Production of GaAs Single Crystal Using the manufacturing apparatus shown in FIG. 2, a C (carbon)-doped semi-insulating GaAs single crystal includ-ing a straight body portion having a diameter of 156 mm and a length of 200 mm is produced in accordance with the VB method. A GaAs polycrystal is used as the GaAs source material. B₂O₃ is used as the sealing member. A temperature distribution in the manufacturing apparatus is adjusted such that a temperature gradient in a crystal growth direction at a crystal growth interface is 2° C./cm, and the GaAs single crystal is grown. Next, the grown GaAs single crystal is cooled to the room temperature (25° C.) at 25° C./min. A temperature difference in the GaAs single crystal at this time is 20±0.2° C. in the entire straight body portion of the GaAs single crystal. An outer circumferential surface of the cooled GaAs single crystal is ground, to thereby produce the GaAs single crystal including the straight body portion having a diameter of 152.4 mm.

Production of GaAs Single Crystal substrate The GaAs single crystal obtained above is sliced along a plane of the straight body portion perpendicular to a central axis, and front and rear main surfaces are both mirror-finished by mechanical polishing and chemical mechanical polishing (CMP), to thereby produce two (one from the seed crystal side and one from the finally solidified portion side) GaAs single crystal substrates each having a diameter of 152.4 mm and a thickness of 700 µm. There is no affected layer on both of the polished front and rear main surfaces. Various types of cleaning that can maintain the mirror surfaces after polishing may be performed. For each of the GaAs single crystal substrates obtained as described above, a type (compressive or tensile) of a residual strain in a tangential direction in an outer circumferential portion is evaluated from a Raman shift by measuring a Raman spectrum using a Raman spectrophotometer (HR evolution manufactured by HORIBA, Ltd.). Determination of an ori-entation of the residual strain in the tangential direction does not specify the magnitude, and thus, a measurement method other than the Raman shift may be used as long as it can determine the orientation. Evaluation of an average value of the magnitude of the residual strain in the outer circumfer-ential portion represented by an absolute value ?Sr−St? of a difference between a radial strain component Sr and a tangential strain component St can be performed based on, for example, the photoelastic method described in Appl. Phys. Lett. 47 (1985) pp. 365-367. Specifically, a light irradiation diameter on each substrate main surface is φ100 µm. The above-described average value of the magnitude of the residual strain in the outer circumferential portion is calculated by scanning and measuring the entire main sur-face at 0.5 mm-pitch square lattice points such that the center of the substrate main surface is included in the measurement points, and averaging all of the measurement values included in the outer circumferential portion extend-ing between an inner circumference located 10 mm inward from an outer circumference toward a center and a location located 5 mm inward from the outer circumference. The crystallinity is evaluated by an average EPD (etch pit density) in the entire location located 5 mm inward from the outer circumferential surface. Specifically, molten potas-sium hydroxide is used as an etchant. The EPD can be obtained by magnifying the main surface of the GaAs single crystal substrate by 100 times with a microscope and count-ing the number of etch pits within a field of view of 1 mm square (meaning a square of 1 mmxl mm; the same applies to the description below). An average value of the EPD can be obtained by counting the number of etch pits at 5 mm intervals along each of four directions equivalent to a <110> direction from the center of the main surface and averaging the resultant numbers. Furthermore, the average value of the EPD can also be obtained by counting the number of etch pits at 5 mm intervals along each of four directions equiva-lent to a <100> direction from the center of the main surface and averaging the resultant numbers.

Evaluation of Whether or Not Slip Occurs By applying a thermal history similar to a thermal history when a semiconductor layer is grown on the above-de-scribed GaAs single crystal substrate, evaluation of whether or not slip occurs is made. Specifically, a temperature of the above-described GaAs single crystal substrate is elevated to 600° C. at a rate of 40° C./min under a AsH₃ (arsine) atmosphere in an OMVPE (organic metal vapor phase growth) furnace, and the temperature is kept for 10 minutes, and then, the above-described GaAs single crystal substrate is cooled at a setting of 100° C./min. Thereafter, observation of whether or not slip occurs in the GaAs single crystal substrate is made with a differential interference microscope. The results are shown in Table 1. The invention claimed is:

