Skip to content

In-Substrate Imaging of Diamond–hBN FET Current via Wide-Field Quantum Diamond Microscopy

FieldValue
JournalACS Applied Electronic Materials
Published2026-06-11
AuthorsAnuj Bathla, Subrat Kumar Pradhan, Ajit Kumar Dash, Prabhat Anand, M Girish Chandra
DOI10.1021/acsaelm.6c00194

Wide-field NV magnetometry resolved current flow beneath an hBN gate in an operating hydrogen-terminated diamond FET with about 1 µm spatial resolution. Under the 532 nm laser required for NV readout, peak drain current rose from approximately 600 µA to nearly 900 µA. The apparent threshold voltage shifted by about 8.60 V at Vds = -15 V. The maps showed contact current crowding, a nearly uniform channel current beneath uniform hBN, and spatial modulation beneath a nonuniform flake.

Current imaging resolves the buried channel

Section titled “Current imaging resolves the buried channel”

At Vds = -12 V and Vgs = -5 V, Device 1 carried stable current during magnetic imaging. The source and drain interface produced localized magnetic field lobes and a reconstructed injection hotspot. The hotspot broadened as current entered the two-dimensional hole gas, consistent with spreading resistance at the contact edge. Beneath the hBN flake, the magnetic profile was smooth and the reconstructed current was nearly uniform across the channel width. Current streamlines followed the channel direction.

Device 2 used a nonuniform hBN flake and was measured at Vds = -19 V and Vgs = -5 V. Its magnetic and reconstructed current maps contained regions of stronger and weaker current. The authors associate these variations with thicker or partly delaminated hBN, which weakens local gate coupling and leaves more current in the channel. These comparisons separate dielectric quality from the current paths seen in Device 1.

Device fabrication puts the sensor below the channel

Section titled “Device fabrication puts the sensor below the channel”

The devices were fabricated on the NV side of electronic-grade single-crystal (100) diamond. A near-surface NV layer lay about 1 µm below the surface, placing the sensor close to the two-dimensional hole gas. Six devices were made on one substrate. Four used hBN gate flakes, and two were ungated. The source and drain pattern defined a channel about 10 µm long.

Hydrogen termination used an MPCVD system with hydrogen plasma for 30 to 40 minutes. Microwave power was 2 to 3 kW, chamber pressure was 700 Torr, and substrate temperature was 700 to 750 °C. A hydrogen flow of 100 to 500 sccm continued during cooling. Air exposure for 12 to 24 hours then produced the surface-transfer-doped hole channel.

Exfoliated hBN flakes were 20 to 30 nm thick and were dry-transferred over the channel. The final gate used a 30 nm Ti layer beneath 100 nm Au. Earlier source and drain patterning also used Ti/Au pads. The paper therefore combined hydrogen-terminated diamond, a van der Waals dielectric, and metal contacts in one top-gated structure.

Magnetic imaging and current reconstruction

Section titled “Magnetic imaging and current reconstruction”

Wide-field sensing used continuous-wave optically detected magnetic resonance. A 532 nm laser at about 0.4 W illuminated the NV layer through a 100× objective with NA = 0.9. The microscope supplied 36× magnification and an approximately 100 µm × 100 µm field of view. Microwave excitation near 2.87 GHz was applied through a PCB loop antenna, and a lock-in CMOS camera collected NV photoluminescence. Measurements ran under ambient laboratory conditions.

The local field along each selected NV axis came from the resonance-frequency difference between current-on and current-off states. A Sm-Co static field lifted the NV transition degeneracy. The authors measured projections along three accessible NV axes, recovered the full vector field, and reconstructed the two-dimensional current density. The inverse problem was ill-posed, so the reconstruction used Green-function kernels, a divergence penalty, a BM3D denoising prior, and an ADMM iterative solution.

Laser illumination changes the transfer curve

Section titled “Laser illumination changes the transfer curve”

Without laser illumination, the transfer curve showed a drain current that increased as Vgs became more negative. The channel saturated at larger negative Vds. Its positive threshold indicated depletion-mode operation with a conductive surface channel at zero gate bias. The reported bias range covered Vds = 0 to -15 V and Vgs = +3 to -9 V.

With the laser on, drain current increased over the full gate range. Peak current rose from approximately 600 µA to nearly 900 µA. The apparent threshold shifted toward smaller positive values by about 8.60 V at Vds = -15 V, with a comparable shift at Vds = -10 V. Near Vgs = -8 V, the current showed a slight reduction in the high-current regime. The authors suggest self-heating and a corresponding mobility reduction.

