Wednesday, November 18, 2026
CS Fusion The compound semiconductor reference
GaN

Breakdown Field of Semiconductors: Si, SiC, GaN, Ga2O3 and Diamond Compared

The breakdown field of a semiconductor, also called the critical electric field, is the field at which avalanche multiplication runs away and the material stops blocking voltage. At 300 K it ranges from about 0.3 MV/cm in silicon to about 2–3 MV/cm in 4H-SiC, 3.3 MV/cm in GaN, about 8 MV/cm in β-Ga2O3 and about 10 MV/cm in diamond.

Key takeaways

  • Critical field rises steeply with bandgap. An empirical fit puts it at roughly Eg2 to Eg2.5.
  • Diamond has the highest critical field of the common candidates (about 10 MV/cm), followed by AlN and β-Ga2O3.
  • Drift-layer thickness for a given voltage falls as 1/Ec. Ideal specific on-resistance falls as 1/Ec3, which is why a 10 times higher field matters so much.
  • Critical field is not a single constant. It rises with doping and depends on crystal direction, so treat every published value as approximate.

Critical field by material

The table lists room-temperature values. Si, GaAs, SiC, GaN, AlN and diamond figures follow the Ioffe NSM archive and standard device texts. β-Ga2O3 is not in the Ioffe archive; its value is the widely cited theoretical estimate from the review literature, and measured device fields so far are lower.

Material (300 K)Bandgap (eV)Critical field (MV/cm)Field relative to SiNotes
Silicon1.12~0.31×Ioffe value; lower at the light doping of high-voltage drift layers
GaAs1.42~0.4~1.3×Used for RF, rarely for power
4H-SiC3.26~2–3~7–10×Depends on doping; Ioffe lists a 3–5 MV/cm range
GaN (wurtzite)3.39~3.3~11×Commonly cited theoretical value
β-Ga2O3~4.8~8 (estimated)~27×Theoretical estimate; not in Ioffe archive
Diamond5.47~10~33×Literature spans roughly 5–20 MV/cm
AlN (wurtzite)~6.2~12~40×Estimates vary widely; doping is very hard
Source: Ioffe Institute NSM archive; Baliga, Fundamentals of Power Semiconductor Devices; Kimoto and Cooper, Fundamentals of Silicon Carbide Technology; review literature on Ga2O3. Ratios use the central value of each range.

For bandgap, mobility and thermal data on the same materials, see our semiconductor properties table and the bandgap explainer, Decoding the Bandgap.

What is the breakdown field of GaN?

Wurtzite GaN has a critical field of about 3.3 MV/cm at 300 K, the figure most papers and the Ioffe archive use. That is about 11 times silicon. Real GaN HEMTs rarely reach it, because breakdown in a lateral device is usually set by field crowding at the gate edge, the buffer and the substrate rather than by the bulk GaN limit. Field plates and thick buffers exist to push devices closer to the material ceiling. Our GaN properties table lists the full data set.

Which semiconductor has the highest critical field?

Among materials with real device work behind them, diamond leads at about 10 MV/cm. AlN may be higher, with estimates around 12 MV/cm or more, but it is extremely hard to dope n-type or p-type, so AlN power devices remain research projects. β-Ga2O3 sits at about 8 MV/cm in theory and has the advantage of melt-grown substrates, but it has no practical p-type doping and poor thermal conductivity (roughly 10–27 W/m·K depending on crystal direction).

So the highest field and the most usable field are different questions. For production power devices today, 4H-SiC and GaN are the practical high-field materials. Our wide-bandgap semiconductor guide covers why.

How critical field scales with bandgap

A wider bandgap means a carrier must gain more energy from the field before it can knock a new electron-hole pair loose by impact ionization. Higher ionization thresholds mean a higher field before avalanche. Hudgins and co-authors fitted published data in their 2003 IEEE Transactions on Power Electronics assessment of wide-bandgap power devices and found the critical field rising roughly as the square of the bandgap, with a slightly steeper exponent (about 2.5) for direct-gap materials.

Treat it as a trend line, not a law. Phonon energy, band structure and crystal direction all move the real value. The fit does explain the table: tripling the bandgap from silicon to SiC or GaN buys roughly a ten times higher field.

Why breakdown field matters for power devices

A power transistor has a lightly doped drift region that holds off the voltage when the device is off. Its thickness and doping set the on-resistance when the device is on. Textbook relations for an ideal non-punch-through drift region (Baliga) are:

  • Drift thickness: WD = 2·BV / Ec. Higher field means a thinner layer.
  • Drift doping: ND = ε·Ec2 / (2q·BV). Higher field allows much heavier doping.
  • Specific on-resistance: Ron,sp = 4·BV2 / (ε·μ·Ec3). The cube makes the field the dominant term.

