JPH0665217B2 - Transistor - Google Patents

Description

The present invention relates to a semiconductor device that operates by injecting and running minority carriers in a semiconductor crystal, specifically, a novel structure in a bipolar transistor or the like.

At present, bipolar transistors for high frequencies are exclusively Si np
n-transistors are in practical use. Here, a large factor limiting the high frequency characteristics is a large base spreading resistance, and it is desired to reduce this and improve the high frequency characteristics. In addition, since III-V compound semiconductors such as GaAs and InAs have a high electron mobility, it can be applied to a bipolar transistor to improve the speed of operation compared to a Si transistor. However, in the III-V compound semiconductor, since the mobility of holes is small, the base resistance is further increased in the structure similar to that of the conventional Si transistor, and even if the III-V compound semiconductor is used, the high frequency characteristics are high. It is considered that there is a high possibility that it will not improve. The means for reducing the base resistance is to increase the carrier concentration of the base or increase the base width. However, if the base concentration is increased, it is necessary to increase the carrier density of the emitter, that is, the impurity concentration, in order not to reduce the emitter injection efficiency, and the base concentration is unnecessarily increased from the viewpoint of the reverse breakdown voltage of the base-emitter. It is not possible. Increasing the base width causes deterioration of high frequency characteristics. Therefore, in the bipolar transistor having the conventional structure, the base spreading resistance cannot be effectively reduced without impairing the high frequency characteristics.

INDUSTRIAL APPLICABILITY As described above, the present invention provides a bipolar transistor having a novel structure, which brings about a large reduction in base spreading resistance, which is not possible with conventional devices, and improves the high frequency characteristics more than the effect of reducing the base spreading resistance. It is provided.

The bipolar transistor of the present invention has an energy band phase diagram in its thermal equilibrium state shown in FIG.
The base 12 includes a p-type layer 12A of a first semiconductor and a p-type layer 12B of a second semiconductor, which are alternately repeated. The energy gap of the first semiconductor is equal to the energy gap of the second semiconductor. And the sum of the electron affinity and the energy gap of the first semiconductor is substantially equal to the sum of the electron affinity and the energy gap of the second semiconductor, and the band gap of the semiconductor of the emitter 11 is the first. 2 or more of the band gap of the semiconductor, and
The thickness of each layer of the first semiconductor is sufficiently thin so that electrons can pass through each layer by a tunnel effect. The first semiconductor layer 12A is the second semiconductor layer 1
It forms a potential barrier for 2B electrons. In FIG. 1, E _C , E _F, and E _V represent energy levels in the conduction band, Fermi level, and valence band, respectively, and 11 is an emitter and 13 is a collector. In the bipolar transistor of the present invention, the electrons injected into the base are the first
The semiconductor p-type layer 12A rapidly moves due to the tunnel effect, and the second semiconductor p-type layer 12 travels by diffusion like a normal bipolar transistor. Since the transit time due to the tunnel effect is extremely short and can be ignored, the base resistance can be reduced by the amount of the p-type layer of the first semiconductor without increasing the transit time in the base. Not only that, but the thickness of the unit layer of the p-type layer 12B of the second semiconductor is made thin and the number of repetitions of each layer is appropriately increased, so that the base spreading resistance is greatly reduced as compared with the conventional element. In addition, the base traveling time is shortened, and the high frequency characteristics can be greatly improved.

Further, in the present invention, when the minority carriers traveling in the base are holes, as shown in FIG. 2, the n-type layer 22A of the first semiconductor and the n-type layer 22B of the second semiconductor alternate. An energy gap of the first semiconductor is larger than an energy gap of the second semiconductor, and an electron affinity of the first semiconductor is larger than that of the second semiconductor. It is almost equal to the affinity, the band gap of the semiconductor of the emitter 21 is greater than or equal to the band gap of the second semiconductor, and the thickness of each layer of the first semiconductor is such that holes pass through each layer by tunnel effect. It is characterized by being made sufficiently thin so that it can be made. The first semiconductor layer 22A forms a potential barrier against holes in the second semiconductor layer 22B. Where 21 is the emitter and 23
Is a collector.