Supporting Information attributes most of the optical change to NV charge-state cycling. Each completed cycle generates one electron-hole pair. Photo-generated holes can diffuse to the hydrogen-terminated surface and raise the hole density of the 2DHG. The calculated NV contribution is large enough to explain a substantial part of the threshold shift and current increase. Surface and hBN-interface charging may add to the effect.

Gate sweep exposes local dielectric control

Section titled “Gate sweep exposes local dielectric control”

Gate-dependent imaging was performed at Vds = -10 V while Vgs was swept. Each selected operating point had a corresponding magnetic image and transfer-curve value. The overall NV magnetic signal decreased as the drain current fell with gating. Spatially, current suppression began earlier and was stronger near the hBN edges than in the channel center. The authors relate this response to local thickness changes or imperfect hBN-diamond coupling.

The authors identify their experiment as the first direct magnetic imaging of current inside an operating diamond FET with an hBN gate. Earlier wide-field NV work imaged fields from metallic microstructures, graphene, and microcoils. The new result places sensing inside the active material and records both electrical current and local field at each selected bias. Their uniform-versus-nonuniform hBN comparison also provides a device-specific reference for separating contact effects from dielectric effects.

Several limits remain. The experiment covered six fabricated devices, with four gated and two ungated. The authors state that a standoff comparable to or larger than the spatial sampling scale suppresses high-spatial-frequency field components. Regularization was therefore required to recover stable current maps. The optical readout also changed the channel electrostatics, so laser-off measurements are needed to separate imaging artifacts from the intrinsic transfer curve. The photo-induced model is consistent with the data, but surface and interface charging may also contribute.

Implications for diamond device engineering

Section titled “Implications for diamond device engineering”

Diamond selection for this measurement requires more than a conductive bulk substrate. The active surface needs hydrogen termination, while the sensing layer needs near-surface NV centers at about 1 µm depth. The buried channel must remain close enough to that layer for magnetic contrast and current reconstruction. hBN thickness, adhesion, and coverage affect local gate efficiency, as the two devices show.

Device engineering must also account for the 532 nm measurement light. Laser illumination can change the apparent threshold and channel current. The paper treats optical modulation through NV photoionization as a coupling mechanism between the sensing layer and surface transport. It does not establish an optimized laser-assisted operating mode.

For buried-transport diagnosis, simultaneous current and field measurements reveal contact crowding, edge suppression, and local dielectric defects that terminal current alone cannot locate. The authors also state that gate-dependent imaging can capture pinch-off, percolation, and localized breakdown in real space.

Operando wide-field NV current imaging of an hBN-gated FET requires an electronic-grade single-crystal (100) diamond substrate, MPCVD hydrogen termination, and a near-surface NV layer about 1 µm deep. The device also uses Ti/Au metallization, and the 6CCVD diamond materials catalog lists MPCVD-grown single-crystal and polycrystalline plates and wafers up to 125 mm with Au and Ti options. The NV layer and hBN-integrated FET stack remain device-specific requirements for engineering support.

View Original Abstract

We demonstrate wide-field magnetic imaging of current flow in hydrogen-terminated diamond field-effect transistors (FETs) through in-substrate nitrogen-vacancy (NV) centers. Hydrogen termination of the diamond surface induces a two-dimensional hole gas (2DHG), while an ensemble of near-surface NV centers located ∼1 μm below the surface enables operando magnetic imaging of current flow with micrometer-scale spatial resolution. The FETs were electrically characterized over a range of drain-source biases, V ds = 0 to −15 V, and gate voltages, V gs = +3 to −9 V, followed by in situ wide-field NV magnetometry during device operation. Magnetic field maps and reconstructed current-density distributions directly visualize current injection at the source-drain contacts and transport beneath the hBN-gated channel. Magnetic field maps reveal current-density variations in the channel region owing to non-uniformities or defects in the gate dielectric. In addition, we observe a pronounced enhancement of the drain current (∼600 to 900 μA) and a shift in the apparent threshold voltage during laser illumination, reflecting photo-induced changes in channel electrostatics. By correlating gate-dependent magnetic images with simultaneous electrical measurements, we directly link spatial current distributions to FET transfer characteristics, providing new insight into buried interface transport and non-uniform gating effects in the transistor channel. As the methodology is compatible with top-gated FETs, it can be used to map channel current distributions with micrometer resolution in emerging channel materials, such as 2D materials and wide-bandgap channels, and establish wide-field NV magnetometry as a powerful platform for probing charge transport in transistors and van der Waals dielectric heterostructures.