Epitaxial-silicon-wafer manufacturing method and epitaxial silicon wafer

Abstract

A manufacturing method of an epitaxial silicon wafer including a silicon wafer doped with boron and having a resistivity of 100 m.Math.cm or less and an epitaxial film formed on the silicon wafer includes: growing the epitaxial film on the silicon wafer; and applying a heat treatment on the epitaxial silicon wafer at a temperature of less than 900 degrees C.

Claims

1. A manufacturing method of an epitaxial silicon wafer comprising: a silicon wafer doped with boron and having a resistivity of 100 m.Math.cm or less; and an epitaxial film provided on a surface of the silicon wafer, the method comprising: growing the epitaxial film on the silicon wafer; and applying a heat treatment on the epitaxial silicon wafer at a temperature of less than 900 degrees C.; wherein an average oxygen concentration of the epitaxial film after the heat treatment is 1.710.sup.17 atoms/cm.sup.3 (according to ASTM F-121, 1979) or more; wherein the heat treatment is performed to satisfy a formula (1)
t3.7110.sup.56X.sup.7.03Y.sup.0.27 Z.sup.3.34T.sup.16.37 (1), where: X (10.sup.17 atoms/cm.sup.3) represents an oxygen concentration of the silicon wafer before growing the epitaxial film; Y (.Math.cm) represents a resistivity of the silicon wafer before growing the epitaxial film; Z (m) represents a film thickness of the epitaxial film; T (degrees C.) represents a temperature of the heat treatment; and t (min) represents a time for the heat-treatment.

2. The manufacturing method of an epitaxial silicon wafer according to claim 1, wherein an oxygen concentration of the silicon wafer before being subjected to the heat treatment is 810.sup.17 atoms/cm.sup.3 or more and 1810.sup.17 atoms/cm.sup.3 or less (according to ASTM F-121, 1979), and a film thickness of the epitaxial film is 0.5 m or more and 8.0 m or less.

Description

BRIEF DESCRIPTION OF DRAWING(S)

(1) FIG. 1 is a flow chart showing a manufacturing method of an epitaxial silicon wafer according to an exemplary embodiment of the invention.

(2) FIG. 2 is a cross section showing the epitaxial silicon water according to the above exemplary embodiment.

(3) FIG. 3 is a graph showing a depth profile of an oxygen concentration in an example (Experiment 1), in which a heat treatment is applied at 850 degrees C.

(4) FIG. 4 is a graph showing a depth profile of an oxygen concentration in an example (Experiment 2), in which a heat treatment is applied at 900 degrees C.

(5) FIG. 5 is a graph showing a depth profile of a depth differential of the oxygen concentration in the example (Experiment 1), in which the heat treatment is applied at 850 degrees C.

(6) FIG. 6 is a graph showing a depth profile of a depth differential of the oxygen concentration in the example (Experiment 2), in which the heat treatment is applied at 900 degrees C.

(7) FIG. 7 is a graph showing a depth profile of a boron concentration in Experiment

(8) FIG. 8 is a graph showing a result of a stress test in Experiment 4.

(9) FIG. 9 is a graph showing a result of a stress test in an example (Experiment 5), in which a silicon wafer having resistivity of 5 m.Math.cm was used.

(10) FIG. 10 is a graph showing a result of a stress test in an example (Experiment in which a silicon wafer having resistivity of 10 m.Math.cm was used.

(11) FIG. 11 is a graph showing a result of a stress test in an example (Experiment 5), in which a silicon wafer having resistivity of 100 m.Math.cm was used.

(12) FIG. 12 is a graph showing a result of a stress test in an example (Experiment 6), in which a silicon wafer having resistivity of 5 m.Math.cm was used.

(13) FIG. 13 is a graph showing a result of a stress test in an example (Experiment 6), in which a silicon wafer having resistivity of 10 m.Math.cm was used.

(14) FIG. 14 is a graph showing a result of a stress test in an example (Experiment 6), in which a silicon wafer having resistivity of 100 .Math.cm was used. FIG. 15 is a graph showing a result of a stress test in an example (Experiment 7), in which a silicon wafer having resistivity of 5 m.Math.cm was used.

