JPS58114455A - semiconductor equipment - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a semiconductor device that operates by injecting minority carriers into a semiconductor crystal and causing them to travel.

Currently, as bipolar transistors used as high-frequency semiconductor devices, Si npn (npn) transistors have been put into practical use. The key elements to improve high frequency characteristics are:
The goal is to lower the base spreading resistance, increase the emitter injection efficiency, and increase the current amplification factor.On the other hand, GaAs s
■-v compound semiconductors such as InAa have high electron mobility, so they are applied to piborad 2 transistors, and St)? Improvements in speed compared to transistors are being considered.

However, since the m-v compound semiconductor has a low hole mobility, a structure similar to that of a conventional St) transistor will have a large pace resistance. Also, ■-■ Compound semiconductors with high electron mobility are of the direct transition type, so when electrons in the minority + q ria are injected into the base, there is a high probability of recombination and the emitter injection efficiency decreases. . Therefore, in order to lower the base resistance, increasing the base impurity concentration and increasing the base thickness will reduce the mobility of electrons in the base region and increase the rate of recombination. Not only does the emitter injection efficiency decrease and the current amplification factor exceed the effect of increasing the impurity concentration, but also the base transit time is not shortened even though a -v compound semiconductor is used. Injecting minority carriers into bipolar transistors using compound semiconductors as described above,
An object of the present invention is to provide a semiconductor device having a novel structure with excellent high speed and which can eliminate the disadvantages of a running semiconductor device and exhibit the advantage of high electron mobility. The semiconductor device of the present invention is shown in FIG. As shown in the energy band diagram 11 in the thermal equilibrium state of sl, the region 12 where electrons, which are minority carriers, are injected and travels is pm
Twelve high-concentration (p) regions and high N& or p-low 11 degree (p-) regions 12B are alternately repeated, and the electron affinity of the semiconductor in the regions 12 is high #lIJ[
Or smaller than the value of the semiconductor of p- region 12B,
Furthermore, the thickness of the r region 12A is sufficiently thin so that electrons can pass through the region by tunneling effect.

Here, k, Ey, and Ev represent the energy levels of the conduction band, Fermi level, and valence band, respectively, region 11 is the electron supply electrode, region 13 is the electron collection electrode, and in the case of a bipolar transistor, , are the emitter and collector, respectively.In the semiconductor device of the present invention, electrons, which are minority carriers, move quickly through tunnels in regions where the probability of recombination is high, and those that travel by diffusion have electron mobility. Since it is in a high purity region with a large value, the transit time is extremely short and there is little recombination of minority carriers. The concentration and repetition rate of the drop region is increased, resulting in a large reduction in the resistance of region 12.

According to the present invention, when the minority carrier is a hole, the second
As shown in the figure, holes, which are minority carriers, are injected,
What is the driving area 22? area and high Nf or n
-area=1 are alternately repeated, and? the sum of the electron affinity and energy gap of the semiconductor in the region is greater than the value of the semiconductor in the high purity or n-region;
Furthermore, the thickness of the male region 22A is required to be sufficiently thin so that holes can pass through the wrinkled region due to the tunnel effect. Here, 21 is the hole supply electrode, and the hole collector is i.
It is an electrode.

Next, in the case of a bipolar 2 transistor, the present invention will be explained in detail, and the operation and effects of the semiconductor device of the present invention will be explained in detail.
(b) shows the case in which the transistor is biased to operate. Here, black circles represent electrons, and white circles represent holes. The base 32 is, for example, p"-GaA with an effective acceptor density of 1x10"ox-"
sx8bl -x32m and effective acceptor density I
10 cm 17) p-Gay In'1-yAs
Consists of alternating layers of 32B. Here, the thickness of each unit layer of 32 people and 32B is, for example, 50X. Further, the emitter 31 has, for example, an effective donor density [I X 1G1
7cs+-”n-Gayll-y As s:2
The V-lid 33 has an effective donor density I
-Gay In1-yAs. Here GaAs
The electron affinity of x 5bt-x is Gay Inl -y
As (F) is smaller than the scissor value, and under the condition of lattice constant matching, if the range of 0.4≦X<1 is selected, p”-
The work function of G&X 8b
Since it is almost the same as the wrinkle value of yAi, as shown in Figure 3,
A periodic potential well is formed in the conduction band in the base region. A specific example of I and y is, for example, using InP as the substrate, InPK: lattice matching xmo, 5.
7-0.47.

