Transistors

Researchers at the National University of Singapore (NUS), led by Assistant Professor Ahmet Avsar of the university's Centre for Advanced 2D Materials, have combined a record-fidelity graphene spin-transport platform with magnetic-proximity band engineering to move graphene closer to practical spin-logic and spin-memory devices, across two complementary studies. The work targets one of graphene spintronics' core limitations: interfacial disorder at the electrical contacts that inject and detect spin, which has historically scrambled spin information before it can be read out electrically.

The first study rebuilt the graphene spin-device fabrication process around an inert-glovebox van der Waals assembly, laminating and cleaning the stack to produce atomically flat hexagonal boron nitride (h-BN) tunnel barriers rather than the oxide barriers more commonly used in graphene spin valves. That interface quality translated directly into device performance: nonlocal spin signals reached up to 1.6 kΩ at 2.5 K, spin polarization approached 90% (89% in the lead device), spin lifetime measured about 2.04 nanoseconds with a spin diffusion length of about 4.74 μm, and gate-tunable magnetoresistance exceeded 80%. Critically for eventual device use, the effect persisted at room temperature, where the same device retained a nonlocal spin resistance of about 160 Ω and roughly 42% spin polarization.

UK-based Paragraf has launched two new products expanding its graphene sensing portfolio: the PMF2002 Graphene Field-Effect Transistor (GFET) Platform and the Enterprise Reader data acquisition system, designed to work together to take users from sensor development through to data analysis.

The PMF2002 builds on Paragraf's earlier PMF2000 platform with a 12-channel multiplexing architecture that uses a common gate with separate source and drain connections, allowing simultaneous measurements across multiple sensing channels from a single sample while requiring less sample material. The sensing surface can be functionalized by customers, letting the platform be adapted for target-specific detection across applications including water quality monitoring, soil analysis, gas detection, and molecular biosensing.

A research team led by the University of California, San Diego, has demonstrated the first direct integration of an insect olfactory receptor with a graphene field-effect transistor (gFET), creating a label-free biosensor capable of selectively detecting small organic compounds (SOCs) with high sensitivity.

The work centers on MhOR5, an odorant receptor from Machilis hrabei, which was produced at scale for the first time. The researchers obtained the protein in a stable tetrameric form with approximately 80% purity, maintaining structural integrity for up to 3 months under frozen storage. This stability enabled consistent device functionalization and testing. To translate biological recognition into an electrical signal, the team fabricated wafer-scale graphene transistors and subjected 6,655 devices to electrical screening, with more than 80% passing quality control in each batch. Each chip contained an array of fifteen graphene channels (100 × 10 µm), along with coplanar gate and sensing electrodes to ensure stable liquid-gated operation.

A team of researchers, led by IMB-CNM-CSIC and ICN2, has developed a graphene-based bidirectional neural interface that can simultaneously record and modulate brain activity, overcoming a longstanding limitation in neurotechnology. The device combines graphene solution-gated field-effect transistors (gSGFETs) with nanoporous reduced graphene oxide (rGO) microelectrodes in a single, flexible platform, enabling both high-sensitivity monitoring and effective stimulation.

Neural interfaces are already used clinically to treat neurological disorders, but most current systems remain unidirectional. They typically deliver stimulation using fixed parameters, without the ability to adapt in real time to ongoing brain activity. Even in systems that can both stimulate and record, performance is often constrained - particularly when it comes to detecting very low-frequency signals, which are increasingly recognized as important biomarkers. The newly reported device addresses these challenges by integrating two complementary graphene technologies. 

Researchers from Princeton University, University of Michigan, California State University and Japan's National Institute for Materials Science have introduced Qumus, an embodied AI system that can autonomously create graphene and fabricate atomically thin graphene devices in a robotic mini-laboratory.