(15) FIG. 16 is a graph showing a result of a stress test in an example (Experiment 7), in which a silicon wafer having resistivity of 10 m.Math.cm was used.

(16) 17 is a graph showing a result of a stress test in an example (Experiment 7), in which a silicon wafer having resistivity of 100 mcm was used.

(17) FIG. 18 is a graph showing a result of a stress test in Experiment 8.

(18) FIG. 19 is a graph showing a result of a stress test in Experiment 9.

DESCRIPTION OF EMBODIMENT(S)

(19) Exemplary embodiment(s) of the invention will be described below with reference to the attached drawings.

(20) FIG. 1 is a flow chart showing a manufacturing method of an epitaxial silicon wafer. FIG. 2 is a cross section showing the epitaxial silicon wafer.

(21) As shown in FIG. 1, in order to manufacture an epitaxial silicon wafer 1 shown in FIG. 2, a silicon wafer preparation step is initially performed (Step S1).

(22) The silicon wafer preparation step includes all the steps required for preparing a silicon wafer 2 with a mirror-polished surface 2 from a monocrystal ingot pulled up by CZ process, MCZ (Magnetic field applied Czochralski) process and the like, the steps including slicing, chamfering, grinding, lapping, etching, polishing and washing. At this time, the oxygen concentration of the silicon wafer 2 is preferably 810.sup.17 atoms/cm.sup.3 or more and 1810.sup.17 atoms/cm.sup.3 or less (according to ASTM F-121, 1979). When the oxygen concentration in the silicon wafer is within the above range, the oxygen concentration in the epitaxial film can be increased to the desired range in the below-described heat-treatment step.

(23) Boron is added to the silicon wafer 2 so that the resistivity of the silicon wafer 2 is adjusted to be 100 m.Math.cm or less, preferably in a range from 5 m.Math.cm to 100 m.Math.cm.

(24) Next, an epitaxial film formation step for forming an epitaxial film 3 on the silicon wafer 2 is performed (Step S2).

(25) The silicon water 2 is placed in a reaction container of an epitaxial apparatus (not shown) and the temperature inside the reaction container is raised from an ambient temperature to a target temperature. The target temperature is preferably set in a range from 1050 degrees C. to 1280 degrees C. When the temperature inside the reaction container reaches the above target temperature, the epitaxial film 3 is grown on the surface 21 of the silicon wafer 2. For instance, a growth gas such as trichlorosilane is introduced into the reaction container and the epitaxial film 3 is formed in the atmosphere of the growth gas. It should be noted that a necessary dopant such as boron and phosphorus may be added during the film formation.

(26) It is preferable that the epitaxial film formation step is performed until a film thickness T of the epitaxial film 3 becomes 0.5 m or more and 8.0 m or less. When the film thickness T of the epitaxial film 3 reaches the above range, the temperature of the epitaxial silicon wafer 1 is lowered from the temperature at which the epitaxial film 3 is grown to the ambient temperature.

(27) Subsequently, a heat-treatment step for thermally treating the epitaxial silicon wafer 1 is performed (Step S3). In the heat-treatment step, the heat treatment conditions are controlled so that the temperature becomes less than 900 degrees C.

(28) Further, it is preferable that the heat treatment time is controlled in a state that the temperature is within the above range.

(29) Specifically, the heat treatment time is controlled so that a formula (1) below is satisfied, where X (10.sup.17 atom/cm.sup.3) represents the oxygen concentration of the silicon wafer 2 before the epitaxial film formation step, Y (.Math.cm) represents the resistivity of the silicon wafer 2 before the epitaxial film formation step, Z (m) represents the film thickness of the epitaxial film 3, T (degrees C.) represents the heat treatment temperature, and t (min) represents the heat treatment time.
t3.7110.sup.56X.sup.7.03Y.sup.0.27Z.sup.3.34T.sup.16.7 (1)

(30) When the heat treatment time is equal to or more than t obtained in the formula (1), the epitaxial silicon wafer 1, in which an average oxygen concentration in the epitaxial film 3 is adjusted to be 1.710.sup.17 atoms/cm.sup.3 (according to ASTM F-121, 1979) or more and the dislocation extension is eliminated, can be manufactured.

(31) Advantage(s) of Exemplary Embodiment(s)

(32) The above-described exemplary embodiment offers the following of effects and advantages.

(33) (1) Since the low-resistivity silicon wafer whose resistivity is 100 m.Math.cm or less is used and the heat-treatment step is performed under a temperature of less than 900 degrees C., the enhanced diffusion effect of oxygen to the epitaxial film by boron can be caused. Thus, the average oxygen concentration in the epitaxial film can be sufficiently increased and thus an epitaxial silicon water capable of restraining the dislocation extension can be manufactured.

(34) (2) Since the heat-treatment step is performed at a temperature of less than 900 degrees C., the thermal diffusion of boron from the silicon wafer into the epitaxial film can be restrained.

(35) (3) With a simple process of substituting the oxygen concentration of the silicon wafer 2, the resistivity of the silicon wafer 2, the film thickness of the epitaxial film 3 and the heat treatment temperature in the above formula (1) to calculate the heat treatment time, the epitaxial silicon wafer 1 capable of restraining the dislocation extension can be manufactured.

(36) Modification(s)

(37) It should be noted that the scope of the invention is not limited to the above exemplary embodiment(s), but various improvements and design modifications are possible as long as such improvements and design modifications are compatible with an object of the invention.

(38) Specifically, without using the heat treatment time calculated based on the formula (1) in the heat-treatment step, the heat treatment conditions may be set within the temperature range of less than 900 degrees C. so that the epitaxial silicon wafer 1 whose average oxygen concentration of the epitaxial film 3 is adjusted to be 1.710.sup.17 atoms/cm.sup.3 (according to ASTM F-121, 1979) or more is producible, based on experiments performed under a plurality of conditions.

(39) Further, the oxygen concentration of the silicon wafer 2 may be less than 810.sup.17 atoms/cm.sup.3 or may exceed 1810.sup.17 atoms/cm.sup.3 (according to ASTM F-121, 1979),

EXAMPLES

(40) Next, the invention will be described in detail below with reference to Examples. However, it should be understood that the scope of the invention is by no means limited by the Examples.

Experiment 1

(41) A monocrystal ingot was manufactured using the CZ (Czochralski) process from silicon melt doped with boron. A silicon wafer was cut from the monocrystal ingot. The oxygen concentration of the silicon wafer (sometimes referred to as substrate oxygen concentration hereinafter) was 1110.sup.17 atoms/cm.sup.3. The resistivity of the silicon wafer (sometimes referred to as substrate resistivity hereinafter) was 5 m.Math.cm. Another silicon wafer with the substrate resistivity of 10 .Math.cm was prepared.

(42) Next, a (100) surface of the silicon wafer was processed to provide a mirror polished surface. Then, an epitaxial film of 3 m film thickness (sometimes referred to as epitaxial film thickness hereinafter) was grown on the mirror polished surface. The epitaxial film was grown under a gas atmosphere (e.g. trichlorosilane) at a temperature of approximately 1150 degrees C.

(43) Then, the heat-treatment step was performed on the wafer on which the epitaxial film had been grown, in which the wafer was held for 60 minutes at 850 degrees C. in a non-acidic atmosphere to obtain an epitaxial silicon wafer.

(44) Another epitaxial silicon wafer without being subjected to the heat treatment was also prepared.

(45) Stress tests were applied on the manufactured epitaxial silicon wafers. Initially, a measurement sample pieces (length: 3 cm, width: 1.5 cm) were cut from each of the epitaxial silicon wafers. Next, a 5 g load (measured by a micro Vickers hardness tester) was applied on a surface of each of the measurement samples (surface of the epitaxial film) and held for 10 seconds to provide indentations of 3 m. depth. Then, the measurement samples are subjected to a three-point bending test (distance between support points: 2 cm, test temperature: 800 degrees C.). At this time, a 5N load was applied so that a tensile force acted on the surface of each of the measurement samples.

(46) Subsequently, a 2 m Wright-etching was applied on each of the measurement samples having been cooled to the ambient temperature to check the presence of dislocation pits generated from the indentations provided on the epitaxial film and observed on the surface of the epitaxial film using an optical microscope. The measurement results are shown in Table 1.

(47) Further, depth profiles of the oxygen concentration for the manufactured epitaxial silicon wafers subjected to the heat treatment were measured. The oxygen concentration was measured using an SIMS (secondary ion mass spectroscope). The depth profiles are shown in FIG. 3.

(48) TABLE-US-00001 TABLE 1 Epitaxial Film Substrate Oxygen Heat-Treatment Dislo- Film Concen- Step cation Thick- tration Temper- Extension ness (10.sup.17 ature Time No: (m) atoms/cm.sup.3) Resistivity ( C.) (Min.) Yes: x 3 11 5 m .Math. cm N/A N/A x 3 11 5 m .Math. cm 850 60 3 11 10 .Math. cm N/A N/A x 3 11 10 .Math. cm 850 60 x

(49) As shown in Table 1, it is found that the dislocation extension from the indentations did not occur and the strength of the epitaxial film was increased when the heat treatment was applied after forming the epitaxial film and the substrate resistivity was 5 m.Math.cm. On the other hand, when the substrate resistivity was 10 .Math.cm, it is found that the dislocation extension from the indentations was observed and a strength of the epitaxial film was low even after the heat-treatment step at the same heat treatment temperature.

(50) It should be noted that all of the wafers not having been subjected to the heat-treatment step exhibited the dislocation extension at both of the resistivity levels, and it is found that the strength of the epitaxial film was low.

(51) As shown in FIG. 3, when the two instances are compared, it is found that, though the oxygen concentration of the substrate of an instance in which the substrate resistivity was 5 m.Math.cm shows a decreasing profile, the profile also shows that the oxygen concentration in the epitaxial film locally increased.

Experiment 2

(52) The epitaxial silicon wafer was manufactured under the same conditions as those in Experiment 1 except that the heat treatment temperature was changed to 900 degrees C. The depth profile of the oxygen concentration was measured for the manufactured epitaxial silicon wafers that were subjected to the heat-treatment step. The depth profiles are shown in FIG. 4.

(53) TABLE-US-00002 TABLE 2 Epitaxial Film Substrate Oxygen Heat-Treatment Film Concen- Step Thick- tration Temper- ness (10.sup.17 ature Time (m) atoms/cm.sup.3) Resistivity ( C.) (Min.) 3 11 5 m .Math. cm N/A N/A 3 11 5 m .Math. cm 900 60 3 11 10 .Math. cm N/A N/A 3 11 10 .Math. cm 900 60

(54) As shown in FIG. 4, when the heat treatment temperature was 900 degrees C., oxygen-concentration profile of the epitaxial film on a low-resistivity substrate did not show the locally increasing oxygen concentration profile.

(55) Next, the depth profiles of depth differentials (atoms/c of the oxygen concentration for the manufactured epitaxial silicon wafers subjected to the heat treatment in the Experiments 1 and 2 were measured. The depth profiles are shown in FIGS. 5 and 6. As shown in FIG. 5, in the Experiment 1 in which the heat treatment at 850 degrees C. was applied, a local peak of the depth profile of the depth differential of the oxygen concentration was observed in the vicinity of an interface in an instance in which a silicon wafer of the substrate resistivity of 5 m.Math.cm was used. The local peak of the depth profile of the depth differential of the oxygen concentration shows that the oxygen concentration locally increases in the vicinity of the interface, and it is speculated that Table 1 shows no dislocation extension was caused from the indentations and thus the strength of the epitaxial film was high.

