The drift layer

Same voltage, a tenth of the depth: why silicon carbide switches the power that silicon cannot.

1,200 V
Mode

A MOSFET that blocks a kilovolt has to hold that voltage somewhere. It holds it in the drift layer, a slab of lightly doped crystal under the gate, and the slab's thickness and doping are set by the one number the crystal cannot be argued out of: the field at which it breaks down.

  1. 01Silicon carbide's bandgap is three times wider,
  2. 02so it takes about ten times the field to break it down;
  3. 03the same voltage then fits in a tenth of the depth, doped a hundred times harder,
  4. 04and the drift layer's resistance falls about a thousandfold.

Drift layer only. A whole SiC MOSFET adds its channel, neck and substrate: devices of 1030 V at 0.94 mΩ·cm² and 1800 V at 2.0 mΩ·cm² have been reported, and at 1.2 kV the channel alone is about half.

Idealised: one-dimensional non-punch-through drift region at room temperature; silicon after Baliga, 4H-SiC critical fields and resistance after Kimoto. Illustrative, not a device model.

Sources

  1. B. J. Baliga, Fundamentals of Power Semiconductor Devices (Springer)Silicon ideal drift region: N = 2×10¹⁸ · V^(−4/3) cm⁻³, W = 2.58×10⁻⁶ · V^(7/6) cm, Ec = 4010 · N^(1/8) V/cm, R = 5.93×10⁻⁹ · V^2.5 Ω·cm²
  2. T. Kimoto, “Updated trade-off relationship between specific on-resistance and breakdown voltage in 4H-SiC{0001} unipolar devices”, Jpn. J. Appl. Phys. (2019)Critical field 2.00, 2.50, 3.34, 5.01 MV/cm at 10¹⁵–10¹⁸ cm⁻³; the SiC limit 0.20, 2.5, 39 mΩ·cm² at 1, 3, 10 kV
  3. T. Kimoto and H. Watanabe, “Defect engineering in SiC technology for high-voltage power devices”, Appl. Phys. Express (2020)Reported SiC MOSFETs of 1030 V at 0.94 mΩ·cm² and 1800 V at 2.0 mΩ·cm²; the channel is about 48% of a 1.2 kV device
  4. T. Kimoto and J. A. Cooper, Fundamentals of Silicon Carbide Technology (Wiley, 2014)Bandgap at room temperature: Si 1.12 eV, 4H-SiC 3.26 eV
四宫格