Production of GaAs Single Crystal Using the manufacturing apparatus shown in FIG. 2, a C (carbon)-doped semi-insulating GaAs single crystal includ-ing a straight body portion having a diameter of 156 mm and a length of 200 mm is produced in accordance with the VB method. A GaAs polycrystal is used as the GaAs source material. B₂O₃ is used as the sealing member. A temperature distribution in the manufacturing apparatus is adjusted such that a temperature gradient in a crystal growth direction at a crystal growth interface is 2° C./cm, and the GaAs single crystal is grown. Next, the grown GaAs single crystal is cooled to the room temperature (25° C.) at 25° C./min. A temperature difference in the GaAs single crystal at this time is 20±0.2° C. in the entire straight body portion of the GaAs single crystal. An outer circumferential surface of the cooled GaAs single crystal is ground, to thereby produce the GaAs single crystal including the straight body portion having a diameter of 152.4 mm.

Production of GaAs Single Crystal substrate The GaAs single crystal obtained above is sliced along a plane of the straight body portion perpendicular to a central axis, and front and rear main surfaces are both mirror-finished by mechanical polishing and chemical mechanical polishing (CMP), to thereby produce two (one from the seed crystal side and one from the finally solidified portion side) GaAs single crystal substrates each having a diameter of 152.4 mm and a thickness of 700 µm. There is no affected layer on both of the polished front and rear main surfaces. Various types of cleaning that can maintain the mirror surfaces after polishing may be performed. For each of the GaAs single crystal substrates obtained as described above, a type (compressive or tensile) of a residual strain in a tangential direction in an outer circumferential portion is evaluated from a Raman shift by measuring a Raman spectrum using a Raman spectrophotometer (HR evolution manufactured by HORIBA, Ltd.). Determination of an ori-entation of the residual strain in the tangential direction does not specify the magnitude, and thus, a measurement method other than the Raman shift may be used as long as it can determine the orientation. Evaluation of an average value of the magnitude of the residual strain in the outer circumfer-ential portion represented by an absolute value ?Sr−St? of a difference between a radial strain component Sr and a tangential strain component St can be performed based on, for example, the photoelastic method described in Appl. Phys. Lett. 47 (1985) pp. 365-367. Specifically, a light irradiation diameter on each substrate main surface is φ100 µm. The above-described average value of the magnitude of the residual strain in the outer circumferential portion is calculated by scanning and measuring the entire main sur-face at 0.5 mm-pitch square lattice points such that the center of the substrate main surface is included in the measurement points, and averaging all of the measurement values included in the outer circumferential portion extend-ing between an inner circumference located 10 mm inward from an outer circumference toward a center and a location located 5 mm inward from the outer circumference. The crystallinity is evaluated by an average EPD (etch pit density) in the entire location located 5 mm inward from the outer circumferential surface. Specifically, molten potas-sium hydroxide is used as an etchant. The EPD can be obtained by magnifying the main surface of the GaAs single crystal substrate by 100 times with a microscope and count-ing the number of etch pits within a field of view of 1 mm square (meaning a square of 1 mmxl mm; the same applies to the description below). An average value of the EPD can be obtained by counting the number of etch pits at 5 mm intervals along each of four directions equivalent to a <110> direction from the center of the main surface and averaging the resultant numbers. Furthermore, the average value of the EPD can also be obtained by counting the number of etch pits at 5 mm intervals along each of four directions equiva-lent to a <100> direction from the center of the main surface and averaging the resultant numbers.

Evaluation of Whether or Not Slip Occurs By applying a thermal history similar to a thermal history when a semiconductor layer is grown on the above-de-scribed GaAs single crystal substrate, evaluation of whether or not slip occurs is made. Specifically, a temperature of the above-described GaAs single crystal substrate is elevated to 600° C. at a rate of 40° C./min under a AsH₃ (arsine) atmosphere in an OMVPE (organic metal vapor phase growth) furnace, and the temperature is kept for 10 minutes, and then, the above-described GaAs single crystal substrate is cooled at a setting of 100° C./min. Thereafter, observation of whether or not slip occurs in the GaAs single crystal substrate is made with a differential interference microscope. The results are shown in Table 1. The invention claimed is:

Production of GaAs Single Crystal Using the manufacturing apparatus shown in FIG. 2, a C (carbon)-doped semi-insulating GaAs single crystal includ-ing a straight body portion having a diameter of 156 mm and a length of 200 mm is produced in accordance with the VB method. A GaAs polycrystal is used as the GaAs source material. B₂O₃ is used as the sealing member. A temperature distribution in the manufacturing apparatus is adjusted such that a temperature gradient in a crystal growth direction at a crystal growth interface is 2° C./cm, and the GaAs single crystal is grown. Next, the grown GaAs single crystal is cooled to the room temperature (25° C.) at 25° C./min. A temperature difference in the GaAs single crystal at this time is 20±0.2° C. in the entire straight body portion of the GaAs single crystal. An outer circumferential surface of the cooled GaAs single crystal is ground, to thereby produce the GaAs single crystal including the straight body portion having a diameter of 152.4 mm.

Production of GaAs Single Crystal substrate The GaAs single crystal obtained above is sliced along a plane of the straight body portion perpendicular to a central axis, and front and rear main surfaces are both mirror-finished by mechanical polishing and chemical mechanical polishing (CMP), to thereby produce two (one from the seed crystal side and one from the finally solidified portion side) GaAs single crystal substrates each having a diameter of 152.4 mm and a thickness of 700 µm. There is no affected layer on both of the polished front and rear main surfaces. Various types of cleaning that can maintain the mirror surfaces after polishing may be performed. For each of the GaAs single crystal substrates obtained as described above, a type (compressive or tensile) of a residual strain in a tangential direction in an outer circumferential portion is evaluated from a Raman shift by measuring a Raman spectrum using a Raman spectrophotometer (HR evolution manufactured by HORIBA, Ltd.). Determination of an ori-entation of the residual strain in the tangential direction does not specify the magnitude, and thus, a measurement method other than the Raman shift may be used as long as it can determine the orientation. Evaluation of an average value of the magnitude of the residual strain in the outer circumfer-ential portion represented by an absolute value ?Sr−St? of a difference between a radial strain component Sr and a tangential strain component St can be performed based on, for example, the photoelastic method described in Appl. Phys. Lett. 47 (1985) pp. 365-367. Specifically, a light irradiation diameter on each substrate main surface is φ100 µm. The above-described average value of the magnitude of the residual strain in the outer circumferential portion is calculated by scanning and measuring the entire main sur-face at 0.5 mm-pitch square lattice points such that the center of the substrate main surface is included in the measurement points, and averaging all of the measurement values included in the outer circumferential portion extend-ing between an inner circumference located 10 mm inward from an outer circumference toward a center and a location located 5 mm inward from the outer circumference. The crystallinity is evaluated by an average EPD (etch pit density) in the entire location located 5 mm inward from the outer circumferential surface. Specifically, molten potas-sium hydroxide is used as an etchant. The EPD can be obtained by magnifying the main surface of the GaAs single crystal substrate by 100 times with a microscope and count-ing the number of etch pits within a field of view of 1 mm square (meaning a square of 1 mmxl mm; the same applies to the description below). An average value of the EPD can be obtained by counting the number of etch pits at 5 mm intervals along each of four directions equivalent to a <110> direction from the center of the main surface and averaging the resultant numbers. Furthermore, the average value of the EPD can also be obtained by counting the number of etch pits at 5 mm intervals along each of four directions equiva-lent to a <100> direction from the center of the main surface and averaging the resultant numbers.

Evaluation of Whether or Not Slip Occurs By applying a thermal history similar to a thermal history when a semiconductor layer is grown on the above-de-scribed GaAs single crystal substrate, evaluation of whether or not slip occurs is made. Specifically, a temperature of the above-described GaAs single crystal substrate is elevated to 600° C. at a rate of 40° C./min under a AsH₃ (arsine) atmosphere in an OMVPE (organic metal vapor phase growth) furnace, and the temperature is kept for 10 minutes, and then, the above-described GaAs single crystal substrate is cooled at a setting of 100° C./min. Thereafter, observation of whether or not slip occurs in the GaAs single crystal substrate is made with a differential interference microscope. The results are shown in Table 1. The invention claimed is:

Production of GaAs Single Crystal Using the manufacturing apparatus shown in FIG. 2, a C (carbon)-doped semi-insulating GaAs single crystal includ-ing a straight body portion having a diameter of 156 mm and a length of 200 mm is produced in accordance with the VB method. A GaAs polycrystal is used as the GaAs source material. B₂O₃ is used as the sealing member. A temperature distribution in the manufacturing apparatus is adjusted such that a temperature gradient in a crystal growth direction at a crystal growth interface is 2° C./cm, and the GaAs single crystal is grown. Next, the grown GaAs single crystal is cooled to the room temperature (25° C.) at 25° C./min. A temperature difference in the GaAs single crystal at this time is 20±0.2° C. in the entire straight body portion of the GaAs single crystal. An outer circumferential surface of the cooled GaAs single crystal is ground, to thereby produce the GaAs single crystal including the straight body portion having a diameter of 152.4 mm.

Production of GaAs Single Crystal substrate The GaAs single crystal obtained above is sliced along a plane of the straight body portion perpendicular to a central axis, and front and rear main surfaces are both mirror-finished by mechanical polishing and chemical mechanical polishing (CMP), to thereby produce two (one from the seed crystal side and one from the finally solidified portion side) GaAs single crystal substrates each having a diameter of 152.4 mm and a thickness of 700 µm. There is no affected layer on both of the polished front and rear main surfaces. Various types of cleaning that can maintain the mirror surfaces after polishing may be performed. For each of the GaAs single crystal substrates obtained as described above, a type (compressive or tensile) of a residual strain in a tangential direction in an outer circumferential portion is evaluated from a Raman shift by measuring a Raman spectrum using a Raman spectrophotometer (HR evolution manufactured by HORIBA, Ltd.). Determination of an ori-entation of the residual strain in the tangential direction does not specify the magnitude, and thus, a measurement method other than the Raman shift may be used as long as it can determine the orientation. Evaluation of an average value of the magnitude of the residual strain in the outer circumfer-ential portion represented by an absolute value ?Sr−St? of a difference between a radial strain component Sr and a tangential strain component St can be performed based on, for example, the photoelastic method described in Appl. Phys. Lett. 47 (1985) pp. 365-367. Specifically, a light irradiation diameter on each substrate main surface is φ100 µm. The above-described average value of the magnitude of the residual strain in the outer circumferential portion is calculated by scanning and measuring the entire main sur-face at 0.5 mm-pitch square lattice points such that the center of the substrate main surface is included in the measurement points, and averaging all of the measurement values included in the outer circumferential portion extend-ing between an inner circumference located 10 mm inward from an outer circumference toward a center and a location located 5 mm inward from the outer circumference. The crystallinity is evaluated by an average EPD (etch pit density) in the entire location located 5 mm inward from the outer circumferential surface. Specifically, molten potas-sium hydroxide is used as an etchant. The EPD can be obtained by magnifying the main surface of the GaAs single crystal substrate by 100 times with a microscope and count-ing the number of etch pits within a field of view of 1 mm square (meaning a square of 1 mmxl mm; the same applies to the description below). An average value of the EPD can be obtained by counting the number of etch pits at 5 mm intervals along each of four directions equivalent to a <110> direction from the center of the main surface and averaging the resultant numbers. Furthermore, the average value of the EPD can also be obtained by counting the number of etch pits at 5 mm intervals along each of four directions equiva-lent to a <100> direction from the center of the main surface and averaging the resultant numbers.

Evaluation of Whether or Not Slip Occurs By applying a thermal history similar to a thermal history when a semiconductor layer is grown on the above-de-scribed GaAs single crystal substrate, evaluation of whether or not slip occurs is made. Specifically, a temperature of the above-described GaAs single crystal substrate is elevated to 600° C. at a rate of 40° C./min under a AsH₃ (arsine) atmosphere in an OMVPE (organic metal vapor phase growth) furnace, and the temperature is kept for 10 minutes, and then, the above-described GaAs single crystal substrate is cooled at a setting of 100° C./min. Thereafter, observation of whether or not slip occurs in the GaAs single crystal substrate is made with a differential interference microscope. The results are shown in Table 1. The invention claimed is: