Cryogenic temperature dependence and hysteresis of surface-trap-induced gate leakage in GaN high-electron-mobility transistors

Basic Information

• Paper ID: 2510.14456
• Title: Cryogenic temperature dependence and hysteresis of surface-trap-induced gate leakage in GaN high-electron-mobility transistors
• Authors: Ching-Yang Pan, Shi-Kai Lin, Yu-An Chen, Pei-hsun Jiang (National Taiwan Normal University)
• Classification: cond-mat.mes-hall, physics.app-ph
• Publication Date: October 16, 2024
• Paper Link: https://arxiv.org/abs/2510.14456

Abstract

This study systematically maps various mechanisms of surface-trap-induced gate leakage in GaN high-electron-mobility transistors (HEMTs) across a temperature range from room temperature to cryogenic conditions. Two-dimensional variable-range hopping (2D-VRH) conduction is observed at small gate biases. At higher reverse gate biases, leakage above 220 K is dominated by Poole-Frenkel emission (PFE), but gradually transitions to trap-assisted tunneling (TAT) below 220 K due to trap freezing effects. The trap barrier height extracted from upward gate sweeps is 0.65 V, approximately 12% higher than that from downward sweeps. The gate leakage current exhibits clockwise hysteresis loops as a function of gate bias above 220 K, while counterclockwise hysteresis loops appear below 220 K. This remarkable reversal of hysteresis direction is fully explained through trap mechanisms.

Research Background and Motivation

1. Problems to be addressed:
   • Gate leakage mechanisms in GaN HEMTs at cryogenic temperatures remain insufficiently studied
   • Surface-trap-induced gate leakage poses serious reliability concerns for devices
   • Systematic investigation of gate leakage hysteresis behavior at cryogenic temperatures is lacking
2. Importance of the problem:
   • Growing demand for cryogenic electronic devices in aerospace, quantum computing, and superconducting systems
   • GaN HEMTs exhibit superior performance at low temperatures, but trap effects limit device reliability
   • Gate leakage reduces breakdown voltage and increases off-state power consumption
3. Limitations of existing approaches:
   • Previous studies primarily focused on gate leakage temperature dependence above room temperature
   • Limited in-depth research on surface gate leakage at cryogenic temperatures
   • Insufficient understanding of the physical mechanisms underlying gate leakage hysteresis
4. Research motivation:
   • Provide theoretical foundation for development and application of cryogenic GaN devices
   • Reveal physical mechanisms of gate leakage across different temperature regimes
   • Establish correlations between trap mechanisms and hysteresis behavior

Core Contributions

1. Systematically mapped various gate leakage mechanisms in GaN HEMTs across a temperature range from 300 K to 1.5 K
2. Discovered and explained the mechanism transition from Poole-Frenkel emission to trap-assisted tunneling around 220 K
3. First observation of temperature-dependent reversal of gate leakage hysteresis loop direction around 220 K
4. Provided precise measurement of trap barrier height (0.65 V) and its relationship with sweep direction
5. Established a physical correlation model between different leakage mechanisms and hysteresis behavior

Methodology Details

Device Structure and Fabrication

• Substrate: 4 μm thick GaN/AlGaN superlattice buffer layer on Si(111) substrate
• Heterostructure: GaN (200 nm)/Al₀.₂Ga₀.₈N (25 nm) forming a two-dimensional electron gas channel
• Electrodes: Ti/Al/Ti/Au source-drain electrodes, Ni/Au/Ti gate electrode
• Passivation: 450 nm SiO₂ passivation layer
• Device dimensions: Gate width 100 μm, gate length 3-5 μm, source-gate distance 3-5 μm, gate-drain distance 20-30 μm

Testing Methods

• Temperature range: Systematic measurements from 300 K to 1.5 K
• Electrical characterization: I-V characteristics, C-V characteristics, mobility and carrier concentration measurements
• Gate leakage testing: Bidirectional gate voltage sweeps, testing at different sweep rates
• Stress testing: Time-dependent measurements at fixed gate bias

Theoretical Models

1. Two-Dimensional Variable-Range Hopping (2D-VRH): $I_{2D-VRH} = I_0 \exp\left[-\left(\frac{T_0}{T}\right)^{1/3}\right]$
2. Poole-Frenkel Emission (PFE): $I_{PFE} \propto E \exp\left[\frac{q}{kT}\left(\sqrt{\frac{qE}{\pi\epsilon}} - \phi_{PFE}\right)\right]$
3. Trap-Assisted Tunneling (TAT): $I_{TAT} \propto \exp\left(-\frac{8\pi\sqrt{2m^*q}}{3h}\frac{\phi_{TAT}^{3/2}}{E}\right)$

Experimental Setup

Device Parameters

• Representative sample: LG = 5 μm, LSG = 3 μm, LGD = 30 μm
• Room temperature carrier density: ~9×10¹² cm⁻²
• Room temperature mobility: 1108 cm²/(V·s)
• Mobility at 1.5 K: 2323 cm²/(V·s)

