JP2008270760A - Semiconductor wafer and manufacturing method, and semiconductor element - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a semiconductor wafer having satisfactory surface smoothness for GaInNAsSb on an InP substrate and a semiconductor laminated structure with low crystal defect density, and to provide its manufacturing method and a semiconductor element. <P>SOLUTION: The method for manufacturing a semiconductor wafer comprises the steps of fixing the InP substrate 1 to a substrate-fixing part of an MBE device, and growing Ga<SB>1-x</SB>In<SB>x</SB>N<SB>y</SB>Sb<SB>z</SB>As<SB>1-y</SB>(0.4≤x≤0.8, 0<y≤0.2, 0≤z≤0.1), with a lattice constant difference to the InP substrate of not lower than -0.5% nor higher than +0.5% on the InP substrate, up to a film thickness of not smaller than 0.5 μm with a substrate-surface temperature measured by a pyrometer of >490°C and ≤530°C. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a semiconductor wafer, a manufacturing method thereof, and a semiconductor element.

  A light-receiving element capable of sensing near-infrared light is used for the analysis of animals and plants or substances derived from animals and plants, environmental gas analysis, nighttime surveillance cameras, and the like. For this light receiving element, a compound semiconductor having an absorption edge wavelength of about 1.7 μm, such as InGaAs, is used, but it is desirable to expand the light receiving sensitivity to a longer wavelength region. There is also a high demand for light-emitting elements in the above wavelength range. Focusing on GaInNAs containing N as a compound semiconductor having a band gap energy corresponding to the above wavelength range or longer wavelength side, a semiconductor laser of a light emitting unit incorporating a GaInNAs layer (thickness 7 nm) in a quantum well structure, A photodiode having a non-strained light absorption layer GaInNAs (film thickness: 0.5 μm) has been proposed (Patent Document 1).

Although the above-described GaInNAs can in principle increase the light receiving sensitivity to the long wavelength side in the near infrared region, GaInNAs with good crystallinity is difficult to form and has not yet been put into practical use. For this reason, in a semiconductor laser in which a GaInNAs layer (film thickness: 7 nm) is arranged in a strained single quantum well active layer on a GaAs substrate, studies have been made to improve the crystallinity of the GaInNAs (Patent Document 2). In this examination, the growth temperature when a GaInNAs well layer is grown on the GaAs barrier layer is investigated, and when the growth temperature is 350 ° C. to 490 ° C., a GaInNAs layer (film thickness 7 nm) having a smooth surface can be obtained. I can do it. It is stated that when the growth temperature is higher or lower than the above temperature range, the GaInNAs surface is undulated and the surface becomes rough. Note that the growth temperature usually refers to a temperature obtained by measuring the substrate surface with a pyrometer (usually provided in a film forming apparatus).
JP-A-9-219563 JP-A-11-87848

  In spite of the efforts to improve the crystallinity of GaInNAs as described above, a method effective only for the compound semiconductor to be tested has been studied. Is insufficient. For this reason, there is a demand for accumulation of experimental data that makes it possible to find a more basic cause and find a versatile improvement method. In particular, in the near infrared region, it is desired to improve the crystallinity of an epitaxially grown layer of GaInNAs on an InP substrate, which occupies an important industrial position, particularly to improve the surface smoothness. The present invention provides a semiconductor wafer having a low crystal defect density, a semiconductor element, and a method for manufacturing the semiconductor wafer, which can be obtained by forming a GaInNAs layer having good crystallinity and surface smoothness on an InP substrate. Objective.

The method of manufacturing a semiconductor device according to the present invention includes a step of attaching an InP substrate to a substrate mounting portion of an MBE (Molecular Beam Epitaxy) apparatus, and a lattice constant difference between the InP substrate on the InP substrate is −0.5% or more +0. Ga _1-x In _x N _y Sb _z As _1-y (0.4 ≦ x ≦ 0.8, 0 <y ≦ 0.2, 0 ≦ z ≦ 0.1) in the range of 5% or less And a step of growing to a film thickness of 0.5 μm or more in a state where the substrate surface temperature of the measurement exceeds 490 ° C. and is 530 ° C. or less.

