JPH0445976B2 - - Google Patents

Description

[Detailed description of the invention]

(1) Technical Field of the Invention The present invention relates to a method for manufacturing a semiconductor device. For details, please refer to the previous patent applications filed by the applicant of this patent application (Japanese Patent Application No. 55-82035 and Japanese Patent Application No. 54-171026).
The present invention relates to a method of manufacturing a semiconductor device having an improved structure of a high electron mobility transistor according to the present invention. (2) Background of the technology A high electron mobility transistor is a transistor with high electron mobility, which is a transistor with a high electron mobility that consists of a group of electrons (two-dimensional electron gas) accumulated in the vicinity of a single heterojunction surface formed by joining two types of semiconductors with different electron affinities. An active semiconductor device in which the electron concentration is controlled by a control electrode to control the impedance of a conductive path formed by the group of accumulated electrons between a pair of input and output electrodes provided elsewhere. A major feature of high electron mobility transistors is that the electron mobility of the above-mentioned stored electron group (two-dimensional electron gas) is maintained at low temperatures, e.g., below 77 It becomes extremely large at high temperatures. In other words, the above accumulated electron group (two-dimensional electron gas) is
Although it is in the semiconductor layer (channel layer) with high electron affinity, there is a very thin layer in the vicinity of the heterojunction.
Since it is accumulated within a range of about 100㎓, it is spatially separated from the layer (electron supply layer) made of a semiconductor with a small electron affinity, and its electron mobility is not affected by impurity scattering. Therefore, at low temperatures where the effect of impurity scattering prevents the increase in electron mobility, an extremely large electron mobility, that is, 1.5 × 10 ⁵ cm ² /VS at 77〓,
It has been experimentally confirmed that the value is approximately 5×10 ⁵ /VS at 5〓. Incidentally, the electron mobility of such a group of accumulated electrons (two-dimensional electron gas) is more or less influenced by impurity scattering based on ionized impurities present near the heterojunction. Impurities that cause this adverse effect include those contained in a semiconductor with a low electron affinity that serves as an electron supply layer, ie, an electron source, and those contained in a channel layer, ie, a semiconductor with a high electron affinity. In conventional high electron mobility transistors, the electron supply layer contains n-type impurities to serve as an electron source, and only the channel layer does not contain impurities. Therefore, in order to increase electron mobility by reducing the influence of scattering due to impurities contained in a semiconductor with low electron affinity, it is necessary to increase the electron surface concentration of the accumulated electron group (two-dimensional electron gas). This is because the screening effect of the electron scattering potential becomes greater as the electron surface concentration increases. In order to increase the electron plane concentration of the accumulated electron group (two-dimensional electron gas), it is sufficient to increase the n-type impurity concentration of the electron supply layer, and thereby the electron mobility can also be increased. However, on the other hand, if the n-type impurity concentration in the electron supply layer is increased, the concentration of n-type impurities diffused into the channel layer during high-temperature processes also increases, so the influence of impurity scattering based on the impurities present in this channel layer becomes large. As a result, electron mobility decreases. Therefore, the impurity concentration in the electron supply layer cannot be increased indefinitely. (3) Conventional technology and problems High electron mobility transistors using conventional technology have the effect of increasing the electron surface concentration of accumulated electron groups (two-dimensional electron gas), but the amount of impurities that diffuse into the channel layer increases. A layer (buffer layer) made of the same semiconductor as the electron supply layer and containing no impurities is interposed between the electron supply layer containing n-type impurities and the channel layer containing no impurities. Ta. by the way,
The larger the thickness of this buffer layer is, the more the impurities in the electron supply layer can be prevented from diffusing into the channel layer, but on the other hand, the larger the thickness is, the more the difference in electron affinity Since the effect of the resulting potential gap is attenuated, the electronic yield of the accumulated electron group (two-dimensional electron gas) decreases. In other words, the effect of the buffer layer naturally has a limit. Therefore, if the electron source of the accumulated electron group (two-dimensional electron gas) could be found in another element rather than the n-type impurity contained in the electron supply layer, the influence of impurity scattering could be minimized, and Since the electron surface concentration of the stored electron group (two-dimensional electron gas) can be made sufficiently large, the electron mobility should be able to be made even higher than that of the high electron mobility transistor in the prior art. (4) Purpose of the Invention The purpose of the present invention is to satisfy this requirement, and instead of relying on the n-type impurity contained in the electron supply layer as an electron source, it relies on other elements,
