MOSFET Threshold Voltage Characteristics

Technical Background of MOSFET Threshold Voltage

Metal-Oxide-Semiconductor Field-Effect Transistors (MOSFETs) are core switching devices in power electronic systems, widely used in power conversion, motor drives, and renewable energy applications. The threshold voltage (Vth) - defined as the minimum gate-source voltage required to form a conductive channel between the source and drain of the MOSFET - is a pivotal performance parameter. It directly determines the device's turn-on characteristics, static power consumption, and switching speed. For instance, in low-voltage power management ICs (PMICs), a precise Vth (deviation ≤ ±0.1V) is essential to ensure stable circuit operation and low standby power. This test focuses on the threshold voltage characteristics of MOSFETs under different operating conditions and process parameters, with all data derived from standardized laboratory tests (no brand-related information involved). The baseline test environment is 25℃ (room temperature) and 50%RH, and the test equipment includes a precision semiconductor parameter analyzer (voltage measurement accuracy: 0.001V), a temperature-controlled test chamber, and a capacitance-voltage (C-V) tester.

Test Methods for Threshold Voltage

This test adopts the "constant drain current method" (also known as the "linear region method") to measure Vth, which is widely recognized in the industry for its accuracy and repeatability. The specific process is as follows: first, select 3 groups of n-channel MOSFET samples with the same package (QFN-8) and rated parameters (drain-source voltage Vds = 30V, drain current Id = 10A), differing only in gate oxide layer thickness (10nm, 15nm, 20nm) and substrate doping concentration (1×10¹⁷ cm⁻³, 5×10¹⁷ cm⁻³, 1×10¹⁸ cm⁻³); then, fix the samples on a thermal stage to maintain uniform temperature, and connect the test probes to the gate (G), source (S), drain (D), and body (B) electrodes (body and source are short-circuited to eliminate body effect interference); next, set the drain-source voltage Vds to 0.1V (ensuring the device operates in the linear region) and gradually increase the gate-source voltage Vgs from 0V to 5V, recording the drain current Id in real time; finally, Vth is defined as the Vgs value when Id reaches 10μA (a standard test current for low-voltage MOSFETs).

To comprehensively characterize Vth stability, supplementary tests are conducted: 1. Temperature dependence: tests at -40℃, 0℃, 25℃, 75℃, 125℃; 2. Long-term reliability: 1000-hour continuous bias test (Vgs = 0.8×Vth, Vds = 15V, 125℃); 3. Process consistency: 50 samples per group, measuring Vth deviation. The test error is controlled within ±0.005V, and all operations comply with the JEDEC JESD22-A109 international standard.

Threshold Voltage Characteristics Data

1. Baseline (25℃) Vth data: For MOSFETs with 15nm gate oxide layer, the Vth of the 1×10¹⁷ cm⁻³ doping concentration sample is 0.85V, 5×10¹⁷ cm⁻³ is 1.12V, and 1×10¹⁸ cm⁻³ is 1.48V - showing a positive correlation with substrate doping concentration (higher doping increases Vth). For samples with 1×10¹⁷ cm⁻³ doping, the Vth of 10nm gate oxide is 0.68V, 15nm is 0.85V, and 20nm is 1.02V - indicating a positive correlation with gate oxide thickness (thicker oxide increases Vth).

2. Temperature dependence data: For the 15nm/5×10¹⁷ cm⁻³ sample, Vth is 1.35V at -40℃, 1.12V at 25℃, and 0.98V at 125℃, with a temperature coefficient α_Vth of -0.0028V/℃. The 10nm/1×10¹⁷ cm⁻³ sample has α_Vth = -0.0032V/℃ (Vth = 0.82V at -40℃, 0.68V at 25℃, 0.55V at 125℃), and the 20nm/1×10¹⁸ cm⁻³ sample has α_Vth = -0.0022V/℃ (Vth = 1.65V at -40℃, 1.02V at 25℃, 0.88V at 125℃). All samples exhibit negative temperature coefficients, meaning Vth decreases as temperature rises - a typical characteristic of MOSFETs.

