Researchers at the University of Hong Kong (HKU) and the Centre for Advanced Semiconductors and Integrated Circuits (CASIC) have announced a major breakthrough in cryogenic electronics: the development of a programmable neuromorphic hardware platform that can operate at temperatures as low as 10 millikelvin (-459.65°F), close to absolute zero. This innovation holds significant promise for the future of quantum computing and deep-space exploration.

Harnessing negative differential resistance in Silicon Carbide

The research team, led by Professor Yuhao Zhang and PhD student Xin Yang from HKU’s Department of Electrical and Computer Engineering, has successfully harnessed negative differential resistance (NDR) in industry-standard Silicon Carbide (SiC) MOSFETs.

Their approach enables a single transistor to mimic the energy-efficient “spiking” behavior of biological neurons at ultra-low temperatures. This is the first time such behavior has been demonstrated in a standard transistor at cryogenic levels.

Addressing quantum computing’s wiring bottleneck

Quantum computers require extremely sensitive qubits, maintained at temperatures near absolute zero, to function effectively. However, traditional silicon-based controllers generate significant heat and consume high power, forcing them to be positioned several meters (over a mile in large installations) away from the qubits. This distance introduces a wiring bottleneck that restricts scalability and performance.

“Our work introduces a hardware platform that can be integrated alongside quantum processors,” said Prof. Zhang. “By using unique carrier dynamics in silicon carbide, we can create circuits that are thousands of times more energy-efficient than conventional electronics, significantly reducing the thermal load on cryogenic systems.”

Stable and scalable technology

The key discovery is that SiC MOSFETs, when cooled below 2 Kelvin (-456.67°F), exhibit a strong “S-shape” NDR behavior due to electron-donor impact ionization (EDII). Unlike other technologies that require heat to function, this mechanism is intrinsic to the material, making it stable and repeatable even across different manufacturing batches.

“This is a robust and scalable approach,” said Mr. Yang. “Silicon carbide is already widely used in electric vehicles and power grids, so we can use existing industrial foundries to produce these cryogenic chips on 300-mm (11.8-inch) wafers.”

Enabling complex cryogenic networks

The study shows that these artificial neurons can be cascaded to form larger networks, allowing for complex, local data processing at ultra-low temperatures. This advancement is expected to greatly improve quantum error correction and real-time quantum control, both crucial for advancing quantum computing technology.

Applications in deep-space exploration

Beyond quantum computing, these rugged circuits have potential uses in deep-space missions, where electronics must withstand extreme cold, such as on the lunar surface or in the outer solar system.

The full findings appear in Nature Communications, in a paper titled “Cryogenic neuromorphic circuits using gate-controlled negative differential resistance in silicon carbide.”

The world’s first nuclear clocks have ticked. A team of physicists has demonstrated a working timekeeping device regulated not by orbiting electrons — as in conventional atomic clocks — but by transitions within an atomic nucleus itself, a feat decades in the making.

How a nuclear clock differs from an atomic clock

Standard atomic clocks measure time by tracking the frequency at which electrons in atoms like cesium or rubidium flip between energy states. These devices are extraordinarily accurate, losing roughly 1 second every 300 million years. A nuclear clock operates on a fundamentally different principle: it uses the energy transition of a nucleus rather than its electron shell.

The specific nucleus involved is thorium-229, which has a uniquely low-energy nuclear transition — known as an isomeric transition — that falls within the range of ultraviolet lasers.

Most nuclear transitions release gamma rays at energies far too high for laser control, making them impractical for timekeeping. Thorium-229 is a rare exception: its ground state and isomeric state are separated by only about 8.4 electron volts, allowing researchers to drive the transition optically.

In the nuclear clock, a laser is tuned to match this precise transition frequency. When the nucleus absorbs and re-emits photons at that frequency, those oscillations serve as the clock’s ticks.

Because the nucleus is thousands of times smaller than an electron orbit, it is far less sensitive to external electromagnetic interference — a property that could, in principle, allow nuclear clocks to surpass atomic clocks in long-term stability.

Why thorium-229 took so long to harness

The nuclear clock concept was first proposed in the 1970s, but the exact energy of the thorium-229 isomeric transition was not known with sufficient precision to target with a laser until recently. A series of measurements over the past decade progressively narrowed the uncertainty.

In 2024, a team at JILA — a joint institute of the University of Colorado Boulder and the National Institute of Standards and Technology (NIST) — pinned down the transition frequency precisely enough to enable laser excitation of the nucleus.

