Schottky Barrier Height Analysis of Diamond SPIND Using High Temperature Operation up to 873 K

At a Glance

Metadata	Details
Publication Date	2020-01-01
Journal	IEEE Journal of the Electron Devices Society
Authors	Mohamadali Malakoutian, Manpuneet Benipal, Franz A. Koeck, R. J. Nemanich, Srabanti Chowdhury
Institutions	Diamond Materials (United States), Stanford University
Citations	25
Analysis	Full AI Review Included

Executive Summary

This research validates the performance and stability of a Diamond Schottky PIN Diode (SPIND) under extreme high-temperature conditions, addressing non-ideal electrical behavior crucial for reliable power electronics design.

Extreme Temperature Stability: The diamond SPIND demonstrated stable operation up to 873 K (600 °C) over 10 thermal cycles (totaling 120 hours) with no degradation in I-V characteristics or contact stability.
High Current Density: The device achieved an excellent forward current density of greater than 3000 A/cm² at 8 V, confirming its suitability for high-power applications.
Improved Conduction: Specific On-Resistance (R_on.sp) decreased significantly from 92 mΩ.cm² at 298 K to 5.7 mΩ.cm² at 873 K (at 1.5 V bias), attributed to increased active dopants at high temperatures.
Non-Ideal Behavior Analysis: The conventional Thermionic Emission (TE) model failed, yielding an extracted Richardson’s constant (A**) three orders of magnitude smaller than the theoretical value (90 A/cm².K²).
Inhomogeneous Barrier Model: Tung’s modified TE model, which accounts for inhomogeneous Schottky Barrier Heights (SBHs), successfully resolved the discrepancy, yielding an A** value (81.16 ± 18.9 A/cm².K²) close to the theoretical limit.
Experimental Validation: Conductive Atomic Force Microscopy (C-AFM) confirmed the presence of inhomogeneous SBHs by visualizing localized, voltage-dependent current patches across the contact area.

Technical Specifications

Parameter	Value	Unit	Context
Maximum Operating Temperature	873	K	Stable operation limit tested
Maximum Forward Current Density	>3000	A/cm²	Device #1, at 8 V
Rectification Factor (298 K)	>10⁹	Dimensionless	Device #1
Rectification Factor (873 K)	~1	Dimensionless	Device #1
Specific On-Resistance (R_on.sp)	5.7	mΩ.cm²	Minimum value achieved at 873 K, 1.5 V bias
Substrate Material	(100) Type IIb Diamond	N/A	Highly Boron-doped (p-type)
Substrate Doping Concentration	~1 x 10²⁰	cm^-3	Boron concentration
N-layer Doping Concentration	<10¹⁹	cm^-3	Phosphorus-doped
Ideal Zero-Bias SBH (Φ_B0)	1.53	eV	Extrapolated from 523-873 K data (Tung’s model)
Extracted Richardson’s Constant (A**)	81.16 ± 18.9	A/cm².K²	Using Modified Richardson’s Plot (Tung’s Model)
Theoretical Richardson’s Constant (A*)	90	A/cm².K²	For holes in diamond

Key Methodologies

The Diamond SPIND fabrication involved specialized CVD growth and mesa isolation techniques optimized for high-temperature performance:

Substrate Preparation: Used (100)-oriented Type IIb diamond substrates grown by High-Pressure, High-Temperature (HPHT) technique, heavily doped with Boron (~1 x 10²⁰ cm^-3).
CVD Layer Growth: Intrinsic (i) and n-type (Phosphorus-doped, <10¹⁹ cm^-3) layers were grown via Chemical Vapor Deposition (CVD).
- i-layer: Grown in multiple stages, reaching pyrometer temperatures up to 807 °C (Stage 3).
- n-layer: Grown at high temperature (1025 °C) and pressure (75 Torr) using CH₄ and TMP/H₂ precursors.
Mesa Isolation: Al hard mask was patterned, followed by Inductively Coupled Plasma / Reactive Ion Etching (ICP/RIE) using O₂/SF₆ plasma. The etch stopped mid-i-layer to minimize sidewall leakage.
Surface Treatment: Samples were immersed in a hot H₂SO₄:HNO₃ (3:1) acid mixture at 220 °C to achieve oxygen termination and remove surface conductive layers.
Contact Metallization: Ti/Pt/Au (50nm/50nm/150nm) contacts were deposited on the n-side via e-beam evaporation and patterned using lift-off. A backside contact was utilized on the conductive p-type substrate.
Contact Annealing: Contacts were annealed at 850 °C to ensure stable, low-resistance interfaces.
Electrical Testing: I-V characteristics were measured from 298 K to 873 K using a high-temperature stage (Linkam T95-PE) and Keithley source/meters.
Microscopic Analysis: Conductive Atomic Force Microscopy (C-AFM, Asylum Research MFP3D) was employed to map local current flow and visualize the inhomogeneous Schottky barrier patches.

Commercial Applications

The demonstrated high-temperature stability, high current density, and robust contact performance make this diamond SPIND technology highly relevant for extreme environment and high-power applications.

High-Power Electronics: Used in power conversion systems, high-voltage switching, and motor drives where high current density and low R_on.sp are critical.
Extreme Environment Operation: Suitable for devices operating in harsh thermal environments, such as downhole drilling equipment, nuclear reactors, and high-temperature industrial processing.
Aerospace and Defense: Specifically cited for potential use in environments like the Venus atmosphere (~737 K), requiring electronics that function reliably above 500 °C.
High-Frequency/RF Power: Diamond’s superior thermal conductivity allows for efficient heat dissipation, enabling high-power density RF devices that maintain performance under heavy load.
Solid-State Lighting (SSL) Drivers: Used in high-efficiency drivers where thermal management is crucial for longevity and performance.

View Original Abstract

In this work, the high temperature performance of a diamond Schottky PIN diode is reported in the range of 298-873 K. The diamond diode exhibited an explicit rectification up to 723 K with an excellent forward current density of >3000 A/cm2. The stability of the diode was investigated by exposing the sample to high temperature cycles (up to 873 K) for more than 10 times (totaling up to 120 hours), which exhibited no change between the I-V characteristics measured in each cycle. The dependence of ideality factor and Schottky barrier height on temperature along with an extracted Richardson’s constant much smaller than the theoretical value (0.0461 A/cm2.K2), motivated us to study the possible reason for this anomaly. A modified thermionic emission model following Tung’s analysis was used to explain the experimental observations. The model assumed the presence of inhomogeneous Schottky barrier heights leading to a reduced effective area and yielded a Richardson’s constant closer to the theoretical value. Conductive atomic force microscopy studies were conducted, which concurred with the electrical data and confirmed the presence of inhomogeneous Schottky barrier heights.

Tech Support

Original Source

DOI: https://doi.org/10.1109/jeds.2020.2999269

References

2017 - Schottky-barrier inhomogeneities in WC/p-diamond at high temperature