QUARTZ GLASS CRUCIBLE AND MANUFACTURING METHOD THEREOF
20230010489 · 2023-01-12
Inventors
- Masami OHARA (Akita-shi, Akita, JP)
- Hiroshi KISHI (Akita-shi, Akita, JP)
- Eriko KITAHARA (Akita-shi, Akita, JP)
Cpc classification
C30B15/10
CHEMISTRY; METALLURGY
C03B19/095
CHEMISTRY; METALLURGY
C03B2201/54
CHEMISTRY; METALLURGY
C03C2201/54
CHEMISTRY; METALLURGY
C03B20/00
CHEMISTRY; METALLURGY
C03C3/06
CHEMISTRY; METALLURGY
International classification
Abstract
A quartz glass crucible (1) has a structure wherein a peak of a distribution of a total concentration of Na, K, and Ca in a depth direction from an inner surface (10i) of the crucible is present at a position deeper than the inner surface (10i). In an exemplary embodiment, the quartz glass crucible is capable of improving the yield of a silicon single crystal by suppressing peeling-off of brown rings.
Claims
1. A quartz glass crucible characterized in that a peak of a distribution of a total concentration of Na, K, and Ca in a depth direction from an inner surface of the crucible is present at a position deeper than the inner surface.
2. The quartz glass crucible according to claim 1, wherein the peak of the total concentration of Na, K, and Ca is present within a depth range of 32 μm or less from the inner surface.
3. The quartz glass crucible according to claim 2, wherein the peak of the total concentration of Na, K, and Ca is present within a depth range of 16 μm or more and 32 μm or less from the inner surface.
4. The quartz glass crucible according to claim 3, wherein the peak value of the total concentration of Na, K, and Ca within the range of 16 μm or more and 32 μm or less from the inner surface is 2 times or more and 19 times or less an average value of the total concentration of Na, K, and Ca within a depth range of 0 μm or more and 8 μm or less from the inner surface.
5. The quartz glass crucible according to claim 3, wherein an average value of the total concentration of Na, K, and Ca within a depth range of 32 μm or more and 1000 μm or less from the inner surface is 0.6 times or more and 1 time or less the average value of the total concentration of Na, K, and Ca within a depth range of 0 μm or more and 8 μm or less from the inner surface.
6. The quartz glass crucible according to claim 5, wherein a total concentration of Na, K, and Ca within a depth range of 32 μm or more and 1000 μm or less from the inner surface has a negative concentration gradient with the depth direction as a positive direction.
7. The quartz glass crucible according to claim 3, wherein an average value of a total concentration of Li, Al, Na, K, and Ca within a depth range of 0 μm or more and 8 μm or less from the crucible inner surface is 3.6×10.sup.16 atoms/cc or more and 5.5×10.sup.17 atoms/cc or less.
8. The quartz glass crucible according to claim 1 comprising: a transparent layer made of silica glass containing no bubbles and constituting the inner surface; and a bubble layer made of silica glass containing a large number of bubbles and provided outside the transparent layer, wherein a thickness of the transparent layer is 1 mm or more.
9. A manufacturing method of a quartz glass crucible comprising: producing a quartz glass crucible by arc-melting raw material quartz powder deposited on an inner surface of a rotating mold; washing an inner surface of the quartz glass crucible with pure water thereby reducing a total concentration of Na, K, and Ca contained in a silica glass around the inner surface as compared to that before washing; and etching the inner surface with washing liquid containing hydrofluoric acid.
10. The manufacturing method of a quartz glass crucible according to claim 9, wherein a specific resistance of the pure water used in the washing step of the inner surface of the quartz glass crucible with the pure water is 17 MΩ cm or more, the amount of water to be used per quartz glass crucible is 125 liters or more, and the water temperature is 45 to 99° C.
11. The manufacturing method of a quartz glass crucible according to claim 9, wherein the degree of etching at the inner surface is 5 μm or more and 10 μm or less and, whereby a peak of the total concentration of Na, K, and Ca is set within a depth range of 16 μm or more and 32 μm or less from the inner surface.
12. The quartz glass crucible according to claim 2 comprising: a transparent layer made of silica glass containing no bubbles and constituting the inner surface; and a bubble layer made of silica glass containing a large number of bubbles and provided outside the transparent layer, wherein a thickness of the transparent layer is 1 mm or more.
13. The quartz glass crucible according to claim 3 comprising: a transparent layer made of silica glass containing no bubbles and constituting the inner surface; and a bubble layer made of silica glass containing a large number of bubbles and provided outside the transparent layer, wherein a thickness of the transparent layer is 1 mm or more.
14. The quartz glass crucible according to claim 4, wherein an average value of the total concentration of Na, K, and Ca within a depth range of 32 μm or more and 1000 μm or less from the inner surface is 0.6 times or more and 1 time or less the average value of the total concentration of Na, K, and Ca within a depth range of 0 μm or more and 8 μm or less from the inner surface.