Next, specific examples of the present invention will be described, and the operation and effects of the transistor of the present invention will be described in detail. In the first example, the P ⁺ layer 11A of the first semiconductor unit that constitutes the base is a Ga _0.7 Al _0.3 As layer having an effective acceptor density of 1 × 10 ¹⁸ cm ^{-3 and a} thickness of 50 Å,
The p ⁺ layer 11B of the second semiconductor unit has an effective acceptor density of 1
× 10 ¹⁸ cm ^-3 It is a 50 Å thick GaAs layer. The emitter 11 is composed of n ⁺ -GaAs having an effective donor density of 2 × 10 ¹⁸ cm ⁻³ , and the collector is composed of n ⁻ -GaAs having an effective donor density of 1 × 10 ¹⁵ cm ⁻³ . FIG. 3 shows an example of the structure of this transistor. 31 is an emitter electrode, 32 is a base electrode, and the main body of the transistor is formed on an n ⁺ -GaAs substrate 33 which also serves as a collector region. 34 is a collector electrode. Multilayer crystals of 11, 12 and 13 can be grown, for example, by molecular beam epitaxy (MBE). The energy band state diagram of this transistor in the thermal equilibrium state is shown in FIG. 4 (a), and the state diagram when the transistor is operated by biasing each electrode is shown in FIG. 4 (b). Here, black circles represent electrons and white circles represent holes. Ga _0.7 Al _0.3 As of the first semiconductor of the base is Ga _0.7 Al _0.3 As of the second semiconductor.
The energy gap is about 0.4 eV larger than As, and the electron affinity is about 0.4 eV smaller. Therefore, potential wells are repeated in the conduction band in the base. In the thermal equilibrium state, the bottom of the potential well of the conduction band of the base, that is, P ⁺ -GaAs, is caused by the contact-potential difference (Built-in Potential) of the pn junction of the emitter-base junction.
The conduction band of 12B has a higher energy than the conductor of the emitter 11, and electrons are not injected into the base. Next, Fig. 3 (b)
Forward biasing the base reduces the energy difference between the conduction band of the p ⁺ -GaAs and the conduction band of the emitter, and
Since the p ⁺ -GaAlAs layer 12A is sufficiently thin, electrons having a larger energy than the bottom of the potential well of the conduction band of the base among the electrons in the emitter are p ⁺ due to the tunnel effect.
Electrons are injected into the base through the barrier of the GaAlAs layer. The injected electrons successively move to the collector side in the base by diffusion in the p ⁺ -GaAlAs layer 12A and in the p ⁺ -GaAs layer 12B by diffusion, and when they reach the base-collector junction, they become collector by the electric field of the junction. Collected and collector current flows.
The transistor operation, that is, the modulation of the collector current, is performed by changing the injection amount of electrons from the emitter to the base due to the minute potential displacement between the base and the emitter.

As is clear from the above operation principle, in the transistor of the present invention, the time for which an electron passes through the p ⁺ layer of the first semiconductor is short and it hardly contributes to the increase of the transit time of the electron in the base. The base resistance is reduced by the p ⁺ layer of the first semiconductor without increasing the transit time as compared with the transistor of FIG. In particular, the thinner the p ⁺ layer of the second semiconductor and the larger the number of repetitions, the greater the effect. Here, in order to realize such a conduction band potential well in the base, the energy gap of the first semiconductor is set to the second
It is necessary that the sum of the electron affinity and the energy gap of the first semiconductor be the first value in order that the potential well has a flat bottom and the electrons travel quickly. It is desirable that the value is substantially equal to the value of the semiconductor of No. 2, that is, the difference in the energy gap between the two is substantially the difference in electron affinity. Considering this point, in the above embodiment, the first semiconductor is Ga _0.7 Al _0.3 As, and the second semiconductor is Ga _0.7 Al _0.3 As.
GaAs is used as the semiconductor.

Next, as a second embodiment, a structure for further enhancing the effect of the transistor of the present invention will be described. Fifth
As shown in the energy band state diagram of the thermal equilibrium state in the figure, in this example, as the first semiconductor of the base region, a material having a sum of electron affinity and energy gap smaller than the value of the second semiconductor is used, and The doping level of the p ⁺ layer of the first semiconductor is made higher than the doping level of the p ⁺ layer of the second semiconductor according to Here, the semiconductor of the emitter is the same as the second semiconductor, so p ^{+ of} the first semiconductor is
Even if the doping level of the layer is increased, the emitter injection efficiency is not reduced, so that the base resistance can be further reduced. In the case of this example, for example, in the first semiconductor of the first embodiment,
Instead of GaAlAs, GaAlAsSb with a little Sb added may be used. Further, as shown in FIG. 6, as the emitter, by using not only the first semiconductor of the base but also the semiconductor of which the sum of energy gap and electron affinity is larger than that of the second semiconductor, the injection of holes into the emitter is extremely small. can do. Therefore, the doping level of the impurity of the emitter can be lowered and the doping level of the p ⁺ layer of the second semiconductor of the base can be increased while keeping the emitter injection efficiency at about 1, so that the emitter junction capacitance is reduced and the base resistance is further reduced. It can be made small and the high frequency characteristics are greatly improved. Specifically, ZnSe or the like may be used as the emitter instead of GaAs in the above embodiment. Further, in order to further shorten the transit time of electrons in the p ⁺ layer of the second semiconductor of the base and further reduce the transit time of the base, as shown in FIG. 7, doping of each p ⁺ layer in the base is performed. The level is gradually reduced from the emitter side to the collector side. Then, an electric field is generated in the base in the direction from the emitter to the collector, and the injected electrons are accelerated by this electric field to further reduce the base transit time.