The denominator, ε·μ·Ec3, is the Baliga figure of merit. A tenfold rise in critical field cuts ideal on-resistance by a factor of about 1,000 before mobility and permittivity are counted.

The effect on drift-layer thickness: a 1,200 V example

Plugging the table values into WD = 2·BV / Ec for a 1,200 V blocking rating gives the ideal minimum drift thickness. Real devices add margin, and silicon’s field falls at the low doping such a layer needs, so the silicon figure is optimistic.

MaterialEc used (MV/cm)Ideal drift thickness at 1,200 V (µm)Relative to Si
Silicon0.3~801×
4H-SiC2.5~10~1/8
GaN3.3~7~1/11
β-Ga2O38~3~1/27
Diamond10~2.4~1/33
Calculated by CS Fusion from WD = 2·BV/Ec (Baliga). Values are idealised and rounded.

The SiC result, about 10 µm, matches the order of epitaxial layer thickness used for 1,200 V SiC MOSFETs. That thin layer is the reason a SiC MOSFET can replace a silicon IGBT at 1,200 V while still conducting as a simple resistor. Our SiC MOSFET vs Si IGBT comparison follows that through to losses.

Why published critical fields disagree

  • Doping dependence. Critical field rises slowly with doping. A figure measured on a heavily doped junction overstates what a lightly doped drift layer sees.
  • Anisotropy. Hexagonal crystals such as 4H-SiC and GaN have different impact-ionization rates along and across the c-axis.
  • Theory vs measurement. Many wide-bandgap values, including GaN’s 3.3 MV/cm and Ga2O3‘s 8 MV/cm, started as estimates. Measured fields in real devices are usually lower because of edge effects and defects.
  • Temperature. In most materials avalanche breakdown voltage rises with temperature because phonon scattering slows carriers. A 300 K value is a reference point, not an operating guarantee.

Pick a material by field and by practicality

  • Pick silicon when the voltage is modest, cost rules and a mature supply chain matters more than size.
  • Pick 4H-SiC when you need vertical devices from 650 V to several kV with good thermal conductivity (about 370 W/m·K) and both n- and p-type doping. See silicon carbide and its properties.
  • Pick GaN when you need fast lateral switches, mostly at 650 V and below.
  • Watch Ga2O3 and diamond if you work on research-stage ultra-high-voltage devices. Both still face doping, substrate or thermal hurdles.

Frequently asked questions

Is breakdown field the same as dielectric strength?

They describe the same idea, the maximum field a material withstands. In semiconductors the term usually means the avalanche critical field. In insulators such as SiO2, dielectric strength is higher, roughly 10 MV/cm, and the failure mechanism is different.

What is the breakdown field of 4H-SiC?

About 2–3 MV/cm at the doping levels used in power-device drift layers, roughly 7 to 10 times silicon. The Ioffe archive lists a broader 3–5 MV/cm range, which reflects heavier doping and different measurement methods.

Why does a higher critical field lower on-resistance so much?

Because ideal specific on-resistance scales as 1/Ec3. A higher field allows a thinner drift layer and much heavier doping at the same time, and both cut resistance.

Does breakdown field change with temperature?

Yes. For avalanche breakdown, the breakdown voltage generally rises with temperature because carriers lose more energy to lattice vibrations. Tunnelling breakdown in very heavily doped junctions shows the opposite trend.

Sources

  • Ioffe Institute, New Semiconductor Materials (NSM) archive: Si, GaAs, SiC, GaN, AlN and diamond pages. http://www.ioffe.ru/SVA/NSM/Semicond/
  • B. J. Baliga, Fundamentals of Power Semiconductor Devices, Springer (drift-region relations and Baliga figure of merit).
  • T. Kimoto and J. A. Cooper, Fundamentals of Silicon Carbide Technology, Wiley-IEEE Press (4H-SiC impact ionization and critical field).
  • J. L. Hudgins et al., “An assessment of wide bandgap semiconductors for power devices,” IEEE Transactions on Power Electronics, 2003 (critical field vs bandgap fit).
  • S. M. Sze and K. K. Ng, Physics of Semiconductor Devices, 3rd ed., Wiley (avalanche breakdown and its temperature dependence).
  • Review literature on β-Ga2O3 materials and devices (estimated critical field and thermal conductivity).
Malcare WordPress Security