15. The quartz glass crucible according to claim 4, wherein an average value of a total concentration of Li, Al, Na, K, and Ca within a depth range of 0 μm or more and 8 μm or less from the crucible inner surface is 3.6×10.sup.16 atoms/cc or more and 5.5×10.sup.17 atoms/cc or less.
16. The quartz glass crucible according to claim 4 comprising: a transparent layer made of silica glass containing no bubbles and constituting the inner surface; and a bubble layer made of silica glass containing a large number of bubbles and provided outside the transparent layer, wherein a thickness of the transparent layer is 1 mm or more.
17. The quartz glass crucible according to claim 5, wherein an average value of a total concentration of Li, Al, Na, K, and Ca within a depth range of 0 μm or more and 8 μm or less from the crucible inner surface is 3.6×10.sup.16 atoms/cc or more and 5.5×10.sup.17 atoms/cc or less.
18. The quartz glass crucible according to claim 5 comprising: a transparent layer made of silica glass containing no bubbles and constituting the inner surface; and a bubble layer made of silica glass containing a large number of bubbles and provided outside the transparent layer, wherein a thickness of the transparent layer is 1 mm or more.
19. The quartz glass crucible according to claim 6, wherein an average value of a total concentration of Li, Al, Na, K, and Ca within a depth range of 0 μm or more and 8 μm or less from the crucible inner surface is 3.6×10.sup.16 atoms/cc or more and 5.5×10.sup.17 atoms/cc or less.
20. The quartz glass crucible according to claim 6 comprising: a transparent layer made of silica glass containing no bubbles and constituting the inner surface; and a bubble layer made of silica glass containing a large number of bubbles and provided outside the transparent layer, wherein a thickness of the transparent layer is 1 mm or more.
Description
BRIEF DESCRIPTION OF THE DRAWINGS
[0030]
[0031]
[0032]
[0033]
[0034]
[0035]
[0036]
[0037]
[0038]
MODE FOR CARRYING OUT THE INVENTION
[0039] Hereinafter, a preferred embodiment of the present invention will be described in detail with reference to the accompanying drawings.
[0040]
[0041] As illustrated in
[0042] The diameter of the quartz glass crucible 1 is preferably 22 inches (about 560 mm) or more and, more particularly, 32 inches (about 800 mm) or more, although it varies depending on the diameter of a silicon single crystal ingot to be pulled up. Such a large-diameter crucible is preferably used for pulling up a large-sized silicon single crystal ingot having a diameter of 300 mm or more and has no adverse effects on the quality of the single crystal even with long-term usage. While the thickness of the quartz glass crucible 1 slightly varies from one portion to another, the thickness of the side wall portion 10a of a crucible having a diameter of 22 inches or more is preferably 7 mm or more, and the thickness of the side wall portion 10a of a large-sized crucible having a diameter of 32 inches or more is preferably 10 mm or more. This allows a large amount of silicon melt to be stably held under high temperatures.
[0043] The quartz glass crucible 1 has a double-layer structure and includes a transparent layer 11 made of silica glass containing no bubbles and a bubble layer 12 (opaque layer) made of silica glass containing a large number of micro-bubbles and provided outside the transparent layer 11.
[0044] The transparent layer 11 is a layer constituting an inner surface 10i of the crucible that contacts a silicon melt and is provided for preventing the single crystal yield from being reduced due to the bubbles in the silica glass. The thickness of the transparent layer 11 is preferably 1 to 12 mm and is set to an adequate value in this range for each location in the crucible so as to prevent the transparent layer 11 from being completely eliminated due to erosion during a single crystal pulling-up process and the bubble layer 12 from being exposed. Like the bubble layer 12, the transparent layer 11 is preferably formed over the entire surface of the crucible from the side wall portion 10a to the bottom portion 10b; however, it may be omitted at the upper end portion (rim portion) of the crucible that does not contact the silicon melt.
[0045] The transparent layer 11 constitutes the inner side of the crucible and has a bubble content of 0.1 vol % or less. The phrase “containing no bubbles” in regard to the transparent layer 11 means that the transparent layer 11 has a bubble content and bubble size to the extent that does not reduce the single crystal yield. When bubbles exist in the vicinity of the inner surface of the crucible, the bubbles in the vicinity of the crucible inner surface cannot be confined within the silica glass due to erosion of the crucible inner surface. This may cause the bubbles in the silica glass to burst due to thermal expansion during a crystal pulling-up process, resulting in peel-off of crucible pieces (quartz pieces). When the crucible pieces released into the silicon melt are transported by melt convection to the growth interface of a single crystal to be entrapped into the single crystal, dislocations may be generated in the single crystal. Alternatively, when the bubbles released into the melt due to the erosion of the crucible inner surface float to the solid/liquid interface to be entrapped in the single crystal, pinholes may be generated. The average diameter of the bubbles in the transparent layer 11 is preferably 100 μm or less.
