JPH04251934A - Semiconductor device - Google Patents

Abstract

PURPOSE:To provide an integrated circuit using an InP substrate on which high-speed, high-voltage heterojunction bipolar transistors are formed. CONSTITUTION:A semiconductor device includes heterojunction bipolar transistors formed on an InP substrate. The bipolar transistor has an InP emitter region and base region, which form a heterojunction. The base region is composed of an InGaAs molecular layer and an InP molecular layer, forming a superlattice with a local band structure of a uniform composition. The average composition ratio of InGaAs in the superlattice increases in the direction from emitter to collector.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a semiconductor device, and more particularly to a semiconductor device using an InP substrate and suitable for constructing an integrated circuit including a hetero-bipolar transistor.

Heterojunction bipolar transistors have advantages over homojunction bipolar transistors, such as the ability to easily obtain a high current amplification factor, and have been actively researched and developed in recent years. In particular, InP in the emitter and InP in the base.
InGaAs/InP heterojunction bipolar transistors using GaAs are attracting attention for their potential applications in high-speed optical communications, such as driving InGaAsP lasers by taking advantage of their high current drive capability.

0003

2. Description of the Related Art A conventional technique will be explained with reference to FIG. FIG. 3(A) shows the band structure of a homobipolar transistor according to the prior art.

[0004] The case of an npn type homobipolar transistor in which the emitter, base, and collector are formed of the same material will be shown. In a homobipolar transistor, the conduction band energy Ec and the valence band energy E in each region are
The difference in v is almost uniform. When a forward bias is applied between the emitter and the base, the potential of the base region is lowered relative to the potential of the emitter region, as shown by the broken line in the figure. When the energy in the base region drops to a certain point,
Electrons are injected from the emitter region into the base region. Electrons injected into the base region move through the base region by diffusion and reach the collector region.

Note that, at the same time, holes are injected from the base region to the emitter region. This forms the base current. The operating speed of such a homobipolar transistor is mainly determined by the travel time of electrons after leaving the emitter region until reaching the low resistivity region of the collector region. When a bias voltage is applied to the collector region,
An electric field is generated in the collector region and carriers are accelerated and transported, so the operating speed is mainly determined by the transport time in the base region.

[0006] In order to increase the operating speed, it is effective to narrow the base region and shorten the base travel time during which carriers pass through the base region by diffusion. Figure 3(B)
shows one structure of a heterobipolar transistor suitable for forming on an InP substrate. Emitter region is n-type I
The base region is made of p-type InGaAs, and the collector region is made of n-type InGaAs. The band gaps of the base and collector regions are narrower than the band gaps of the emitter region. Therefore, when a forward bias is applied to the base region and the potential is lowered as shown by the broken line, electrons are injected from the emitter region to the base region, but at this time, holes are injected from the base region to the emitter region. rarely occurs. Therefore, an extremely high current amplification factor can be achieved.

Furthermore, InGaAs has high electron mobility and can transport electrons at high speed. Therefore, a high-speed transistor with a current cutoff frequency fT exceeding 140 GHz can be realized.

[0008] Since the bandgap of the collector region is narrow, it is difficult to increase the collector breakdown voltage. That is, when a high voltage is applied to the collector, tunneling occurs from the valence band of the base region to the collector region.

The structure of FIG. 3B is called a single heterojunction bipolar transistor because one heterojunction exists between the emitter region and the base region. FIG. 3C shows a second configuration of a heterobipolar transistor according to the prior art.

The emitter region is made of n-type InP, the base region is made of p-type InGaAs, and the collector region is made of n-type InP. In this configuration, the emitter region and the base region are similar to the configuration in FIG. 3(B), and it is possible to obtain a high current amplification factor.

Furthermore, since the collector region is formed of InP with a wide bandgap, the collector breakdown voltage can be increased.