The current demonstration confirms that thorium-229 nuclei embedded in a calcium fluoride crystal can be repeatedly excited and interrogated using vacuum ultraviolet light.

The calcium fluoride host is transparent at the required wavelength and holds the nuclei in a rigid lattice that minimizes vibrational noise. The clock signal emerges from the collective oscillation of many nuclei simultaneously, which improves the measurement statistics.

According to researchers involved in the work, the present prototype is not yet more accurate than the best optical atomic clocks, but the architecture is now proven. Refinements in laser linewidth and crystal quality are expected to further improve precision.

Probing physics beyond the standard model

The appeal of a nuclear clock extends beyond timekeeping. Because the nucleus interacts strongly with the weak and strong fundamental forces, a thorium-229 clock is sensitive to variations in fundamental constants — quantities like the fine-structure constant and the strong-force coupling — in ways that electron-based clocks are not.

This makes the device a potential instrument for detecting ultralight dark matter. Certain dark matter candidates are theorized to cause tiny, periodic fluctuations in fundamental constants as the Earth moves through a dark matter field.

Comparing a nuclear clock against an optical atomic clock over time could reveal such oscillations. The approach complements particle detector experiments and quantum sensors now used to detect phenomena once considered beyond measurement.

A nuclear clock could also serve as a testbed for whether fundamental constants drift over cosmological timescales — a question with direct implications for theories of dark energy and the evolution of the universe.

Engineers at the University of Texas at Austin have built a jacket that captures moisture from the surrounding air and converts it into drinkable water — a development that could have practical consequences for soldiers, hikers, and disaster-relief workers operating far from any water source.

How the textile works

The jacket uses a hygroscopic sorbent woven directly into the fabric. These materials have a strong thermodynamic affinity for water vapor, adsorbing moisture from the atmosphere even at relatively low humidity levels.

Once the sorbent is saturated, a low-grade heat source — potentially body heat or a small solar element integrated into the garment — triggers desorption, releasing the captured water vapor into a compact condensation chamber where it accumulates as liquid. The resulting water is then channeled through a filtration stage before it is considered safe to drink.

The design avoids the energy-intensive refrigeration cycles that conventional atmospheric water generators rely on, which makes a wearable form factor physically plausible.

Traditional atmospheric water-harvesting systems typically require significant electrical input to cool a surface below the dew point; this jacket sidesteps that requirement by using sorption-desorption kinetics.

Material selection and engineering constraints

Choosing the right sorbent is the central engineering problem. The material must adsorb water rapidly, release it cleanly under mild thermal stimulation, survive repeated wetting-and-drying cycles without degrading, and remain flexible enough not to compromise the garment’s wearability.

Metal-organic frameworks (MOFs) and certain hydrogel composites have emerged as candidate materials in the broader atmospheric water-harvesting literature, and similar classes of materials are likely candidates for this type of application, though the specific formulation used by the UT Austin team is detailed in their official announcement.

Yield rate is a genuine constraint. The volume of water a garment-scale sorbent can collect per day is governed by the exposed surface area, the ambient relative humidity, and the cycling frequency of adsorption and desorption.

In arid environments — precisely where a portable water supply is most critical — lower atmospheric humidity directly limits output, so the technology performs best in humid or semi-arid conditions rather than true desert climates.

Structural integration with the garment

Embedding a functional water-harvesting system into a garment without making it impractical to wear requires careful attention to weight distribution, vapor permeability, and thermal management. If the outer layer is too occlusive, it traps sweat, raising the wearer’s core temperature.

The UT Austin design must balance the need to expose the sorbent to ambient air against the thermal comfort requirements of the person wearing it — a tradeoff that textile engineers working on responsive smart fabrics have encountered in other contexts.

Contamination risk and water quality

Atmospheric water harvesting is not inherently free of contamination risk. Air in industrial or wildfire-affected areas can carry particulates, volatile organic compounds, and heavy metals, which a sorbent material may co-adsorb with water vapor.

The filtration stage integrated into the jacket’s collection system is therefore not a minor accessory — it is a critical safety component. Whether the filtration approach is sufficient across a range of realistic environmental conditions, and how the filter itself is serviced or replaced in the field, are questions that field trials will need to address before the technology sees wide deployment.

Potential applications and near-term limitations

Military logistics, backcountry search-and-rescue, and humanitarian response in regions with disrupted water infrastructure are the most direct use cases. Each imposes different durability, laundering, and maintenance requirements.

A garment that can survive the wash cycles of standard military field use while retaining its hygroscopic function is a considerably harder engineering target than a laboratory prototype.

This research was first published in the journal Science Advances.