Moreover, the electron surface concentration of the accumulated electron group (two-dimensional electron gas) is sufficiently increased to increase the electron mobility of the accumulated electron group (two-dimensional electron gas) without increasing the influence of impurity scattering. An object of the present invention is to provide a method for manufacturing a high electron mobility transistor. (5) Structure of the Invention The structure of the present invention is as follows: (a) a first layer 2 made of a semiconductor 1 that does not substantially contain impurities; and a semiconductor 1 that does not substantially contain impurities; and a second layer 3 made of another semiconductor having an electron affinity smaller than the electron affinity of A control electrode 6 that allows transmission of electromagnetic waves is formed on the layer 3 of 2 or the first layer 2, and (c) the control electrode 6 is sandwiched between the control electrode 6 and the second layer that forms the upper layer. 3 or on the first layer 2, input/output electrodes 4 are formed to produce a semiconductor structure; (d) this semiconductor structure is maintained at a low temperature and the control electrode 6 is formed on the first layer 2; Electromagnetic waves are transmitted and irradiated to the second layer 3 of the semiconductor structure, and electrons are generated in the first layer 2 near the interface between the first layer 2 and the second layer 3. group (two-dimensional electron gas)
This is a method for manufacturing a semiconductor device including a step of generating (Claim 1). Further, in the above configuration, the second layer 3
and the control electrode 6 (when the first layer 2 is the lower layer and the second layer 3 is the upper layer), or
Between the second layer 3 and the substrate (when the second layer 3 is the lower layer and the first layer 2 is the upper layer),
By forming the layer 8 of the same semiconductor as the semiconductor constituting the second layer 3, which contains n-type impurities, the concentration of two-dimensional electron gas (electron group) generated is improved and control is easy. (Claim 2)
section). Furthermore, in any of the above configurations, the step of irradiating the electromagnetic waves is performed by housing the semiconductor structure in a low-temperature container, so that the semiconductor structure in which the two-dimensional electron gas (electron group) is generated is The device is
It is practical to continue storing the product in the low-temperature container (claim 3).
section). By the way, in the high electron mobility transistor manufactured by implementing the above manufacturing method, not only the first layer (hereinafter referred to as channel layer) 2 but also the second layer
Since the layer (hereinafter referred to as the electron supply layer) 3 also does not contain any impurities, the influence of impurity scattering is limited to a minimum.On the other hand, at low temperatures, electromagnetic waves such as light are at least applied to the electron supply layer 3. Since the semiconductor constituting the electron supply layer 3 or the shallow impurity level contained therein is excited by the irradiation, a sufficient amount of electron groups is accumulated near the hetero interface between the electron supply layer 3 and the channel layer 2. Since the conductive path 7 is made of two-dimensional electron gas, the electron mobility is higher than that in conventional high electron mobility transistors. In addition, in the high electron mobility transistor manufactured by implementing the above manufacturing method, the surface concentration of the two-dimensional electron gas (conductive path 7) accumulated near the hetero interface between the electron supply layer 3 and the channel layer 2 is as follows: Since it is determined by the amount of free electrons generated by irradiation with electromagnetic waves, the electron supply layer 3 should have a thickness that is sufficient to be expected to have the effect of preventing the above-mentioned impurity scattering, and should not contain impurities. It is desirable not to contain
In addition, in contact with this electron supply layer 3 that does not contain impurities, a layer that is made of the same semiconductor and contains n-type impurities (a layer that is made of the same semiconductor as the semiconductor that contains n-type impurities and constitutes the second layer 3) is added. The addition of layer (hereinafter referred to as additional layer) 8 is effective for increasing electron mobility (Claim 2). In this embodiment, the lower limit of the thickness that the electron supply layer 3 that does not contain impurities should have is a thickness that can suppress the above impurity diffusion, but the combination of semiconductors is gallium arsenide (GaAs), aluminum gallium arsenide ( 150 for AlGas)
It is about Å. Note that, as shown in FIG. 2, when the electron supply layer 3 is the upper layer of the channel layer 2, the position where the additional layer 8 containing n-type impurities should be added is made of a semiconductor having a small electron affinity. It is located between the electron supply layer 3 and the control electrode 6, and when the electron supply layer 3 is the lower layer of the channel layer 2, it becomes the lower layer of the electron supply layer 3. In the high electron mobility transistor manufactured by implementing the above manufacturing method, the amount of electrons accumulated near the hetero interface between the electron supply layer 3 and the channel layer 2 remains constant for at least one year as long as it is kept at a low temperature. Although no significant decrease is observed over the period of time, it is unavoidable that it will decrease exponentially with the passage of time. Therefore, it is practical to keep the high electron mobility transistor in a low temperature container (not shown) (claim 3).
section). In addition to this, there is also a means for intermittently irradiating electromagnetic waves such as light, such as a semiconductor light emitting element (not shown) disposed opposite to the control electrode 6.
It is extremely meaningful to provide a drive circuit (not shown) to operate this at desired times from the viewpoint of improving industrial applicability. The above configuration (Claims 1, 2, and 3)