0012

[Problem to be solved by the invention] Figures 3 (A), (B),
In the configuration (C), carriers move through the base region by diffusion. Therefore, the base region transit time is mainly determined by the carrier diffusion rate and the base width. Attempting to operate at higher speeds requires narrowing the base area, but if the base area is made too narrow,
The base resistance increases and the RC in the base region charging circuit
The time constant increases. For this reason, the expected high-speed operation cannot be achieved.

Furthermore, the hetero bipolar transistor shown in FIG. 3(B) can provide a higher current amplification factor and faster operating speed than the homo bipolar transistor. However, the bandgap of the collector region is narrow, making it difficult to increase the collector breakdown voltage. Therefore, even if it operates at high speed as a discrete device, the collector breakdown voltage is insufficient as an element in an integrated circuit that performs voltage level shifting. This poses a problem as a transistor structure used in an integrated circuit.

The hetero bipolar transistor shown in FIG. 3(C) has a high collector breakdown voltage and can solve the problems of the hetero bipolar transistor shown in FIG. 3(B). However, in the configuration of FIG. 3C, when the base region is forward biased, a potential barrier tends to occur in the collector region as shown by the broken line. Therefore, electrons injected from the emitter region into the base region are inhibited from traveling by the potential barrier in front of the collector region.

An object of the present invention is to provide a hetero-bipolar transistor that operates at high speed. Another object of the present invention is to provide a heterobipolar transistor with high collector breakdown voltage.

[0016]

[Means for Solving the Problems] A semiconductor device of the present invention includes:
A semiconductor device having a hetero bipolar transistor formed on an InP substrate, having an emitter region made of InP, and a base region forming a heterojunction with the emitter region, wherein the base region of the hetero bipolar transistor is made of InGaAs molecules. A superlattice structure consisting of an InP molecular layer and a local band structure having an approximately average composition;
It is characterized by including a superlattice structure in which the average composition ratio of As increases from the emitter side to the collector side.

Further, the semiconductor of the present invention is a semiconductor device having a hetero bipolar transistor formed on an InP substrate, having an emitter region made of InP, and a base region forming a heterojunction with the emitter region. ,
The collector region of the hetero bipolar transistor is InG.
A superlattice structure composed of a molecular layer of aAs and a molecular layer of InP, in which the local band structure shows a band structure with an approximately average composition, and the average composition ratio of InGaAs in the superlattice structure is far from the base region. It is characterized by including a superlattice structure that decreases according to

[0018]

[Operation] In order to speed up the base travel time without narrowing the base width, it is sufficient to generate an accelerating electric field (drift electric field) within the base region. One way to generate an accelerating electric field in the base region is to generate an electric field with a built-in potential in the base region. InP substrate
One method for generating a built-in electric field in a P/InGaAs-based hetero bipolar transistor is to form the base region from an InGaAsP mixed crystal and gradually change the composition within the base region.

However, it is extremely difficult to precisely control the ratios of In and Ga and As and P in the InGaAsP mixed crystal. In particular, in a hetero-bipolar transistor for high-speed operation, the base width is already on the order of 100 nm or less, and it is extremely difficult to change the composition with high precision over such a short distance.

[0020] Also, one method of improving the collector breakdown voltage without forming a potential barrier in the collector region is to form a composition gradient in the collector region and increase the band gap as the distance from the base region increases. I
In the nP/InGaAs hetero bipolar transistor, the collector region may be formed of InGaAsP and the composition may be gradually changed within the collector region.

However, as mentioned above, InGaAs
It is difficult to control the composition of P with high precision. By the way, a superlattice structure in which the thickness of each layer is sufficiently small compared to the spread of the carrier wave function has a property that its properties are similar to those of a mixed crystal of a superlattice structure forming substance. Therefore, in the superlattice structure, by changing the number of constituent molecular layers, it is possible to achieve the same effect as changing the composition of a mixed crystal. In this case, there is no need to change the composition within each layer.

A superlattice structure is constructed from an atomic layer of InGaAs and an atomic layer of InP, and InGaAs with an average composition
If the composition is configured such that the component increases from the emitter side to the collector side, it is possible to exhibit the same function as an InGaAsP composition gradient mixed crystal in which the InP composition is high on the emitter side and low on the collector side.