Qumus AI architecture and fully robotic minilab. a Defining characteristics of an AI experimentalist. b Key self-evolving modules of Qumus, including LLM-agents, memory and knowledge systems, and skills including instrumental workflows and materials/devices realization recipes. c Qumus architecture for efficient multi-agent collaboration and robust performance. d A compact, fully robotic minilab consisting of vacuum- and temperature-controlled stages for 2D material mechanical exfoliation, optical flake search, flake transfer and stacking, along with robotic arms, storage modules, cameras, and microscope systems. Image from : arXiv

Qumus is built around the complete graphene workflow: from exfoliating bulk crystals to isolating single-layer flakes and stacking them into functional van der Waals (vdW) devices, all without human intervention. The system combines generative AI, computer vision and robotics to handle the labor-intensive steps that typically limit graphene research, such as flake discovery, thickness assessment and submicron alignment during transfer.

Paragraf has announced the launch of its latest Graphene Field Effect Transistor (GFET), the PMF2000, marking a step forward in scalable graphene device manufacturing.

The PMF2000 builds on Paragraf’s existing GFET technology, retaining its hallmark contamination-free graphene while introducing production on six-inch silicon wafers. This transition is enabled by the company’s new large-wafer manufacturing facility in Huntingdon - described as the world’s first graphene foundry. The move to larger wafers improves device yield and consistency, setting a new benchmark for quality and enabling reliable high-volume production.

Researchers at Penn State have developed a graphene-based field-effect transistor (GFET) architecture that improves sensor stability and sensitivity in liquid environments, marking a step toward real-time molecular detection for health, environmental, and industrial applications.

The team engineered a dual‑gate GFET that integrates a high‑κ hafnium dioxide (HfO₂) local back gate with an electrolyte top gate, coupled through a real‑time feedback control loop. This novel configuration enables capacitive signal amplification while suppressing gate leakage and low‑frequency noise - two sources of instability that have long limited the performance of conventional single‑gate GFETs used in liquid sensing.

Researchers from Nanjing University of Aeronautics and Astronautics have demonstrated an atom‑thin sliding ferroelectric transistor that can reliably store 3,024 distinct, non‑volatile polarization states at room temperature - a record for ferroelectric neuromorphic hardware. The device is built from a well‑aligned monolayer graphene channel on hexagonal boron nitride (hBN), forming a moiré superlattice that enables fine, electrical control over ferroelectric polarization and charge localization within just a few atomic layers.

Performance comparison of Gr/hBN device and schematic diagram of its working mechanism. Credit: Nature Electronics (2026)

The transistor consists of an aligned graphene monolayer atop ferroelectric hBN, with source, drain and gate electrodes defined by standard nanofabrication, including electron‑beam evaporation for the metal contacts. Graphene serves as a high‑mobility, atomically thin channel whose Fermi level can be tuned electrostatically, while the underlying hBN provides sliding‑induced ferroelectricity and an atomically flat, low‑disorder interface. The lattice mismatch of about 1.8% between graphene and hBN generates a long‑wavelength moiré potential, which plays a central role in localizing injected carriers and stabilizing multiple polarization configurations.

As power consumption by machine learning technologies rises, demand grows for AI devices with low power consumption and high computational performance. "Physical reservoirs" are AI devices that perform efficient brain-inspired information processing called reservoir computing and are interesting thanks to their low computational load and low power consumption. However, their lower computational performance compared to software processing has thus far been a drawback.

A research team from NIMS, Tokyo University of Science, and Kobe University recently developed an ion-gel/graphene electric double layer (EDL) transistor-based ion-gating reservoir (IGR) that achieved high computational performance comparable to that of deep learning while reducing the computational load by orders of magnitude. By combining graphene, which has high electron mobility and ambipolar behavior, and an ion gel, various responses with different speeds (ions and electrons moving in various manners) develop through complex interactions, enabling the device to respond to input signals with time constants (rates of change) that vary over an extremely wide range.

Researchers from the University of Illinois Chicago, Wayne State, and Northwestern have shown that random defects in graphene transistors can be harnessed for next-generation hardware security. Their work demonstrates how the intrinsic disorder in graphene can generate unique electromagnetic “fingerprints,” signals so tied to each device’s atomic structure that they cannot be copied or predicted.

Traditional digital encryption relies on stored keys that can be stolen or cracked. By contrast, this graphene-based system uses a physical unclonable function (PUF), a hardware identity formed by the material’s natural randomness. When probed wirelessly, each graphene transistor produces a distinctive radio signal encoding its physical quirks - residues, strain, charge variations - into a one-of-a-kind signature.