Measurement Conditions

• Temperature step: 25 K intervals above 200 K, 20 K intervals below 200 K
• Gate voltage sweep range: -10 V to +2 V
• Sweep rate: ±0.05 V per 18-250 ms
• Stress duration: up to 90 seconds

Evaluation Metrics

• Relationship between gate leakage current IG and gate voltage VG
• Trap barrier heights φPFE and φTAT
• Hysteresis loop area and direction
• Temperature dependence exponents

Experimental Results

Main Findings

1. Temperature classification:
   • High temperature region (300-220 K): PFE dominated
   • Intermediate temperature region (200-140 K): PFE to TAT transition
   • Low temperature region (120-1.5 K): TAT dominated, temperature dependence vanishes
2. Leakage mechanism identification:
   • VG > 0.3 V: Schottky thermionic emission
   • -1 V < VG < 0.3 V: 2D-VRH mechanism
   • VG < -4 V: PFE (high temperature) or TAT (low temperature)
3. Trap barrier heights:
   • Upward sweep: φPFE = 0.65 V
   • Downward sweep: φPFE = 0.58 V
   • Difference: 12%

Hysteresis Phenomena

1. Hysteresis direction:
   • T > 220 K: Clockwise hysteresis loops
   • T < 220 K: Counterclockwise hysteresis loops
   • T = 220 K: Minimum hysteresis
2. Sweep rate dependence:
   • High temperature: Hysteresis insensitive to sweep rate
   • Low temperature: Strong sweep rate dependence
3. Time dependence:
   • T > 220 K: IG increases with time
   • T < 220 K: IG decreases with time

Physical Mechanism Explanation

1. PFE mechanism (T > 220 K):
   • Electrons thermally excited from traps to conduction band
   • More trapped electrons provide larger carrier reservoir
   • Produces clockwise hysteresis
2. TAT mechanism (T < 220 K):
   • Trap freezing effect, electrons emit via tunneling
   • Increased trap occupancy suppresses tunneling current
   • Produces counterclockwise hysteresis

Related Work

Main Research Directions

1. Cryogenic characteristics of GaN HEMTs
2. Trap mechanisms in semiconductor devices
3. Temperature dependence analysis of gate leakage
4. Hysteresis phenomena in Schottky diodes

Uniqueness of This Work

• First systematic study of gate leakage in GaN HEMTs across an extremely wide temperature range
• Discovery and explanation of temperature-dependent reversal of hysteresis direction
• Establishment of clear boundaries between different leakage mechanisms

Conclusions and Discussion

Main Conclusions

1. Mechanism transition: 220 K is the critical temperature for PFE to TAT transition
2. Hysteresis reversal: The reversal of hysteresis direction directly corresponds to the change in leakage mechanism
3. Trap freezing effect: Trap freezing at low temperatures is the fundamental cause of TAT mechanism and counterclockwise hysteresis
4. Practical value: Hysteresis measurement can serve as a convenient tool for identifying leakage mechanisms

Limitations

1. Material limitations: Study restricted to MOCVD-grown devices with SiO₂ passivation
2. Temperature range: Signal-to-noise ratio limitations in certain temperature intervals restrict precise analysis
3. Theoretical models: Non-uniform electric field distribution affects precise extraction of TAT barrier height

Future Directions

1. Process optimization: Investigate effects of different passivation materials and growth methods
2. Theory refinement: Develop more accurate models of surface electric field distribution
3. Device applications: Develop reliability enhancement strategies based on trap mechanism understanding

In-Depth Evaluation

Strengths

1. Strong systematicity: Covers an extremely wide temperature range, providing a complete physical picture
2. Clear mechanisms: Multiple theoretical model fits clearly identify different leakage mechanisms
3. Novel phenomena: First discovery of temperature-dependent reversal of hysteresis direction
4. Reasonable explanation: Mechanism explanation based on trap physics is logically clear and convincing
5. Practical value: Provides important reference for design and reliability assessment of cryogenic GaN devices

Weaknesses

1. Sample limitations: Only studied one specific device structure and process condition
2. Model simplification: Some complex trap interactions may be overlooked
3. Quantitative analysis: Parameter extraction partially limited by experimental conditions

Impact

1. Academic value: Provides new perspectives and methods for GaN device physics research
2. Application prospects: Important guidance for cryogenic applications such as quantum computing
3. Methodological contribution: Hysteresis measurement as a mechanism identification tool has promotion value

Applicable Scenarios

1. Cryogenic electronics: Aerospace, quantum computing systems
2. Power electronics: High-efficiency power converters
3. RF applications: Low-noise amplifiers, power amplifiers
4. Device reliability: Failure analysis and lifetime prediction

References

The paper cites 93 relevant references covering multiple fields including GaN device physics, trap mechanisms, and cryogenic electronics, providing solid theoretical foundation and comparative references for the research.

Overall Assessment: This is a high-quality experimental physics paper that systematically and thoroughly investigates gate leakage mechanisms in GaN HEMTs at cryogenic temperatures, discovers novel physical phenomena, and provides reasonable theoretical explanations. The research results are of significant importance for advancing GaN device applications in extreme environments.