(56) On the other hand, both of the instance shown in FIG. 5 in which a silicon wafer of the substrate resistivity of 10 .Math.cm was used and the instance shown in FIG. 6 in which the heat treatment at 900 degrees C. was applied in Experiment 2 show broad peaks,

Experiment 3

(57) An epitaxial silicon wafer was prepared under the same conditions as those in Experiment 1 except that the heat treatment temperature was changed to 1000 degrees C. Further, an epitaxial silicon wafer was prepared at a heat treatment temperature of 850 degrees C. in the same manner as in Experiment 1. The depth profile of boron concentration for the prepared epitaxial silicon wafer was measured. The depth profile of the boron concentration was measured using an SIMS (secondary ion mass spectroscope). The depth profiles are shown in FIG. 7.

(58) As shown in FIG. 7, while the wafer subjected to the heat treatment at 850 degrees C. exhibited small boron diffusion toward the epitaxial film, the wafer subjected to the heat treatment at 1000 degrees C. exhibited large boron diffusion toward the epitaxial film.

Experiment 4

(59) Except that the epitaxial film thickness, the substrate oxygen concentration and the substrate resistivity were set as in Table 3 below, the heat treatment temperature was 890 degrees C. and the heat treatment time was variously changed, the epitaxial silicon wafers were prepared and the stress test was performed under the same conditions as those in Experiment 1, and the dislocation pits observable on the surface of the epitaxial film were measured. The measurement results are shown in Table 3. The results of the stress tests are shown in FIG. 8. It should be noted that the curve in FIG. 8 is an approximate curve showing a border of the presence/absence of the dislocation extension derived from the above formula (1).

(60) TABLE-US-00003 TABLE 3 Epitaxial Film Substrate Oxygen Heat-Treatment Dislo- Film Concen- Step cation Thick- tration Temper- Extension ness (10.sup.17 Resistivity ature Time No: (m) atoms/cm.sup.3) (m .Math. cm) ( C.) (Min.) Yes: x 3 8 5 890 30 x 3 8 5 890 150 3 9 5 890 30 x 3 9 5 890 90 3 9 5 890 150 3 10 5 890 90 3 11 5 890 30

(61) As shown in Table 3, it was found that the dislocation extension from the indentations did not occur and the strength of the epitaxial film was improved when the heat treatment temperature was 890 degrees.

(62) Further, as is clear from FIG. 8, the border between the presence and absence of the dislocation extension is at the approximate curve, and it is observable that the dislocation extension tends to occur when the heat treatment time is shorter than the time indicated by the approximate curve.

Experiment 5

(63) Except that the epitaxial film thickness, the substrate oxygen concentration and the substrate resistivity were set as in Table 4 below, the heat treatment temperature was 850 degrees C. and the heat treatment time was variously changed, the epitaxial silicon wafers were prepared and the stress test was performed under the same conditions as those in Experiment 1, and the dislocation pits observable on the surface of the epitaxial film were measured.

(64) In an instance using a silicon water whose substrate resistivity was 5 m.Math.cm, an average oxygen concentration of the epitaxial film was measured. The measurement results are shown in Table 4. The results of the stress tests are shown in FIGS. 9 to 11. It should be noted that the curves in FIGS. 9-11 are approximate curves derived from the above formula (1).

