Schottky Diode: Metal-Semiconductor Junction, Low VF

A Schottky diode replaces the p-n junction with a direct metal-semiconductor contact. That structural change eliminates minority-carrier storage, which is why Schottky diodes dominate high-frequency rectification, reverse-polarity protection, and switching-regulator freewheeling paths where a silicon p-n diode's forward drop and recovery time would kill efficiency or speed.

Metal-Semiconductor Junction Physics

When a metal is brought into contact with a lightly-doped n-type semiconductor whose work function is lower than the metal's, electrons diffuse from the semiconductor into the metal until the Fermi levels align. This leaves a depletion region on the semiconductor side only — the metal has essentially infinite carrier density, so no depletion forms there. The result is a rectifying (Schottky) barrier with built-in potential:

qΦB = q(ΦM − χS)     (ideal Schottky barrier height, n-type)

where ΦM is the metal work function and χS is the semiconductor electron affinity. In practice, interface states and surface effects pin ΦB below this ideal value, but the qualitative picture holds: a one-sided barrier controlled by majority carriers (electrons, for an n-type substrate) on both sides of the junction.

This is the key structural difference from a p-n junction: conduction is purely majority-carrier. No minority carriers (holes injected into n-type, or electrons into p-type) are stored on the "other side" of the junction during forward conduction, because there is no other side in the same sense — the metal has no minority-carrier population to speak of.

I-V Characteristic and Why VF Is Lower

The forward current still follows a thermionic-emission form of the diode equation:

I = I₀·[exp(V/(n·VT)) − 1]
I₀ = A·A*·T²·exp(−qΦB/kT)

where A* is the effective Richardson constant, A is junction area, VT = kT/q ≈ 25.85 mV at 300 K, and n is close to 1 (typically 1.02–1.1) because there's no minority-carrier recombination current component. The exponential shape is the same as a p-n diode, but I₀ is orders of magnitude larger because ΦB (typically 0.3–0.6 V for common Schottky metals on silicon) is much smaller than the ~1.1 V band-bending seen in a silicon p-n junction. A larger I₀ shifts the entire exponential curve to the left on the V-axis — for the same operating current, VF is lower.

Typical forward drops at moderate current density:

Note SiC Schottky diodes buck the "always lower VF" intuition — they trade forward drop for much higher breakdown voltage and negligible reverse recovery at high temperature, which is a separate design axis (see the SiC MOSFET article for the same wide-bandgap trade-off in switches).

No Minority-Carrier Storage → No Reverse-Recovery Tail

A p-n diode, when switched from forward conduction to reverse bias, must first sweep out the minority-carrier charge stored in the drift region before it can support reverse voltage — this is the reverse-recovery current spike (trr, typically tens of ns to a few µs for silicon rectifiers). A Schottky diode has no such stored charge on the semiconductor's majority-carrier side and no injected minority carriers on the metal side, so its "reverse recovery" is just the discharge of the junction's depletion capacitance — a few hundred picoseconds to a few nanoseconds, dominated by CJ and the external circuit, not by carrier lifetime.

This is the practical reason Schottky diodes are the default choice for:

• Switching-regulator freewheeling/catch diodes at frequencies above a few hundred kHz, where trr losses in a p-n diode would dominate switching loss.
• RF/microwave detectors and mixers, where the majority-carrier transport gives response into the GHz range.
• Reverse-polarity and ORing protection, where the low VF directly reduces power dissipation and voltage drop budget.

Reverse Characteristics and Leakage

The trade-off for low VF is higher reverse leakage current and typically lower reverse breakdown voltage than an equivalent p-n diode. Reverse leakage in a Schottky diode is dominated by thermionic emission back over the (lower) barrier, and it increases strongly with temperature — often the limiting factor in high-temperature applications, since I₀ ∝ T²·exp(−qΦB/kT) grows quickly as T rises. This is why silicon Schottky diodes are usually rated to lower junction temperatures than comparable p-n rectifiers, and why SiC Schottky diodes (larger ΦB, wider bandgap) are preferred where both high temperature and low leakage are required simultaneously.

Worked Example: Freewheeling Diode Power Loss

Compare conduction loss for a silicon p-n rectifier (VF = 0.65 V) vs. a Schottky diode (VF = 0.35 V) in a buck converter freewheeling path carrying an average current of 3 A during the diode's conduction interval, with duty cycle where the diode conducts D' = 0.4 of the switching period.

P_pn      = VF·I·D' = 0.65 V × 3 A × 0.4 = 0.78 W
P_Schottky = 0.35 V × 3 A × 0.4 = 0.42 W

Saved conduction loss: 0.78 − 0.42 = 0.36 W. At a switching frequency of 500 kHz, add the p-n diode's reverse-recovery loss, approximated as:

P_rr ≈ ½·Qrr·VR·fSW

For a typical silicon rectifier Qrr ≈ 50 nC at VR = 12 V:

P_rr ≈ 0.5 × 50 nC × 12 V × 500,000 Hz = 0.15 W

Total p-n diode loss ≈ 0.78 + 0.15 = 0.93 W vs. Schottky's 0.42 W (its recovery loss is negligible, sub-milliwatt, since Qrr is essentially just CJ·VR). That's roughly a 2.2× total loss reduction — consistent with why Schottky rectifiers are near-universal in converters above a few hundred kHz.

Check: units confirm — VF·I gives watts directly (V×A=W), and ½·Qrr·VR·fSW has units of (C·V)·(1/s) = J/s = W. Order of magnitude (sub-watt losses for a few-amp, tens-of-volt converter) matches typical small buck-converter thermal budgets.

Design Implications

• Thermal runaway risk: because I₀ rises steeply with temperature, a Schottky diode carrying too much current in a poorly heat-sunk design can push VF down, current up, temperature up — a positive feedback loop. Derate current well below datasheet peak at elevated ambient.
• Paralleling: like any diode with a negative VF-temperature coefficient region of forward operation, current sharing between paralleled Schottky diodes needs matched parts or forced balancing — a hotter device tends to draw more current.
• Voltage rating trade-off: low-barrier silicon Schottky diodes rarely exceed ~60–100 V reverse rating economically; for higher voltage with Schottky-like switching speed, SiC Schottky diodes extend usefully to 600 V–1700 V class parts.
• Capacitance matters at RF: junction capacitance CJ (a function of area and reverse bias) sets the practical high-frequency ceiling for detector/mixer applications — smaller-area, higher-ΦB Schottky diodes trade some VF for lower CJ and higher fmax.

Key Takeaways

• A Schottky diode's rectifying action comes from a metal-semiconductor barrier (ΦB), not a p-n junction — conduction is purely majority-carrier.
• Lower barrier height (0.3–0.6 V typical, vs. ~1.1 eV silicon bandgap) gives a larger saturation current I₀, which is why forward voltage is lower for the same current.
• No minority-carrier storage means reverse recovery is dominated by junction capacitance discharge (sub-ns to a few ns), not carrier-lifetime sweep-out — the core reason Schottky diodes win at high switching frequency.
• The trade-offs are higher reverse leakage, generally lower breakdown voltage, and strong temperature sensitivity of leakage/I₀ compared to p-n rectifiers.
• SiC Schottky diodes shift the whole trade-off curve, combining wide-bandgap high voltage/temperature capability with the fast-switching benefits of the metal-semiconductor junction, at the cost of a higher VF than low-barrier silicon parts.