In the thermal equilibrium state of 11311(a), the emitter (3
1) -Be x (32) Junction diffusion potential difference (Build
t −1nPot@ntiml ), base 3
The bottom of the potential shell of the conduction band of 2, that is, P--G
The conduction band of almAm 32B has higher energy than the conduction band of the emitter 31, and no electrons are injected into the base. Next, if the bias is applied as shown in FIG. 3(b), the P
- The energy difference between the conduction band of GaInAlB 52B and the conduction band of the emitter 31 decreases, and the F'-GaAsSb layer 32
Since A is sufficiently thin, among the electrons in the emitter, electrons with higher energy than the bottom of the potential well in the conduction band of the base transit through the barrier of the P+ layer 32 due to the tunnel effect, and the electrons are injected into the base. . The injected electrons move through the base 32 to the collector 3341 one after another, using the tunnel effect in P''JI32A and diffusion in the P layer 32B, and after reaching the base-collector junction, the process is similar to that of a normal bibo-2 transistor. The operation of the 0 transistor, in which the collector current flows through the collector 33 by the electric field of the junction, is performed by changing the amount of electrons injected from the emitter to the base by a minute displacement of the base-emitter potential. As can be understood from the above operation am The region that travels by diffusion is a high purity region with high electron mobility.Therefore, the base transit time is short, and if the thickness of the unit layer of the high purity or r region 32B is made thin, the transit time will be reduced. Here, the electron transit time is long in regions with high purity or few holes in the y region, so the recombination rate of electrons injected into the base can be extremely small.On the other hand, Regarding the base spreading resistance, by increasing the effective accessor density, that is, the hole density, of the r region 32A of the base 32 and increasing the number of repetitions of the unit layer, the base transit time and recombination rate of electrons, which are carriers, can be improved. It can be significantly reduced without resulting in a significant increase.

In order to make the effects of the present invention noticeable, it is necessary to form a potential well as steep as possible in the conduction band of the base region, but it is also desirable that the bottom of the potential well be flat so that electrons can easily diffuse. For this reason, it is desirable that the electron affinity of the semiconductor in the r region 32M is smaller than that of the P- region 32B, and that the work functions of both regions, that is, the energy difference between the vacuum level and the 7 Hermi level, are approximately the same. Considering this point, in this example, 32
P+-GaAszSbl-1 in A and PG in 32B
ay In1-, As is used. In this system, the electron affinity of GaAazSbl-z is much smaller than that of GayInl-ymS over a large range of compositions;
In addition, the sum of the electron affinity and the energy gap of the former is also smaller than the corresponding value of the latter, and furthermore, these energy differences vary greatly depending on the composition, so that depending on the doping level of the r region 32B, there is a steep slope as shown in No. 31ii. There is a large degree of freedom in design to realize a potential well.

Furthermore, if a semiconductor with high hole mobility is used for the r region 32A and a semiconductor with high electron mobility is used as the semiconductor for the P- region 32B, the characteristics will be better. In the above-mentioned embodiment, this point is also taken into account, and GaI is used as the semiconductor for the f-region 32B.An embodiment of the biborder transistor according to the present invention can be manufactured with the structure shown in FIG. 4, for example. Here, 4L and 43 are an emitter electrode, a base electrode, and a collector electrode, respectively, and 44 is a male substrate forming a part of the collector region, for example, InP can be used. The multilayer crystal structure of 31, 32, 33 can also be formed by molecular beam epitaxial growth (MBE).

In the above example, the sum of the electron affinity and the energy gap of GajnAs used for the emitter, that is, the energy difference between the vacuum level and the upper end of the valence band, is larger than that of the semiconductor QaAsSb in the t region of the base, so the positive The barrier height for holes is larger than the barrier height for injected electrons, that is, the energy difference between the bottom of the conduction band between the emitter 31 and the P-GaI layer 32B in the base, so that electrons are injected from the base to the emitter. In other words, in a normal S-junction transistor, the base convergence cannot be made larger than the emitter concentration in order to keep the emitter injection efficiency high. In the example, we will introduce P”-GaAs while keeping the emitter injection efficiency high.
The doping level of Sb is changed to n-GaInA emitter.
The doping level of s can be greater than
Base resistance can be reduced. Furthermore, if InP or the like, which has a larger sum of electron affinity and energy gap than GaInAs, is used as the emitter semiconductor, the barrier to holes will be roughly larger (Fig. 5), and the ensuta injection efficiency will be increased, and the tgA region of the base will be Doping levels can be extremely high. In addition, if the doping level of each P+ slope of the base is gradually decreased from the emitter side to the collector side, high purity or impurity density 1 vas in the Pl region can be eliminated, and as shown in No. 611#c, from the emitter side to the prevert side in the base. An electric field is generated at , and the injected electrons are accelerated by this electric field, further reducing the base transit time.

Next, if the minority carrier is a hole, that is, npn
A specific example in the case of an Ml bipolar transistor will be described. - As an example, as the n area 22A of page 2,
--InP, high purity or In as n-region 22B
It is possible to use GjLio and s8bg, which have a smaller sum of electron, **force and energy gap than P.

At this time, I of the n+ region semiconductor of the base is used as an emitter.
Semiconductors with lower electron affinity than np, e.g. GaAg
By using g5bas, the injection of electrons from the base to the emitter can be reduced, and the emitter efficiency can be increased by increasing the n of the base.
The doping level in the ten regions can be increased. Also each in the base? As in the case of npn (npn) transistors, if the doping level of the region is gradually reduced from the emitter side to the collector side, the base transit time can be reduced.

The present invention has been described above with respect to the case where the base is repeatedly formed of a highly doped region and a high purity or low impurity density region. However, considering the operating principle of the present invention, the present invention can also be applied to a case where each of the above regions is a single layer. It is clear that it falls within the scope of

[Brief explanation of drawings]

Figures 1 to 6 are diagrams for explaining the semiconductor device of the present invention. jlE1 diagram shows a case where the minority carrier is an electron, FIG. 2 shows a case where the minority carrier is a hole, and FIG. 3 shows a specific example in which the present invention is applied to an npn Ml bipolar transistor. (b) shows the case where a bias is applied. Figure 4 shows a specific structural example of the IIK3 diagram, and Figures 5 and 6 show examples that make the present invention more effective. are the conduction band,
In the 0 diagram showing the valence band and Fermi level, 11.21; Supply electrode: 12A22: Area where minority carriers run: 1 & 23: Collection electrode; 31: Emitter; 32: Base; 33: Collector; 41: Emitter Electrode; 42: Base electrode: 43: Collector electrode: 44: Also serves as collector region? Substrate: 12Ah32A:p
"Region: 12 & 32B: High purity or p-region:
22A:? Region: 22B: High purity or n-region; Black circle: Electron; White circle: Hole.

Claims (1)

[Claims] 1. The region where electrons, which are minority carriers, are injected and travel is alternately a high pH concentration (p+) region and a high purity or p-type low*degree (p″″) region. It consists of repeating, armpit? The electron affinity of the semiconductor in the region is smaller than that of the high-purity or p'''' region conductor, and the thickness of each unit layer in the region is thin enough to allow electrons to pass through the region by tunneling effect. 2. A semiconductor device characterized in that a region in which holes, which are minority carriers, are injected and travels is a high purity (n+) region and a high purity or all low concentration (n−) region. regions are alternately repeated, and the sum of the electron affinity and energy gap of the semiconductor in the region is greater than that of the high purity or n-region semiconductor, and The thickness is
411FiR is a semiconductor device that has a thin structure so that holes can pass through the region due to the tunnel effect.