[0023] In this way, a drift electric field can be generated in the base region and carriers can be transported at high speed. In addition, the collector region is made of an InGaAs molecular layer,
If it is composed of an InP molecular layer and its average composition is such that the InGaAs component decreases as it moves away from the base region, a collector region with high breakdown voltage can be formed without forming a potential barrier on the base region side. be able to.

[0024]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Examples of the present invention will be described below with reference to the drawings. FIG. 1(A) is a schematic diagram for schematically explaining the properties of the superlattice SL. InGaAs1 region and In
A superlattice structure 2 in which molecular layers of both are arranged alternately is arranged between the P region 3 and the P region 3. Furthermore, in this superlattice structure 2, the ratio of the number of InGaAs molecular layers to the number of InP molecular layers is gradually changed so that the InGaAs component is high on the InGaAs region side and the InP component is high on the InP region side. Then, the band structure within the superlattice structure is
As shown by the pseudo broken lines Eco and Evo, the composition is approximately determined by the average composition. That is, InGaAs
By laminating a molecular layer and an InP molecular layer, InG
It can exhibit properties equivalent to those of aAsP mixed crystal.

FIG. 1B schematically shows a case where such a superlattice structure is employed in the base region of a hetero-bipolar transistor. Emitter region 5 is made of InP, and collector region 7 is made of InGaAs. Note that the composition of this collector region is selected so as to be lattice matched to the InP substrate. The base region 6 is formed of a superlattice stack of alternating InP molecular layers and InGaAs molecular layers, and is selected so that its average composition gradually changes. That is, the P component of InGaAsP is set to be high in the portion of the base region in contact with the emitter region, and the P component is selected to decrease toward the collector side. Due to such an average composition change, the energy Eco of the pseudo conduction band gradually decreases within the base region toward the collector region, as shown by the broken line.

At this time, a drift electric field is generated in the base region, and carriers injected from the emitter region into the base region are accelerated by the drift electric field and transported at high speed toward the collector region. Note that by keeping the InP component on the emitter region side of the base region 6 above a predetermined value, the substantial advantages of a hetero-bipolar transistor can be obtained.

FIG. 1C shows a configuration in which the above-mentioned superlattice structure is employed in the collector region. A portion of the collector region 7 in contact with the base region is formed with a superlattice structure in which an InP molecular layer and an InGaAs molecular layer are laminated. Further, in this superlattice structure, the average composition is selected such that the InP component increases as the distance from the base region increases. Therefore, the pseudo conduction band energy Eco within this superlattice structure gradually increases as shown by the broken line.

With this configuration, a potential barrier is not generated in the portion of the collector region in contact with the base region, and the band gap of the collector region is gradually increased as the distance from the base region increases. Collector breakdown voltage can be increased.

Note that the collector region which is more than a certain distance away from the base region can be an InP region. FIG. 2 shows the structure of a hetero-bipolar transistor according to a more specific embodiment of the present invention.

FIG. 2(A) shows the cross-sectional structure, and FIG. 2(B)
indicates a planar configuration. n+ type InGaA for forming an ohmic contact on the semi-insulating InP substrate 11
An s layer 12 is formed on which an n- type InP region 13 is formed.
A collector region 13 composed of a and n- type superlattice regions 13b is laminated. The n- type superlattice layer 13b is an InP/In layer in which the InP component gradually increases as the distance from the base region increases, as shown in FIG. 1(C).
It is formed of a GaAs superlattice layer. For example, 1 of the superlattice
The unit is composed of 10 molecular layers, and the number of InP molecular layers in each unit is 2, 4, 8, 10 from top to bottom.
(The order of epitaxial growth is reversed.) Note that the composition can be gradually changed by repeating each unit a predetermined number of times depending on the thickness of the collector region. For example, repeat 3 times (2, 2, 4
), (4, 4, 4), ... The thickness of one unit of the superlattice must be chosen such that within each unit the wave function sufficiently reaches the neighboring unit. For example, 1 unit is 1
0 molecular layer.

A base region 14 formed of a p-type superlattice is arranged above the collector region 13. This p-type base region 14 is formed of a composition gradient superlattice region as shown in FIG. 1(B). For example, one unit is made of 10 molecular layers, and the InGaAs composition ratio on the top surface is selected so as to fully obtain the advantages of a hetero-bipolar transistor and to generate a sufficient built-in potential gradient toward the collector side.

For example, the molar ratio of InGaAs is selected to be 0.4 on the emitter side and approximately 1 on the collector side. The width of the base region 14 is, for example, 50 to 7
The thickness is approximately 0 nm. Within this base width, the average composition of the superlattice is designed so that the InGaAs component gradually increases from the emitter side to the collector side. p
On the mold base region 14, an n+ type InGa layer is formed to facilitate forming ohmic contact with the n type InP layer 15a.
An emitter region 15 made of an As layer 15b is formed.

In this way, the emitter region 15 and the base region 14 form a heterojunction, and a heterobipolar transistor is constructed in which the composition of the base region 14 and part of the collector region 13 gradually changes.

On the surface of the n+ type InGaAs layer 12,
A collector electrode 16 is formed, a base electrode 17 is formed on the surface of the p-type base region 14, and an n+-type InGaA
An emitter electrode 18 is formed on the s-layer 15b.

The regions where these electrodes are to be formed are shown in FIG.
), it is formed in a stepwise manner to expose a sufficient area for electrode formation. For example, emitter region 15
The plane has a size of approximately 4×6 μm.

[0036] One unit of the superlattice is, for example, about 10 molecular layers or less, and by stacking superlattice units with the same composition or gradually changing composition, a superlattice structure that exhibits the same effect as a composition gradient mixed crystal can be created. form. The compositional gradient can have any desired shape, such as linear or exponential variation.

The impurity concentration of the n+ type InGaAs layer is:
For example, the impurity concentration is about 1019 cm-3, and the impurity concentration of n- type InP region 13a is about 1017 cm-3. Moreover, within the superlattice structure, InGaA
The impurity concentration of the S molecular layer is set high and the impurity concentration of the InP molecular layer is set low, so that a desired impurity concentration is obtained as a whole.

A semiconductor device as shown in FIG. 2 can be manufactured by the following steps. Metal-organic chemical vapor phase epitaxy (M) capable of atomic layer epitaxial growth (ALE)
A semi-insulating InP substrate 11 is installed in the (OCVD) device,
Trimethylindium, Ga for In as source gas
Triethylgallium is used for the gas, arsine is used for the As, phosphine is used for the P, and H2 is used as the carrier gas. These source gases are selectively controlled and supplied onto the substrate,
Perform MOCVD growth or ALE growth. for example,
When MOCVD growth is performed, the substrate temperature is set at about 600° C., and the necessary raw material gases are supplied at the same time. AL
When performing E-growth, the substrate temperature is set to, for example, 350° C., and the gas necessary for molecular layer growth is alternately switched and supplied.

By such growth, a laminated structure as shown in FIG. 2A is formed, an etching mask is formed, and etching is performed so that necessary portions remain. The superlattice structure in the collector region is formed to have a thickness of, for example, about 200 nm, and the superlattice structure in the base region is formed to have a thickness of, for example, about 50 nm.

The hetero bipolar transistor structure as described above allows the base region to be made extremely thin without making it extremely thin.
The base running time of the carrier can be shortened. In addition, without creating a potential barrier in the collector region,
Collector breakdown voltage can be increased.

Other semiconductor elements such as a light emitting element and a light receiving element are formed on the same InP substrate 11 to form an optoelectronic integrated circuit (OE).
IC) device can be configured. Although the present invention has been described above with reference to Examples, the present invention is not limited thereto. For example, it will be obvious to those skilled in the art that various changes, improvements, combinations, etc. are possible.

