Two U.S. Department of Energy quantum research centers have jointly demonstrated a set of advances in ion-trap technology that could remove key barriers to building quantum computers with millions of qubits. The work, announced in February 2026, combines ultra-high-fidelity gate operations that skip a costly cooling step with new chip-scale hardware for moving and controlling ions, directly targeting the engineering bottleneck that has kept ion-trap machines small. Taken together, the results suggest that large, practical ion-trap quantum processors may be closer than many in the field expected.
Ultra-High Fidelity Without the Cooling Tax
Ion-trap quantum computers manipulate charged atoms suspended in electromagnetic fields, and their logic operations, called gates, must hit extreme accuracy targets before the machines can run useful algorithms. A persistent obstacle has been that two-qubit gates, the operations most prone to error, typically require ions to be cooled to their quantum ground state before each cycle. That cooling step is slow and resource-intensive, and it becomes a serious bottleneck as systems grow. Researchers addressed this problem head-on with what they call the “smooth gate” method, which reshapes the laser pulses driving the gate so that performance stays high even when ions retain significant thermal energy. Their arXiv preprint reports an estimated two-qubit gate error of 8.4(7) × 10-5, corresponding to fidelity above 99.99%.
What makes that number especially striking is the conditions under which it holds. The team showed that the smooth gate maintains error at or below 5 × 10-4 even when the average phonon occupation, a measure of residual thermal motion, reaches approximately 9.4. In practical terms, that means the gate tolerates nearly ten times more thermal noise than a ground-state-cooled ion would carry, yet still operates well within the error thresholds that quantum error correction schemes demand. Eliminating mandatory ground-state cooling for every gate cycle could dramatically speed up computation and simplify the control electronics needed for each qubit, both of which are critical if ion-trap architectures are to compete at the million-qubit scale.
Chip-Scale Penning Traps Enable 2D Ion Transport
High-fidelity gates alone do not solve the scaling problem. Ions also need to be moved, sorted, and positioned across a processor without losing their quantum information. Most ion traps today use radio-frequency electric fields to confine ions, but RF-driven traps face engineering limits when designers try to pack many trapping zones onto a single chip. A separate line of research published in Nature demonstrated a micro-fabricated surface-electrode Penning trap that replaces RF confinement with a 3 T magnetic field. The device achieved quantum control of an individual ion and, critically, flexible two-dimensional transport of ions above the chip surface with very low heating. This shows that ions can be steered around corners and along complex paths without losing coherence.
Low heating during transport matters because every bit of unwanted energy injected into an ion degrades the qubit it encodes. By using a strong static magnetic field instead of oscillating RF voltages, the Penning trap design avoids a major source of electrical noise that plagues conventional surface traps. The result is a hardware platform where ions can be shuttled across a flat chip in two dimensions, opening the door to grid-like processor layouts rather than the linear chains that dominate current systems. Combined with the smooth gate’s tolerance for thermal noise, this pairing of technologies could relax cooling requirements at both the gate and transport stages of a computation cycle, making it more realistic to orchestrate millions of qubits spread across a dense chip.
Multi-Zone Photonic Control on a Single Device
Even with better gates and quieter transport, a large ion-trap processor needs to perform operations on many qubits simultaneously in different physical locations on the chip. A study accepted by Physical Review X tackled this requirement by demonstrating multi-zone coherent control using integrated photonic components inside a quantum charge-coupled device, or QCCD, architecture. The team transported ions between zones separated by 375 micrometers in just 200 microseconds and showed simultaneous control of two ions in separate zones with low crosstalk, indicating that tightly packed zones can still be addressed independently.
The QCCD model is the leading blueprint for scaling ion-trap computers. It treats the processor like a classical CCD camera chip: ions are shuffled between specialized zones for storage, gate operations, and readout. Embedding the photonic control elements directly into the chip, instead of relying on bulky external optics, is a prerequisite for building devices with hundreds or thousands of zones. The 200-microsecond transport time is fast enough that ion shuttling does not become the dominant time cost in a computation, and the demonstration of parallel operations in separate zones shows the architecture can support the kind of concurrent processing that large algorithms require. When combined with two-dimensional transport and thermally robust gates, this style of integrated control sketches a path toward modular ion-trap tiles that could be networked into far larger machines.
DOE Centers Join Forces on Scaling Strategy
These individual technical results gain strategic weight through a coordinated effort by two U.S. national quantum research hubs. A joint announcement from the Quantum Science Center and the Quantum Systems Accelerator describes how researchers, including teams at Fermi National Accelerator Laboratory, are aligning their work around ion-trap arrays designed for larger systems. Instead of treating each laboratory’s experiment as a standalone demonstration, the centers are deliberately knitting together advances in gates, transport, and control into a coherent roadmap for scaling. That roadmap emphasizes architectures that can be manufactured with semiconductor-style processes and controlled with integrated optics and electronics, not lab-scale optical tables.
The Department of Energy has underscored the urgency of this approach. In a related release, officials stress that while ion-trap systems offer some of the highest fidelities among quantum platforms, scaling them to millions of qubits remains a central unsolved challenge. Today’s leading devices still rely on architectures that were never intended to host such vast numbers of qubits, forcing researchers to rethink everything from chip layout to cryogenic packaging. By pooling expertise across national labs and universities, the DOE centers aim to accelerate the transition from bespoke experimental setups to reproducible, scalable hardware that industry partners can eventually adopt.
Federal Programs Push Toward Million-Qubit Architectures
The federal investment in ion-trap scaling extends beyond DOE-backed centers into broader U.S. research programs. The Defense Advanced Research Projects Agency has launched a multi-year effort focused on quantum error correction and modular architectures, seeking designs that can tolerate realistic noise levels while still performing complex algorithms. Ion-trap systems, with their demonstrated high-fidelity gates and long coherence times, are natural candidates for these error-corrected architectures, but only if they can be engineered to handle the immense overhead of logical qubits. Program calls emphasize the need for integrated control, fast ion shuttling, and robust interconnects between modules, all of which align closely with the recent DOE-backed demonstrations.
Additional federal solicitations, such as a separate funding opportunity for advanced quantum hardware, point toward a strategy that couples fundamental physics experiments with aggressive engineering milestones. By tying support to concrete performance targets (like gate fidelities above 99.99%, microsecond-scale transport, and scalable chip layouts), these programs aim to compress the timeline from laboratory proof-of-principle to deployable technology. The convergence of smooth, thermally tolerant gates, chip-scale Penning traps, and integrated multi-zone photonics suggests that ion-trap quantum computers are moving into a new phase, where the central questions are less about basic feasibility and more about how quickly researchers can turn these building blocks into processors with millions of reliable qubits.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