By the above method, a semiconductor wafer having a semiconductor multilayer structure including a Ga _1-x In _x N _y As _1-y layer with good surface smoothness and a low crystal defect density can be obtained. By including Sb, the surface properties can be particularly improved.

The semiconductor wafer of the present invention is manufactured by the above-described manufacturing method, and Ga _1-x In _x N _y Sb _z As _1-y (0.4 ≦ x ≦ 0.8, 0 <y ≦ 0.2, 0 ≦ Rms roughness measured by AFM (Atom Force Microscopy) with z ≦ 0.1) is 10 nm or less. This _can reduce the lattice defect density at the interface between the upper layer to be epitaxially grown on _{Ga 1-x In x N y} Sb z As 1-y layer. For this reason, for example, in the case of a light receiving element, an increase in dark current due to a high lattice defect density at the interface can be prevented in a structure in which a pn junction or a pin junction is formed.

In addition, the semiconductor wafer of the present invention includes an InP substrate and Ga _1-x In _x N _y Sb _z As _1-y (0.4 ≦ x ≦ 0. 8, 0 <y ≦ 0.2, 0 ≦ z ≦ 0.1) layers. The difference in lattice constant between the Ga _1-x In _x N _y Sb _z As _1-y layer and the InP substrate is in the range of −0.5% or more and + 0.5% or less, and Ga _1−x In _x N and _y Sb _z as _1-y layer, an interface between the upper layer positioned in contact with the _{Ga 1-x in x N y} Sb z as 1-y layer, the irregularities correspond to the following Rms roughness 10nm measured by AFM It is characterized by having.

With the above configuration, it is possible to form a semiconductor element having a laminated structure with a low crystal defect density, including an active layer or a light receiving layer having an absorption edge wavelength corresponding to the long wavelength side in the near infrared region and excellent surface smoothness. it can. Here, the Rms roughness of the interface is determined by removing each layer above the Ga _1-x In _x N _y Sb _z As _1-y layer of the semiconductor element by etching and then AFM (Atomic Force Microscopy). Can be obtained by a built-in automatic measurement operation. In this case, the Rms roughness in the visual field of 5 μm × 5 μm is used. Further, when the material selectivity is not good in etching, the interface in the cross section may be observed to obtain the roughness of the interface in a one-dimensional manner, and it may be determined whether or not the unevenness corresponds to the Rms roughness of 10 nm or less. That is, Rms roughness (5 μm × 5 μm) of 10 nm or less may be substituted by AFM measurement of the unevenness of the interface in the cross section. For observation of the cross section, a transmission electron microscope (TEM) or a high resolution scanning electron microscope (HR-SEM) may be used.

  The semiconductor element of the present invention is manufactured using any one of the above semiconductor wafers. With this configuration, it is possible to obtain a semiconductor device having a laminated structure with a low crystal defect density, including an active layer or a light receiving layer having an absorption edge wavelength corresponding to the long wavelength side in the near infrared region and having excellent surface smoothness. The semiconductor element of the present invention includes other semiconductor elements without being limited to the light receiving element and the light emitting element as long as the above configuration is satisfied.

The semiconductor element can be a photodiode. As a result, a photodiode having a small light current and a light receiving sensitivity on the long wavelength side in the near infrared region can be obtained.

According to the present invention, Ga _1-x In _x N _y Sb _z As _1-y (0...) Having good crystallinity and surface smoothness on an InP substrate that occupies an industrially important position in the near infrared region. 4 ≦ x ≦ 0.8, 0 <y ≦ 0.2, 0 ≦ z ≦ 0.1), and as a result, a semiconductor wafer, a semiconductor element, and a semiconductor having a semiconductor stacked structure with a low crystal defect density A wafer manufacturing method can be obtained.