(65) TABLE-US-00004 TABLE 4 Epitaxial Film Substrate Dislocation Average Oxygen Film Oxygen Heat-Treatment Step Extension Concentration of Thickness Concentration Resistivity Temperature Time No: Epitaxial Film (m) (10.sup.17 atoms/cm.sup.3) (m .Math. cm) ( C.) (Min.) Yes: x (10.sup.17 atoms/cm.sup.3) 3 8 5 850 90 x 1.48 3 8 5 850 150 x 1.68 3 8 5 850 240 1.93 3 9 5 850 30 x 1.40 3 9 5 850 90 x 1.67 3 9 5 850 120 1.79 3 10 5 850 30 x 1.57 3 10 5 850 60 1.72 3 10 5 850 90 1.87 3 11 5 850 30 1.73 3 8 10 850 180 x 1.51 3 9 10 850 90 x 1.59 3 9 10 850 120 1.71 3 10 10 850 30 x 1.66 3 10 10 850 60 x 1.67 3 10 10 850 90 1.72 3 11 10 850 30 x 1.69 3 11 10 850 60 1.82 3 12 10 850 30 1.82 3 8 100 850 210 x 1.38 3 8 100 850 270 x 1.52 3 9 100 850 60 x 1.55 3 9 100 850 180 x 1.58 3 9 100 850 210 1.70 3 10 100 850 60 x 1.67 3 10 100 850 90 x 1.68 3 10 100 850 120 1.71 3 11 100 850 30 x w/o 3 11 100 850 60 x w/o 3 11 100 850 120 w/o 3 12 100 850 30 x w/o 3 12 100 850 60 w/o 3 13 100 850 30 w/o

Experiment 6

(66) Except that the epitaxial film thickness, the substrate oxygen concentration and the substrate resistivity were set as in Table 5 below, the heat treatment temperature was 800 degrees C. and the heat treatment time was variously changed, the epitaxial silicon wafers were prepared and the stress test was performed under the same conditions as those in Experiment 1, and the dislocation pits observable on the surface of the epitaxial film were measured. The measurement results are shown in Table 5. The results of the stress tests are shown in FIGS. 12 to 14. It should be noted that the curves in FIGS. 12-14 are approximate curves derived from the above formula (1).

(67) TABLE-US-00005 TABLE 5 Epitaxial Film Substrate Oxygen Heat-Treatment Dislo- Film Concen- Step cation Thick- tration Temper- Extension ness (10.sup.17 Resistivity ature Time No: (m) atoms/cm.sup.3) (m .Math. cm) ( C.) (Min.) Yes: x 3 9 5 800 300 x 3 10 5 800 90 x 3 10 5 800 210 3 10 5 800 300 3 11 5 800 60 x 3 11 5 800 150 3 12 5 800 30 x 3 12 5 800 60 3 13 5 800 30 3 13 5 800 60 3 10 10 800 120 x 3 10 10 800 270 3 10 10 800 300 3 11 10 800 60 x 3 11 10 800 150 3 11 10 800 180 3 11 10 800 300 3 12 10 800 30 x 3 12 10 800 90 3 13 10 800 30 3 13 10 800 60 3 11 100 800 90 x 3 11 100 800 120 x 3 12 100 800 60 x 3 12 100 800 210 3 12 100 800 300 3 13 100 800 30 x 3 13 100 800 120 3 14 100 800 30 3 14 100 800 60

Experiment 7

(68) Except that the epitaxial film thickness, the substrate oxygen concentration and the substrate resistivity were set as in Table 6 below, the heat treatment temperature was 750 degrees C. and the heat treatment time was variously changed, the epitaxial silicon wafers were prepared and the stress test was performed under the same conditions as those in Experiment 1, and the dislocation pits observable on the surface of the epitaxial film were measured. The measurement results are shown in Table 6. The results of the stress tests are shown in FIGS. 15 to 17, It should be noted that the curves in FIGS. 15-17 are approximate curves derived from the above formula (1).

(69) TABLE-US-00006 TABLE 6 Epitaxial Film Substrate Oxygen Heat-Treatment Dislo- Film Concen- Step cation Thick- tration Temper- Extension ness (10.sup.17 Resistivity ature Time No: (m) atoms/cm.sup.3) (m .Math. cm) ( C.) (Min.) Yes: x 3 11 5 750 180 x 3 11 5 750 210 x 3 11 5 750 330 3 12 5 750 30 x 3 12 5 750 90 x 3 12 5 750 180 3 12 5 750 240 3 13 5 750 30 x 3 13 5 750 60 3 13 5 750 90 3 14 5 750 30 3 12 10 750 60 x 3 12 10 750 120 x 3 12 10 750 270 3 13 10 750 60 x 3 13 10 750 120 3 13 10 750 210 3 14 10 750 30 3 14 10 750 90 3 12 100 750 120 x 3 13 100 750 120 x 3 13 100 750 180 3 13 100 750 300 3 14 100 750 30 x 3 14 100 750 90 3 14 100 750 180 3 14 100 750 300 3 15 100 750 30