[0042]

[Effects of the Invention] As explained above, according to the present invention,
In a heterobipolar transistor, a drift electric field can be generated in the base region.

[0043] Therefore, a high-speed operation hetero-bipolar transistor is provided. Further, a hetero bipolar transistor is provided that has improved collector breakdown voltage without generating a potential barrier in the collector region.

Therefore, it becomes easy to integrate the hetero bipolar transistor into an integrated circuit.

[Brief explanation of the drawing]

FIG. 1 is a conceptual diagram for explaining an embodiment of the present invention. Figure 1 (A) is a conceptual diagram of a band structure to explain the properties of a superlattice, Figure 1 (B) is a conceptual diagram to explain the properties of a base region formed by a superlattice, and Figure 1 (C) is a conceptual diagram of a band structure to explain the properties of a superlattice. FIG. 2 is a conceptual diagram for explaining the properties of a collector region formed by a superlattice.

FIG. 2 is a diagram showing a semiconductor device according to an embodiment of the present invention. FIG. 2(A) is a cross-sectional view, and FIG. 2(B) is a plan view.

FIG. 3 is a conceptual diagram for explaining a bipolar transistor according to the prior art. 3(A) is a conceptual diagram showing the band structure of a homo bipolar transistor, FIG. 3(B) is a conceptual diagram for explaining the first configuration of a hetero bipolar transistor, and FIG. 3(C) is a conceptual diagram showing the first configuration of a hetero bipolar transistor. FIG. 2 is a conceptual diagram for explaining the configuration of No. 2;

[Explanation of symbols]

7, 13 Collector area 6, 14 Base area 5, 15 Emitter area

Claims (5)

[Claims] 1. Formed on an InP substrate, having an emitter region (5) made of InP, and a base region (6).
is a semiconductor device having a hetero bipolar transistor forming a heterojunction with an emitter region (5), wherein the base region (6) of the hetero bipolar transistor is composed of an InGaAs molecular layer and an InP molecular layer,
The superlattice structure includes a superlattice structure in which the local band structure exhibits a band structure with an approximately average composition, and the average composition ratio of InGaAs in the superlattice structure increases from the emitter side to the collector side. Characteristic semiconductor devices. 2. Formed on an InP substrate, having an emitter region (5) made of InP, and a base region (6).
A semiconductor device having a hetero bipolar transistor forming a heterojunction with an emitter region (5), the collector region (7) of the hetero bipolar transistor forming a heterojunction with an emitter region (5).
is a superlattice structure composed of an InGaAs molecular layer and an InP molecular layer, and the local band structure shows a band structure with an approximately average composition, and the InGaAs in the superlattice structure
1. A semiconductor device comprising a superlattice structure in which an average composition ratio of is decreased as the distance from a base region increases. 3. Formed on an InP substrate, having an emitter region (5) made of InP, and a base region (6).
is a semiconductor device having a hetero bipolar transistor forming a heterojunction with an emitter region (5), wherein the base region (6) of the hetero bipolar transistor is composed of an InGaAs molecular layer and an InP molecular layer,
The superlattice structure includes a superlattice structure in which the local band structure shows a band structure with a substantially average composition, and the average composition ratio of InGaAs in the superlattice structure increases from the emitter side to the collector side, The collector region (7) of the hetero bipolar transistor is composed of an InGaAs molecular layer and an InGaAs molecular layer.
A superlattice structure composed of a molecular layer of P, in which the local band structure exhibits a band structure with an approximately average composition, in which the average composition ratio of InGaAs in the superlattice structure decreases as the distance from the base region increases. A semiconductor device characterized by including a structure. 4. The semiconductor device according to claim 1, wherein the superlattice structure of the base region (6) forms a substantial heterojunction with an emitter region formed of InP. 5. Further, the emitter region and the collector region include a low resistivity InGaAs region.
, 4. The semiconductor device according to any one of .