FIG. 1 is a cross-sectional view showing a semiconductor stacked structure constituting a photodiode in a semiconductor element according to an embodiment of the present invention. An InGaAs buffer layer 2 is positioned on an n-type InP substrate 1 doped with S, and Ga _1-x In _x N _y Sb _z As _1-y (0.4 ≦ x ≦ 0.8, 0 < (y ≦ 0.2, 0 ≦ z ≦ 0.1) (hereinafter referred to as “GaInNAsSb”) layer 3 and AlInAs layer 4 are sequentially epitaxially grown. The GaInNAsSb layer 3 when there is no state other than a state not containing Sb (z = 0) is referred to as a GaInNAs layer 3.

  For example, when the semiconductor laminated structure 10 of FIG. 1 is used for a photodiode, as shown in FIG. 2, the GaInNAsSb layer 3 is used as a light receiving layer, and the AlInAs layer 4 thereabove is used as a window layer. Next, an impurity introduction diffusion mask pattern 5 is formed on the AlInAs window layer 4, and p-type impurity Zn or the like is introduced from the impurity introduction diffusion mask pattern 5 so as to reach the GaInNAsSb light-receiving layer 3 to form a p-type region. 15 is formed. By forming the p-type region 15, a pn junction or a pin junction is formed in the GaInNAsSb light receiving layer 3. The p-type region is formed of AuZn so that the p-type electrode 12 is in ohmic contact, and the n-type electrode 11 is formed of AuGeNi on the back surface of the InP substrate 1 so as to be in ohmic contact.

  A reverse bias voltage is applied to the p-part electrode 12 and the n-part electrode 11, and the depletion layer is spread over the GaInNAsSb light-receiving layer 3 and waits for the incidence of light. When light having a wavelength shorter than the absorption edge wavelength (near-wavelength long wavelength side) of the GaInNAsSb light-receiving layer 3 is incident, a photocurrent is generated, and the incidence of light can be detected. If the surface of the GaInNAsSb light-receiving layer 3 is not smooth, the window layer 4 having good crystallinity cannot be grown, and a high lattice defect density is formed at the interface between the GaInNAsSb light-receiving layer 3 and the window layer 4. . For this reason, a portion having a high lattice defect density is located at the end of the pn junction or the pin junction, resulting in an increase in dark current and a deterioration in light receiving sensitivity. For this reason, the Rms roughness of the surface of the GaInNAsSb light receiving layer 3 is preferably 10 nm or less. Alternatively, it is preferable that the roughness of the interface between the GaInNAsSb light-receiving layer 3 and the AlINAs window layer 4 is equivalent to an Rms roughness of 10 nm or less when observed in a cross section.

  Next, a method for manufacturing the semiconductor multilayer structure 10 shown in FIG. 1 will be described with reference to FIGS. FIG. 3 shows the flow of each process of the manufacturing method, and FIG. 4 shows the MBE manufacturing apparatus. First, a standard manufacturing method will be described according to the steps in FIG. In the standard manufacturing method, first, the InP substrate 1 is attached to the substrate attaching portion 33 of the MBE film forming apparatus 30 shown in FIG. 4 (step S1 in FIG. 3). An infrared heating device 33 is built in the board mounting portion 33 and wired so that the set temperature can be controlled from the outside while viewing the temperature display of a pyrometer or the like. Next, the InGaAs buffer layer 2 is epitaxially grown on the InP substrate 1 (step S2). Next, the GaInNAsSb layer 3 is epitaxially grown on the InGaAs buffer layer 2 by MBE (step S3). The laminated structure 10 including the InP substrate is attached to the substrate attachment portion 33 having a rotation and heating mechanism, heated as described above, and brought into a rotation state. In the film formation, a molecular beam cell (E-type electron gun) as an evaporation source is arranged corresponding to the elements constituting the layer. When the solid phase element is In, Ga, Sb, As, In, Molecular beam cells that emit molecular beams of Ga, Sb, and As are arranged separately. In FIG. 4, three molecular beam cells including the gas cell 31 are shown, but some of them are omitted. In order to manufacture the semiconductor element, an upper layer, for example, a window layer is epitaxially grown on the GaInNAsSb layer 3 (step S4). Here, in FIG. 4, the semiconductor multilayer structure 10 at each stage includes the InP substrate 1. Further, the substrate temperature refers to the surface temperature of the semiconductor laminated structure at each stage measured by a pyrometer.

Next, the manufacturing method of the modification 1 and the modification 2 with respect to the above-mentioned standard manufacturing method is demonstrated according to FIG.
(Modification 1): In Modification 1, the InGaAs buffer layer 2 is formed in advance regardless of the film formation method (the MBE film formation apparatus 30 in FIG. 4 may or may not be used). The InP substrate can be attached to the substrate attachment portion 33 (corresponding to the course indicated by the broken line in FIG. 3). In the first modification, the manufacturing method may be used in which the InP substrate on which the InGaAs buffer layer 2 is formed in advance is attached to the substrate attaching portion 33 in step S1, step S2 is skipped, and the process proceeds to step S3.
(Modification 2): In Modification 2, the GaInNAsSb layer 3 can be grown directly on the InP substrate 1 without the InGaAs buffer layer. In the manufacturing method in which the InGaAs buffer layer according to the second modification is omitted, step S2 is omitted in FIG. 3, and the process immediately proceeds from step S1 to step S3. Of course, it is not necessary to attach the InP substrate on which the InGaAs buffer layer has been formed in advance to the substrate mounting portion.
(Modification 3): In Modification 3, an InGaAs buffer layer 2 and a GaInNAsSb layer 3 are grown on an InP substrate 1, and another film forming apparatus (for example, MOVPE (Metal Organic Vapor) is formed.
The upper layer (for example, InP layer) in step 4 is grown in the (Phase Epitaxy) film forming apparatus).

  In order to adjust the chemical composition and the film forming speed, the opening and closing of the cell shutter and the substrate shutter are adjusted, and an attached computer is used for the control. The substrate temperature or the like is measured by a pyrometer. For RHEED (reflection high electron energy diffraction) observation, a RHEED electron gun is arranged so that electrons are incident on the laminated structure 10 at a shallow incident angle, and a fluorescent screen (RHEED screen) for obtaining a diffraction image thereof, and its A camera that captures the diffraction image is provided at a position in the diffraction direction. The RHEED is used for evaluating the crystallinity of the laminated structure 10, grasping the film forming process, and the like. In addition, observation devices such as a mass spectrometer, a beam monitor, and a quartz film thickness meter are attached. Among the molecular beams and the like, those that are not incorporated into the laminated structure 10 are exhausted to the vacuum exhaust system. The liquid nitrogen shroud is used for adsorption of impurities generated by collision of molecular beams. The film forming apparatus communicates with the vacuum exhaust system via a gate valve.

  When forming the GaInNAsSb layer, the substrate temperature is controlled to be over 490 ° C. and below 530 ° C. by controlling the infrared heating device of the substrate mounting portion 33. By maintaining the substrate temperature at a temperature higher than 490 ° C. and lower than 530 ° C., the surface smoothness of the growing GaInNAsSb layer 3 can be improved. As described above, the substrate temperature can be confirmed by a pyrometer.

  In order to introduce nitrogen (N) into the GaInNAsSb layer, nitrogen gas is supplied to the gas line, an excited state of nitrogen is obtained by the nitrogen plasma cell 31, and the laminated structure 10 is irradiated with this excited molecular nitrogen beam. .

  As described above, the surface smoothness of the GaInNAsSb layer 3 is improved by setting the substrate temperature to be higher than 490 ° C. and lower than 530 ° C. However, in that case, using the mechanism incorporated in the AFM based on the AFM image. The surface can be scanned to automatically determine Rms roughness. AFM apparatuses are commercially available from many companies, and there is not much difference in performance regardless of which AFM apparatus is used. In the embodiment of the present invention, the Rms roughness of the surface of the GaInNAsSb layer 3 can be 10 nm or less.

  Further, after the semiconductor element is manufactured, the cross section of the semiconductor multilayer structure 10 is exposed, and the surface roughness can be evaluated at the interface between the GaInNAsSb layer 3 and the AlInAs layer 4 which is the upper layer. FIG. 5 is a diagram for explaining an analogy method for evaluating unevenness of an interface in a cross section. In FIG. 5, the macro length of the interface of one visual field is set to about 2 μm, and the difference Δh or the height difference between the maximum height and the minimum height in the visual field of the GaInNAsSb layer 3 is obtained. This Δh is obtained in at least 30 visual fields, preferably 50 visual fields or more, and is averaged by arithmetic average. In the embodiment of the present invention, the average height difference is preferably 30 nm or less by analogy from the surface Rms of 10 nm or less.

(Growth of GaInNAsSb layer on InP substrate)
Next, in the embodiment of the present invention, the influence of the substrate temperature when forming the GaInNAsSb layer on the InP substrate will be described in comparison with the content disclosed in the above-mentioned Patent Document 2. FIG. 6A is a diagram showing a result of observing the surface of the InGaAs layer with an optical microscope when an InGaAs layer not containing N is epitaxially grown on the InP substrate 1 at a substrate temperature of 542 ° C. to 544 ° C. FIG. 6B is a schematic diagram thereof. According to FIGS. 6A and 6B, when the substrate temperature reaches about 542 ° C., coarse surface defects due to As missing are generated. The diameter of the surface defect is as large as 100 μm. If such a surface defect occurs, the crystallinity is greatly deteriorated and the finished semiconductor laminated structure is hardly usable. The problem of As missing is not mentioned in Patent Document 2, and is a problem peculiar to the present invention.

  Even in GaInNAsSb containing N, when the substrate temperature is high, As is lost and a huge surface defect cannot be prevented. Therefore, the substrate temperature for the growth of the GaInNAsSb layer must be in a temperature range in which As is not lost, and this determines the upper limit of the substrate temperature. In order to prevent GaInNAsSb from coming out of As, it is necessary to set the substrate temperature to 530 ° C. or less based on the knowledge of the growth experiment of GaInNAsSb itself, including the knowledge of the growth of InGaAs.

In the embodiment of the present invention, the substrate temperature for film formation by MBE is over 490 ° C. In Patent Document 2, the temperature range in which GaInNAsSb having a thickness of about 7 nm is grown on a GaAs substrate is clearly different from 350 ° C. or more and 490 ° C. or less. This is considered to be caused by the following crystal difference.
(1) The film thickness of the GaInNAsSb layer in the present invention is 0.5 μm (500 nm) or more, which is very thick compared to the film thickness of 7 nm in Patent Document 2. When the film thickness is extremely thin, if the substrate temperature is raised, aggregation or the like occurs and breaks, making it difficult to form a very thin and flat film. A difference in film thickness is also considered as one cause of the difference in the substrate temperature. That is, the setting of the substrate temperature on the high temperature side is set based on different phenomena between the present invention and the invention disclosed in Patent Document 2.
(2) GaInNAsSb grown on an InP substrate has a smaller difference in lattice constant from the substrate than GaInNAsSb on a GaAs substrate, and crystal distortion due to the substrate is small. The In composition is larger for GaInNAsSb on the InP substrate. For this reason, it is considered that the temperature range in which phase separation into a phase region having a high Ga concentration and a low In concentration and a phase region having a high In concentration and a low Ga concentration is performed on a GaAs substrate is a lower temperature region. . The thin film thickness may also affect the low temperature side shift of the phase separation temperature. As a result, in the invention disclosed in Patent Document 2, the upper limit of the phase separation temperature is about 350 ° C., whereas it is about 490 ° C. on the InP substrate. As a result, in the invention disclosed in Patent Document 2, the lower limit of the substrate temperature is 350 ° C., whereas in the present invention, the lower limit of the substrate temperature is over 490 ° C.
One or both of the above (1) and (2) affect the above-described difference in substrate temperature.

Example 1
Next, the function and effect of the present invention will be described with reference to examples. Using Example 1 of the present invention and Comparative Example 1, the surface smoothness of the GaInNAs layer was investigated. The test specimen had the configuration shown in FIG. 1 and was manufactured by the following procedure.
(Invention Example 1): A Si-doped n-type InGaAs buffer layer 2 was epitaxially grown on an InP substrate 1 having a surface orientation (100) doped with S by MBE. The film thickness was 1.5 μm, the In composition was 53%, and the carrier concentration was 5 × 10 ¹⁶ cm ⁻³ . Next, the GaInNAs layer 3 was epitaxially grown by the MBE method. The group III Ga composition was 46%, the In composition 54%, the group V N composition 1.5%, and the As composition 98.5%. The film thickness was 2.5 μm. Doping is not performed. Next, the AlInAs layer 4 was epitaxially grown by the MBE method. The In composition is 52% and is lattice matched to the InP substrate. The film thickness was 0.6 μm. The substrate temperature for growing the GaInNAs layer 3 was 502 ° C.
(Comparative Example 1): In the same manner as Example 1 of the present invention, an InGaAs layer 2, a GaInNAs layer 3, and an AlInAs layer 4 were sequentially formed on an InP substrate having a surface orientation (100) doped with S by MBE. The semiconductor multilayer structure 10 was formed by epitaxial growth. The composition of each semiconductor layer was the same as Example 1 of the present invention. The difference is that in Comparative Example 1, the substrate temperature when growing the GaInNAs layer 3 was 484 ° C.

  About the test body of the said invention example 1 and the comparative example 1, when the GaInNAs layer 3 was grown, the surface of the GaInNAs layer 3 was observed by AFM about the area | region of 5 micrometers x 5 micrometers. FIG. 7A shows an AFM image and FIG. 7B shows a schematic diagram of the surface of Example 1 of the present invention. For Comparative Example 1, FIG. 8A shows an AFM image, and FIG. 8B shows a schematic diagram thereof. “Zrange” shown in FIG. 7A means a full scale in the Z direction, that is, the uneven direction. Accordingly, in FIG. 7A, the full scale in the concave-convex direction is taken as 50 nm, whereas in FIG. 8A, the full scale in the same direction is taken as 100 nm. That is, in Comparative Example 1, the unevenness is large, and it is indicated that an appropriate AFM image cannot be obtained unless the full scale is doubled that of Example 1 of the present invention. In other words, in FIG. 7A, an enlarged image of fine irregularities on the surface is captured.

  FIG. 7 (a), (b) and FIG. 8 (a), (b) are obvious at a glance, but in Example 1 of the present invention, the dense and uniform irregularities are recognized by the enlarged magnification. In contrast, in Comparative Example 1, coarse projections are intermittently positioned despite the reduction in magnification. When Rms roughness (5 μm × 5 μm) was obtained for both the above-mentioned specimens, it was 3.5 nm in Example 1 of the present invention. On the other hand, in Comparative Example 1, the Rms roughness was 14.7 nm, which is about 4 times that of Example 1 of the present invention. From this example, in the growth of GaInNAs by the MBE method, the surface properties of the GaInNAs layer change drastically due to the difference of 20 ° C. between the substrate temperature of 502 ° C. (Example 1 of the present invention) and the substrate temperature of 484 ° C. (Comparative Example 1). It has been found.

(Example 2)
Further investigation was performed on the influence of the substrate temperature on the crystallinity of GaInNAs by the MBE method. The test method was performed as follows.
(Test method): By an MBE method, an In _0.57 Ga _0.43 As layer (film thickness: 0.15 μm) is formed on the InP substrate (100) as a buffer layer, and Ga _0.43 In _0.57 N _1-x As _x (film thickness 1 μm) was grown. A valve cracking cell was used to supply As. RF (Radio), a nitrogen source
Frequency) Plasma discharge conditions and growth rate were fixed. X-ray rocking curves and photoluminescence (PL) emission intensities of crystals with different growth temperatures (substrate temperatures) were measured, and the results are shown in FIGS. 9 and 10, respectively.
(Result): From the X-ray rocking curve, the diffraction peak of the GaInNAs layer becomes broader at a substrate temperature of 460 ° C. or lower, and the diffraction peak of GaInNAs is broader than the diffraction peak at a temperature higher than 480 ° C. Slightly low. Therefore, in order to obtain the best crystallinity, it is necessary to raise the substrate temperature to over 500 ° C. or 515 ° C. exceeding 490 ° C. Further, the PL emission intensity is much higher at the growth temperature of 500 ° C. than the growth temperature of 480 ° C., and the crystallinity is good. Further, the crystallinity becomes better as the As / III flux ratio, that is, the partial pressure ratio between the group III element (Ga, In) and As is larger.

  According to Example 2, it was confirmed that the crystallinity of GaInNAs was not the best at a substrate temperature of 480 ° C., and good crystallinity was obtained at a higher temperature of 500 ° C. or 515 ° C. On the high temperature side, the problem of As missing at about 540 ° C. shown in FIGS. 6A and 6B must be considered. By combining the phenomena shown in FIG. 6 to FIG. 10, it is rational that there is a substrate temperature condition in which a GaInNAs layer having good surface smoothness can be grown on an InP substrate in a narrow range of 490 ° C. to 530 ° C. I was able to confirm with.

(Example 3)
Next, Example 3 in which the effect of Sb on the surface smoothness was confirmed will be described. Example 2 of the present invention, which is a test body in Example 3, includes the GaInNAsSb layer 3 having the configuration shown in FIG. 1 and with z not zero, and was prepared by the following procedure.
(Invention Example 2): An InGaAs buffer layer 2 was epitaxially grown on the S-doped InP substrate 1 by MBE. The growth conditions (composition, film thickness, doping) of the InGaAs buffer layer 2 are the same as those of Example 1 of the present invention in Example 1 above. Next, the GaInNAsSb layer 3 containing Sb was epitaxially grown by the MBE method. The substrate temperature at this time was 500 ° C. The group III Ga composition was 45%, the In composition 55%, the group V N composition 1.5%, the As composition 95.9%, and the Sb composition 2.6%. The film thickness was 2.5 μm. Doping is not performed. Next, the AlInAs layer 4 was epitaxially grown by the MBE method. The growth conditions of AlInAs are the same as those of Example 1 of the present invention in Example 1.

  With respect to Inventive Example 2, when the GaInNAsSb layer 3 was grown, the surface of the GaInNAsSb layer 3 was observed with an AFM in a 5 μm × 5 μm region. FIG. 11A shows an AFM image and FIG. 11B shows a schematic diagram of the surface of Example 2 of the present invention. In Example 2 of the present invention, since Sb is included, the surface smoothness is excellent, and Zrange 10 nm can be finally recognized by microscopically enlarging larger than in the case of FIGS. 7 and 8. Therefore, by including Sb, the surface smoothness of the GaInNAsSb layer 3 is remarkably improved.

  When Rms roughness (5 μm × 5 μm) was obtained for Example 2 of the present invention, it was 0.27 nm. The measurement results of Rms roughness of Example 2 of the present invention are shown in Table 1 together with Example 1 of the present invention and Comparative Example 1 which are test bodies in Example 1.

  According to Table 1, the Rms roughness is improved to about ¼ by increasing the substrate temperature. By adding Sb to GaInNAs after increasing the substrate temperature, the Rms roughness can be suppressed by 1/10 or more compared to the case where only measures for increasing the substrate temperature are taken.

(Forms other than the embodiment)
1. Form of Semiconductor Element Other than the form as a photodiode, any element may be used. Any film thickness may be used regardless of the light emitting element and the light receiving element as long as the film thickness is 0.5 μm or more.
2. Material of Buffer Layer and Window Layer When a photodiode is used, the lattice matching buffer layer may be anything as long as it is lattice matched to the InP substrate, and most widely, In _x Ga _y Al _1-xy As (0 ≦ x ≦ 1, 0 ≦ y ≦ 1) and InP. The window layer may be anything and is not limited to InAlAs and InP.

  Although the embodiments and examples of the present invention have been described above, the embodiments and examples of the present invention disclosed above are merely examples, and the scope of the present invention is the implementation of these inventions. It is not limited to the form. The scope of the present invention is indicated by the description of the scope of claims, and further includes meanings equivalent to the description of the scope of claims and all modifications within the scope.

  By using the semiconductor wafer, the semiconductor element, and the method for manufacturing a semiconductor wafer of the present invention, a GaInNAsSb layer having excellent surface smoothness can be obtained, and the crystallinity of the semiconductor multilayer structure can be improved. For this reason, for example, a photodiode having a low dark current on the long wavelength side in the near-infrared region and excellent in light receiving sensitivity can be obtained.

It is a figure for demonstrating the epitaxial laminated structure in the semiconductor element of embodiment of this invention. It is a figure which shows the photodiode based on the laminated structure of FIG. It is a figure for demonstrating the manufacturing method of the semiconductor element of embodiment of this invention. It is a figure explaining an MBE film-forming apparatus. It is a figure which shows the method of calculating | requiring the surface roughness of a GaInNAsSb layer in the interface of a GaInNAsSb layer and an upper layer. It is a figure which shows the surface defect resulting from As omission produced when a substrate temperature is higher than embodiment of this invention, (a) is an optical microscope image, (b) is the schematic diagram. (A) is an AFM image of Example 1 of the present invention in Example 1, and (b) is a schematic diagram thereof. (A) is the AFM image of the comparative example 1 in Example 1, (b) is the schematic diagram. FIG. 4 is a diagram showing an X-ray rocking curve showing the influence of the substrate temperature on the crystallinity of GaInNAs in Example 2. It is a figure which shows the photoluminescence light emission intensity which shows the influence of the substrate temperature which acts on the crystallinity of GaInNAs in Example 2. FIG. (A) is an AFM image of Example 2 of the present invention in Example 3, and (b) is a schematic diagram thereof.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 InP substrate, 2 InGaAs layer, 3 GaInNAsSb layer, 4 AlInAs layer, 5 Impurity diffusion mask pattern, 10 Semiconductor laminated structure, 11 n part electrode, 12 p part electrode, 15 p type area | region, 30 MBE film-forming apparatus, 31 Nitrogen plasma cell, 33 substrate mounting part.

Claims (5)

Attaching the InP substrate to the substrate mounting portion of the MBE (Molecular Beam Epitaxy) device;
Ga _1-x In _x N _y Sb _z As _1-y (0.4 ≦ x ≦ 0) in which the lattice constant difference with the InP substrate is in the range of −0.5% to + 0.5% on the InP substrate. .8, 0 <y ≦ 0.2, 0 ≦ z ≦ 0.1) is grown to a film thickness of 0.5 μm or more in a state where the substrate surface temperature measured with a pyrometer is over 490 ° C. and below 530 ° C. A method for producing a semiconductor wafer, comprising: Produced by the production method according to claim _{1, Ga 1-x In x} N y Sb z As 1-y (0.4 ≦ x ≦ 0.8,0 <y ≦ 0.2,0 ≦ z ≦ 0.1 A semiconductor wafer characterized in that the Rms roughness measured by AFM (Atom Force Microscopy) is 10 nm or less. An InP substrate;
Ga _1-x In _x N _y Sb _z As _1-y (0.4 ≦ x ≦ 0.8, 0 <y ≦ 0.2, 0 ≦ z) having a film thickness of 0.5 μm or more located on the InP substrate. ≦ 0.1) layer,
The lattice constant difference between the Ga _1-x In _x N _y Sb _z As _1-y layer and the InP substrate is in the range of −0.5% or more and + 0.5% or less,
Said _{Ga 1-x In x N y} Sb z As 1-y layer, an interface between the upper layer positioned in contact with the _{Ga 1-x In x N y} Sb z As 1-y layer, Rms measured by AFM A semiconductor wafer having irregularities corresponding to a roughness of 10 nm or less.   A semiconductor device manufactured using the semiconductor wafer according to claim 2.   The semiconductor device according to claim 4, wherein the semiconductor device is a photodiode.

Images