This is because the life of the high electron mobility transistor in this fourth embodiment is extremely long compared to the life of the high electron mobility transistor in item 2). The two types of semiconductors that make up a high electron mobility transistor must meet the following conditions: (a) their lattice constants are the same or similar, (b) there is a large difference in electron affinity, and (c) there is a band gap. Although there are a large number of combinations, the present invention is applicable to any of them. Further, in high electron mobility transistors, the channel layer may be configured as an upper layer or as a lower layer of the electron supply layer, and the present invention can be applied to either case. In general, materials with high electron affinity have a small band gap and a long fundamental absorption edge wavelength, so if the electron supply layer is the lower layer,
The electromagnetic waves irradiated to the electron supply layer are partially absorbed and reduced in water in the channel layer, but especially when the channel layer is the upper layer, the thickness of the channel layer is extremely thin, about 1000 Å, so there is no substantial This is because it is durable enough for use. It is desirable that the wavelength of the electromagnetic waves used for irradiation is shorter than the basic absorption edge wavelength of the semiconductor constituting the electron supply layer, but as mentioned above, it also excites shallow impurity levels of impurities that are inevitably included in the semiconductor.
It is not an essential requirement that the wavelength be shorter than the above basic absorption edge wavelength. It goes without saying that the pinch-off voltage can be controlled by controlling the amount of electromagnetic waves such as light. (6) Embodiments Hereinafter, with reference to the drawings, methods for manufacturing high electron mobility transistors according to two embodiments of the present invention will be described to further clarify the structure and unique effects of the present invention. An example of the combination of semiconductors is a combination of gallium arsenide (GaAs) and aluminum gallium arsenide (AlGaAs). First Example (corresponding to Claim 1) See Figure 1 Figure 1 shows a completed high electron mobility transistor manufactured by carrying out the method for manufacturing a semiconductor device according to an example of the present invention. A cross-sectional view in the state is shown. In the figure, 1 is a substrate made of gallium arsenide (GaAs) containing chromium (Cr), etc., and 2 is a gallium arsenide (GaAs) layer substantially free of impurities formed on the substrate 1 with lattice matching. The channel layer is approximately 2000 Å thick. Reference numeral 3 designates an electron supply layer made of an aluminum gallium arsenide (AlGaAs) layer which is substantially free of impurities and is formed on the channel layer 2 in a lattice-matched manner, and has a thickness of approximately 500 Å. The above semiconductor layers 2 and 3 are successively formed using a molecular beam epitaxial growth method. 4
gold/germanium/germanium selectively formed on the input/output electrode (source/drain electrode) formation region.
Input/output electrodes (source/drain electrodes) made of gold (AuGe/Au) layers are heat-treated at approximately 450°C to penetrate through the electron supply layer 3 and alloy to the top of the channel layer 2 to form the alloy layer 5. Input/output electrodes (source/drain electrodes) 4
and channel layer 2 are resistively connected. 6
is a control electrode (gate electrode), which in this embodiment is an aluminum (Al) curtain having a thickness of about 100 Å. With such a thickness, electromagnetic waves such as light are transmitted, so an opening for light irradiation is unnecessary. In the state after the completion of the above manufacturing process, the electron group near the interface between the channel layer 2 and the electron supply layer 3 is not accumulated, but as shown by the arrow,
When irradiated with electromagnetic waves such as light, a group of electrons 7 accumulates in the vicinity of the above-mentioned hetero interface, excluding the lower regions of the input and output electrodes (source and drain electrodes) in the figure, and functions as a normally-on type high electron mobility transistor. I will do it. In addition, after being irradiated with electromagnetic waves such as light, it is kept at a low temperature in order to retain the electron group (two-dimensional electron gas) accumulated near the hetero interface between the electron supply layer 3 and the channel layer 2. Therefore, it must be stored in a low-temperature container (not shown). Here, the electron surface concentration and the electron mobility of the accumulated electron group (two-dimensional electron gas) in the state after electromagnetic wave irradiation can be measured at 77〓 and 5〓,
The results were as shown in the table below. Electron surface concentration and electron mobility measurement temperature in one embodiment of the present invention Electronic surface concentration Electron mobility cm ^-2 cm ² /Vsec 77〓 4.5×10 ¹¹ 200000 5〓 5.0×10 ¹¹ 1200000 The measurement results are shown in the table below. This is a significant improvement over the values of the prior art shown in . Measurement of electronic surface concentration and electron mobility in an example of conventional technology Temperature of electronic surface concentration Electron mobility cm ^-2 cm ² /Vsec 77〓 3.5×10 ¹¹ 160000 5〓 3.4×10 ¹¹ 540000 Second embodiment (Claims) Corresponding to paragraph 2) 2nd
Refer to Figures FIG. 2 shows a cross-sectional view of a completed high electron mobility transistor manufactured by implementing a method for manufacturing a semiconductor device according to another embodiment of the present invention. 1st
The difference from the figure is that the thickness of the electron supply layer 3 is somewhat thinner, at 150 Å, and that an additional layer 8 made of n-type aluminum gallium arsenide (AlGaAs) is formed on the electron supply layer 3 and has a thickness of about 350 Å. There is only one thing. In this case, the accumulated electron group exists before being irradiated with electromagnetic waves such as light, but its value increases as shown in the table below by irradiation with electromagnetic waves such as light.

[Table] Note that it must be kept at a low temperature, so it must be stored in a low temperature container (not shown). Furthermore, it is interesting to note that when the irradiation amount of electromagnetic waves such as light is increased, as shown in Figure 3, the electron surface concentration increases as expected, but the electron mobility and the electron surface concentration are not necessarily related. They are not proportional, and the relationship between the electron surface concentration and electron mobility has a maximum value as shown in FIG. In the figure, curve A
shows the measurement results at 77〓, and curve B shows the measurement results at 5〓. As mentioned above, since the electronic surface concentration decreases exponentially with time, the amount of irradiation of electromagnetic waves such as light may be determined by taking into account the characteristics shown in FIG. (7) Effects of the Invention As explained above, according to the present invention, the electron source does not depend on the n-type impurity contained in the electron supply layer, but instead relies on other elements, specifically, electromagnetic waves such as light. Furthermore, the electron mobility of the storage element group (two-dimensional electron gas) can be increased without increasing the influence of impurity scattering by sufficiently increasing the electron surface concentration of the storage electron group (two-dimensional electron gas). A method for manufacturing a high electron mobility transistor can be provided.

[Brief explanation of the drawing]

FIG. 1 is a cross-sectional view of a high electron mobility transistor manufactured by implementing a method for manufacturing a semiconductor device according to an embodiment of the present invention, and FIG. 2 is a cross-sectional view of a semiconductor device according to another embodiment of the present invention. FIG. 2 is a cross-sectional view of a high electron mobility transistor manufactured by implementing the manufacturing method.
FIG. 3 is a graph showing the relationship between the electron surface concentration radiation amount when the electron supply layer of a high electron mobility transistor manufactured by implementing the method of manufacturing a semiconductor device according to the present invention is irradiated with electromagnetic waves such as light. 4 is a graph showing the relationship between electron mobility and electron surface concentration in a state corresponding to FIG. 3. 1... Semi-insulating semiconductor (gallium arsenide) substrate, 2... First layer (channel layer/i-type gallium arsenide layer), 3... Second layer (electron supply layer/i
type aluminum gallium arsenic layer), 4...
Input/output electrodes (source/drain electrodes), 5...
Alloying region of input/output electrodes, 6... Control electrode (gate electrode), 7... Accumulated electron group (two-dimensional electron gas)
A conductive path consisting of 8...layer of the same semiconductor as the semiconductor constituting the second layer 3 containing n-type impurities (additional layer/n-type aluminum gallium arsenide layer), A...77〓 Curve showing the measured electron mobility versus electronic surface concentration, B...Curve showing the measured electron mobility versus electronic surface concentration at 5〓.

Claims (1)

[Scope of Claims] 1. A first layer 2 made of a semiconductor that does not substantially contain impurities and has an electron affinity that is smaller than that of the semiconductor that does not substantially contain impurities. Second layer 3 made of another semiconductor
a control electrode 6 that allows transmission of electromagnetic waves on the second layer 3 or the first layer 2 forming the upper layer of the semiconductor stack; and forming input/output electrodes 4 on the second layer 3 or the first layer 2 forming the upper layer with the control electrode 6 in between, thereby manufacturing a semiconductor structure, The semiconductor structure is held at a low temperature, and electromagnetic waves are transmitted through the control electrode 6 and irradiated to the second layer 3 of the semiconductor structure, so that the first layer 2 and the second layer 3 are irradiated with electromagnetic waves.
A method for manufacturing a semiconductor device, comprising the step of generating a group of electrons (two-dimensional electron gas) in the first layer 2 near the interface with the first layer 3. 2 between the second layer 3 and the control electrode 6 (when the first layer 2 is the lower layer and the second layer 3 is the upper layer), or between the second layer 3 and the substrate. (when the second layer 3 is the lower layer and the first layer 2 is the upper layer) contains an n-type impurity and
2. The method of manufacturing a semiconductor device according to claim 1, further comprising the step of forming a layer 8 of the same semiconductor as the semiconductor constituting the layer 3. 3. The step of irradiating the electromagnetic wave is performed by housing the semiconductor structure in a low-temperature container, and the semiconductor device in which the electron group (two-dimensional electron gas) is generated is subsequently stored in the low-temperature container. A method of manufacturing a semiconductor device according to claim 1 or 2, characterized in that: