that a single aperture would have merged. Galaxy stellar mass is about 1.3 billion suns, so the black holes are a large fraction of the baryonic budget. The inner pair is expected to merge within a few hundred million years, a channel that could help explain how some holes grew so large so early.
The outer hole could be a recoil remnant from an earlier merger or an infalling companion; the data do not decide. Super-Eddington accretion estimates depend on models of the line-emitting gas. How common such triples are at z~5 is unknown, because similar systems would have been counted as one AGN without spatially resolved spectroscopy.
214 cuprates show giant domain walls, not a uniform crystal
University of Warwick chemists, working at the ESRF, used scanning 3D X-ray diffraction to map a 214 cuprate superconductor long treated as structurally uniform. The crystal is a patchwork of two slightly different atomic arrangements, separated by domain walls hundreds of times wider than expected. Those walls are thick enough to act as distinct regions and, the authors argue in Physical Review Letters, likely work against superconductivity.
Cuprate theory for four decades has assumed a homogeneous lattice. Bulk measurements average over that hidden texture, so some published properties may need reinterpretation. Similar heterogeneity may exist in other cuprates and in related high-pressure superconductors.
The scan does not yet quantify how much the walls suppress the critical temperature or critical current. Whether growth methods can thin or remove the walls is open. The result is a materials constraint, not a new pairing mechanism.
Sources: ScienceDaily/RIKEN (Nature Physics, 25 Oct 2026); ScienceDaily/MPE (Astronomy, 25 Oct 2026); ScienceDaily/University of Warwick (Physical Review Letters, 24 Oct 2026).