Experiment 8

(70) Except that the epitaxial film thickness was 2 m, and the substrate oxygen concentration, the substrate resistivity, the heat treatment temperature and the heat treatment time were set as shown in Table 7 below, epitaxial silicon wafers were manufactured under the same conditions as those in Experiment 1. Further, except that the load applied by a micro Vickers hardness tester was 3 g and the depth of the indentations was 2 m, the stress test was performed under the some conditions as those in Example 1 and the dislocation pits observed on the surface of the epitaxial film were measured, The measurement results are shown in Table 7. The results of the stress tests are shown in FIG. 18. It should be noted that the curve in FIG. 18 is an approximate curve derived from the above formula (1).

(71) TABLE-US-00007 TABLE 7 Epitaxial Film Substrate Oxygen Heat-Treatment Dislo- Film Concen- Step cation Thick- tration Temper- Extension ness (10.sup.17 Resistivity ature Time No: (m) atoms/cm.sup.3) (m .Math. cm) ( C.) (Min.) Yes: x 2 8 5 850 30 x 2 8 5 850 60 2 8 5 850 60 2 8 5 850 90 2 8 5 850 120 2 9 5 850 30 2 10 5 850 30

Experiment 9

(72) Except that the epitaxial film thickness was 4 m, and the substrate oxygen concentration, the substrate resistivity, the heat treatment temperature and the heat treatment time were set as shown in Table 8 below, epitaxial silicon wafers were manufactured under the same conditions as those in Experiment 1. Further, except that the load applied b the micro Vickers hardness tester was 7 g and the depth of the indentations was 4 m, the stress test was performed under the same conditions as those in Example 1 and the dislocation pits observed on the surface of the epitaxial film were measured. The measurement results are shown in Table 8. The results of the stress tests are shown in FIG. 19. It should be noted that the curve in FIG. 19 is an approximate curve derived from the above formula (1).

(73) TABLE-US-00008 TABLE 8 Epitaxial Film Substrate Oxygen Heat-Treatment Dislo- Film Concen- Step cation Thick- tration Temper- Extension ness (10.sup.17 Resistivity ature Time No: (m) atoms/cm.sup.3) (m .Math. cm) ( C.) (Min.) Yes: x 4 8 5 850 180 x 4 8 5 850 240 x 4 9 5 850 150 x 4 9 5 850 210 4 10 5 850 90 x 4 10 5 850 120 4 11 5 850 60 x 4 11 5 850 90 4 12 5 850 30 x 4 12 5 850 60 4 13 5 850 30

(74) As is clear from Tables 4-8 and FIGS. 9-19, the border between the presence and absence of the dislocation extension is at the approximate curves, and it is observable that the dislocation extension tends to occur when the heat treatment time is shorter than the time indicated by the approximate curve.

(75) Further, from the relationship shown in Table 4 between the presence/absence of the dislocation extension and the average oxygen concentration of the epitaxial film, it is deducible that the dislocation extension can be restrained when the average oxygen concentration of the epitaxial film is 1.710.sup.17 atoms/cm.sup.3 or more.

Experiment 10

(76) The epitaxial silicon wafers manufactured in Experiments 5 to 9 were subjected to heat treatments simulating manufacture processes of semiconductor devices (kept at 1000 degrees C. for an hour, kept at 800 degrees C. for two hours, kept at 650 degrees C. for three hours, and kept at 900 degrees C. for an hour). The atmosphere for the heat treatments was a mixture of N.sub.2 and O.sub.2 (containing 3 mass % of O.sub.2).

(77) The wafers showing no dislocation extension after the stress test in the above Experiments 5 to 9 also showed no dislocation extension in a strength test after the device heat treatment in Experiment 10.