Tunnel diodes exploit a strange quantum phenomenon called resonant tunneling to provide a negative resistance forward-bias characteristics. When a small forward-bias voltage is applied across a tunnel diode, it begins to conduct current. (Figure below(b)) As the voltage is increased, the current increases and reaches a peak value called the peak current (IP). If the voltage is increased a little more, the current actually begins to decrease until it reaches a low point called the valley current (IV). If the voltage is increased further yet, the current begins to increase again, this time without decreasing into another “valley.” The schematic symbol for the tunnel diode shown in Figure below(a).
Tunnel diode (a) Schematic symbol. (b) Current vs voltage plot (c) Oscillator.
The forward voltages necessary to drive a tunnel diode to its peak and valley currents are known as peak voltage (VP) and valley voltage (VV), respectively. The region on the graph where current is decreasing while applied voltage is increasing (between VP and VV on the horizontal scale) is known as the region of negative resistance.
Tunnel diodes, also known as Esaki diodes in honor of their Japanese inventor Leo Esaki, are able to transition between peak and valley current levels very quickly, “switching” between high and low states of conduction much faster than even Schottky diodes. Tunnel diode characteristics are also relatively unaffected by changes in temperature.
Reverse breakdown voltage versus doping level. After Sze [SGG]
Tunnel diodes are heavily doped in both the P and N regions, 1000 times the level in a rectifier. This can be seen in Figure above. Standard diodes are to the far left, zener diodes near to the left, and tunnel diodes to the right of the dashed line. The heavy doping produces an unusually thin depletion region. This produces an unusually low reverse breakdown voltage with high leakage. The thin depletion region causes high capacitance. To overcome this, the tunnel diode junction area must be tiny. The forward diode characteristic consists of two regions: a normal forward diode characteristic with current rising exponentially beyond VF, 0.3 V for Ge, 0.7 V for Si. Between 0 V and VF is an additional “negative resistance” characteristic peak. This is due to quantum mechanical tunneling involving the dual particle-wave nature of electrons. The depletion region is thin enough compared with the equivalent wavelength of the electron that they can tunnel through. They do not have to overcome the normal forward diode voltage VF. The energy level of the conduction band of the N-type material overlaps the level of the valence band in the P-type region. With increasing voltage, tunneling begins; the levels overlap; current increases, up to a point. As current increases further, the energy levels overlap less; current decreases with increasing voltage. This is the “negative resistance” portion of the curve.
Tunnel diodes are not good rectifiers, as they have relatively high “leakage” current when reverse-biased. Consequently, they find application only in special circuits where their unique tunnel effect has value. To exploit the tunnel effect, these diodes are maintained at a bias voltage somewhere between the peak and valley voltage levels, always in a forward-biased polarity (anode positive, and cathode negative).
Perhaps the most common application of a tunnel diode is in simple high-frequency oscillator circuits as in Figure above(c), where it allows a DC voltage source to contribute power to an LC “tank” circuit, the diode conducting when the voltage across it reaches the peak (tunnel) level and effectively insulating at all other voltages. The resistors bias the tunnel diode at a few tenths of a volt centered on the negative resistance portion of the characteristic curve. The L-C resonant circuit may be a section of waveguide for microwave operation. Oscillation to 5 GHz is possible.
At one time the tunnel diode was the only solid-state microwave amplifier available. Tunnel diodes were popular starting in the 1960’s. They were longer lived than traveling wave tube amplifiers, an important consideration in satellite transmitters. Tunnel diodes are also resistant to radiation because of the heavy doping. Today various transistors operate at microwave frequencies. Even small signal tunnel diodes are expensive and difficult to find today. There is one remaining manufacturer of germanium tunnel diodes, and none for silicon devices. They are sometimes used in military equipment because they are insensitive to radiation and large temperature changes.
There has been some research involving possible integration of silicon tunnel diodes into CMOS integrated circuits. They are thought to be capable of switching at 100 GHz in digital circuits. The sole manufacturer of germanium devices produces them one at a time. A batch process for silicon tunnel diodes must be developed, then integrated with conventional CMOS processes. [SZL]
The Esaki tunnel diode should not be confused with the resonant tunneling diode CH 2, of more complex construction from compound semiconductors. The RTD is a more recent development capable of higher speed.
Article Extracted from Tony R. Kuphaldt Lessons In Electric Circuits — Volume III Chapter 3 under the terms and conditions of the CC BY License.