3. Long-term stability data: After 1000 hours of 125℃ bias test, the Vth of the 15nm/5×10¹⁷ cm⁻³ sample increases by 0.03V (from 1.12V to 1.15V), the 10nm/1×10¹⁷ cm⁻³ sample increases by 0.05V (from 0.68V to 0.73V), and the 20nm/1×10¹⁸ cm⁻³ sample increases by 0.02V. The small Vth drift indicates good long-term reliability, with thicker oxide layers showing better stability.

4. Process consistency data: The Vth deviation of the 15nm/5×10¹⁷ cm⁻³ batch is ±0.04V, the 10nm/1×10¹⁷ cm⁻³ batch is ±0.06V, and the 20nm/1×10¹⁸ cm⁻³ batch is ±0.03V - thinner oxide layers lead to poorer process consistency due to stricter thickness control requirements.

Process Details Affecting Threshold Voltage

The threshold voltage of MOSFETs is closely related to gate oxide growth, substrate doping, and gate metal deposition processes. Key parameters and their effects are as follows: 1. Gate oxide growth: The oxide layer is grown by thermal oxidation at 950℃. A temperature deviation of ±20℃ leads to a 0.05V Vth change (higher temperature reduces oxide thickness, lowering Vth). The oxide layer thickness uniformity must be controlled within ±0.5nm; otherwise, Vth deviation increases by 50%. 2. Substrate doping: Ion implantation is used for doping. The implantation dose deviation of ±10% causes a 0.08V Vth change (higher dose increases Vth). The implantation depth is controlled at 50nm; depth deviation of ±5nm leads to a 0.03V Vth shift. 3. Gate metal work function: The gate metal is titanium nitride (TiN) with a work function of 4.6eV. A work function deviation of ±0.1eV causes a 0.06V Vth change (higher work function increases Vth). 4. Post-implantation annealing: Annealing at 1050℃ for 30 minutes in a nitrogen atmosphere. Insufficient annealing (＜1000℃) leads to incomplete activation of dopants, increasing Vth by 0.1V; excessive annealing (＞1100℃) causes dopant diffusion, reducing Vth by 0.07V.

Current Status of Commercial Application

From the industrial commercialization perspective, silicon-based MOSFETs with Vth of 0.8V-1.5V (gate oxide thickness 15nm-20nm, doping concentration 5×10¹⁷ cm⁻³-1×10¹⁸ cm⁻³) have achieved large-scale commercialization, accounting for about 70% of the low-voltage MOSFET market, and are widely used in consumer electronics and general industrial control. Low-threshold MOSFETs (Vth = 0.5V-0.8V, 10nm gate oxide) have also achieved large-scale commercialization, accounting for about 20% of the market, mainly used in battery management systems (BMS) and low-power PMICs. High-voltage MOSFETs (Vth = 2V-3V, 25nm-30nm gate oxide) are in small-batch mass production, used in electric vehicle on-board chargers. Wide-bandgap MOSFETs (gallium nitride GaN, Vth = 1.2V) are in the sample verification stage, featuring lower on-resistance and faster switching speed than silicon-based devices.

Existing Technical Pain Points

1. Low-temperature Vth increase: At ultra-low temperatures (＜-40℃, such as polar exploration equipment), Vth increases sharply. For example, the 10nm/1×10¹⁷ cm⁻³ sample's Vth reaches 0.82V at -40℃, increasing the turn-on voltage by 20% and increasing driving loss. This cannot meet the requirements of low-temperature electronic systems. 2. High-voltage application limitation: High-voltage MOSFETs require thicker gate oxide layers, leading to higher Vth (＞2V) and increased driving circuit complexity. Reducing Vth by reducing oxide thickness causes breakdown voltage degradation (a 5nm thickness reduction reduces breakdown voltage by 15%). 3. Process consistency for thin oxide layers: The Vth deviation of 10nm gate oxide MOSFETs reaches ±0.06V, which is twice that of 20nm oxide devices. This is due to the difficulty in controlling thin oxide layer thickness and doping uniformity, increasing production costs. 4. Radiation-induced Vth drift: In aerospace radiation environments, ionizing radiation causes Vth to increase by 0.2V-0.3V, leading to device turn-on failure. Current radiation-hardened processes increase costs by 3 times and reduce switching speed by 20%.