Next, in the present invention, a case where the minority carrier at the base is a hole, that is, an example of a pnp type transistor will be described. N ⁺ −GaA as the n ⁺ layer 22A of the first semiconductor of the base
s, n ⁺ -Ge is used as the n ⁺ layer 22B of the second semiconductor. Here, the energy gap of GaAs is about 0.6 eV larger than that of Ge,
Further, since the electron affinities are the same, a potential well for holes in the valence band at the base is formed as shown in FIG. The emitter is p ⁺ −Ge and the collector is p ⁻ −Ge.
Can be used. Where the base of the first semiconductor
If GaInAs with a small amount of In added is used instead of GaAs, the electron affinity becomes larger than that of Ge of the second semiconductor, so that the doping level of the n ⁺ layer of the first semiconductor can be increased. Also, it has a small electron affinity instead of n ⁺ −Ge of the emitter.
With GaAlAs, the injection of electrons from the base to the emitter can be made extremely small. Also, if the doping level of each n ⁺ layer of the base is gradually reduced from the emitter side to the collector side, it is effective to shorten the base transit time.
The same as in the case of n transistors.

Note that, in the above, the multilayer structure in which the base is the first semiconductor layer and the second semiconductor layer has been described, but from the operating principle of the transistor of the present invention, the base is the first semiconductor layer and the second semiconductor layer.
The present invention also includes the case where each of the semiconductor layers is formed by one layer.

[Brief description of drawings]

1 to 7 are views for explaining a bipolar transistor according to the present invention. Here, FIG. 3 shows an example of the specific structure, and the other shows an energy band phase diagram. Note that FIG. 4B shows the energy band in the thermal equilibrium state when biased so that the transistor operates. In the figure, E _F : Fermi level; E _C : conduction band; E _V : valence band;
21: emitter; 12,22: base; 13,23: collector; 12A: first semiconductor p ⁺ layer; 12B: second semiconductor p ⁺ layer; 22A: first semiconductor n ⁺ layer; 22B : N ⁺ layer of the second semiconductor; 31: emitter electrode; 3
2: base electrode; 33: n ⁺ substrate; 34: collector; black circle: electron; white circle: hole

Claims (6)

[Claims] 1. A p-type layer of a first semiconductor and a p-type layer of a second semiconductor.
A base formed by alternately repeating mold layers, the energy gap of the first semiconductor is larger than that of the second semiconductor, and the first semiconductor has an energy gap larger than that of the second semiconductor;
The sum of the electron affinity and the energy gap of the semiconductor is substantially equal to the sum of the electron affinity and the energy gap of the second semiconductor, and the first semiconductor layer has a potential with respect to the electrons of the second semiconductor layer. It forms a barrier, the bandgap of the semiconductor of the emitter is greater than or equal to the bandgap of the second semiconductor, and the thickness of each layer of the first semiconductor is sufficient to allow electrons to pass through each layer by the tunnel effect. A transistor characterized by being made thinner. 2. The transistor according to claim 1, wherein the sum of the electron affinity and the energy gap of the semiconductor of the emitter is larger than the sum of the electron affinity and the energy gap of the first semiconductor and the second semiconductor of the base. . 3. A transistor according to claim 1 or 2, wherein the doping level of each p-type layer of the base is gradually reduced from the emitter side to the collector side. 4. An n-type layer of a first semiconductor and an n-type layer of a second semiconductor.
A base formed by alternately repeating mold layers, the energy gap of the first semiconductor is larger than that of the second semiconductor, and the first semiconductor has an energy gap larger than that of the second semiconductor;
Of the semiconductor of the second semiconductor layer is substantially equal to the electron affinity of the second semiconductor, and the first semiconductor layer forms a potential barrier against holes of the second semiconductor layer. The bandgap is equal to or larger than the bandgap of the second semiconductor, and the thickness of each layer of the first semiconductor is made sufficiently thin so that holes can pass through each layer by a tunnel effect. . 5. The transistor according to claim 4, wherein the electron affinity of the semiconductor of the emitter is smaller than the electron affinity of the first semiconductor and the second semiconductor of the base. 6. The transistor according to claim 4, wherein the doping level of each n-type layer of the base gradually decreases from the emitter side to the collector side.