[0046] The bubble layer 12 is a layer constituting an outer surface 10oof the crucible and is provided for enhancing the heat-retaining property of the silicon melt in the crucible and for dispersing radiant heat from a heater which is provided in a single crystal pull-up unit so as to surround the crucible, so that the silicon melt in the crucible can be heated as uniformly as possible. To this end, the bubble layer 12 is formed over the entire surface of the crucible from the side wall portion 10a to the bottom portion 10b. The thickness of the bubble layer 12 is, although it varies from one portion to another in the crucible, equal to a value obtained by subtracting the thickness of the transparent layer 11 from the thickness of the crucible.
[0047] The bubble content of the bubble layer 12 is higher than that of the transparent layer 11 and is preferably more than 0.1 vol % and 5 vol % or less, and more preferably 1 vol % or more and 4 vol % or less. When the bubble content of the bubble layer 12 is 0.1 vol % or less, the bubble layer 12 cannot exhibit the required heat-retaining function. In addition, when the bubble content of the bubble layer 12 exceeds 5 vol %, the crucible could be deformed due to expansion of the bubbles, which may reduce the single crystal yield, and heat conductivity would become insufficient. In particular, when the bubble content of the bubble layer 12 falls within the range of 1 to 4%, a good balance is maintained between heat retaining property and heat conductivity. The bubble content of the bubble layer 12 can be calculated by measuring the specific gravity of, for example, an opaque silica glass piece cut out from the crucible.
[0048] To prevent contamination of the silicon melt, the silica glass constituting the transparent layer 11 preferably has high purity. Therefore, the quartz glass crucible 1 according to the present embodiment is preferably constituted by two layers of a synthetic quartz glass layer formed from synthetic quartz powder and a natural quartz glass layer formed from natural quartz powder. The synthetic quartz powder can be produced by means of a vapor phase oxidation (dry synthesis method) of silicon tetrachloride (SiCl.sub.4), or hydrolysis of silicon alkoxide (Sol-Gel method). The natural quartz powder is quartz powder produced by pulverizing natural mineral mainly composed of α-quartz.
[0049] Although details will be described later, the two-layer structure of the synthetic quartz glass layer and natural quartz glass layer can be produced by depositing the natural quartz powder along the inner surface of a mold for producing a crucible, depositing the synthetic quartz powder on the deposited natural quartz powder, and melting the thus deposited quartz powder layers by using Joule heat of arc discharge. In the initial stage of the arc melting, bubbles are removed by performing strong evacuation from outside of the deposited layer of the quartz powder to form the transparent layer 11. After that, the evacuation is stopped or weakened to form the bubble layer 12 outside the transparent layer 11. Therefore, the boundary between the synthetic quartz glass layer layer and the natural quartz glass layer does not necessarily coincide with the boundary between the transparent layer 11 and the bubble layer 12; however, like the transparent layer 11, the synthetic quartz glass layer preferably has such a thickness as not to be completely eliminated due to erosion of the crucible inner surface during a crystal pulling-up process.
[0050]
[0051] As illustrated in
[0052] The first surface layer portion Z.sub.1 at the depth position of 0 to 16 μm from the inner surface 10i of the crucible is a layer that first contacts the silicon melt. In the first half (I) of a raw material melting step in which the first surface layer portion Z.sub.1 contacts the silicon melt, many cristobalite cores are generated on the inner surface 10i of the crucible. Metal impurities such as Li, Al, Na, K, and Ca, which abide in the vicinity of the crucible inner surface contribute to the generation of brown ring cores and may cause more brown rings to be generated. Thus, the average value of the total concentration of Li, Al, Na, K, and Ca within a depth range of 0 to 8 μm from the inner surface 10i of the crucible is preferably 3.6×10.sup.16 atoms/cc or more and 5.5×10.sup.17 atoms/cc or less. This can reduce the number of generated brown ring cores.
[0053] The inner surface 10i having a reduced concentration of impurities such as Na needs to be formed particularly at the bottom portion 10b and/or corner portion 10c of the quartz glass crucible 1. This is because the bottom portion 10b and corner portion 10c of the quartz glass crucible 1 contact the silicon melt for a longer period of time than do the side wall portion 10a and are thus likely to generate the brown ring. The side wall portion 10a may have or may not have the inner surface 10i having a reduced concentration of impurities such as Na.
[0054] The inner surface 10i of the crucible is preferably as smooth as possible. In particular, the inner surface 10i of the bottom portion 10b preferably has an arithmetic average roughness Ra of 0.02 to 0.3 μm. This can reduce the number of generated brown ring cores in the first half (I) of the raw material melting step.
[0055] The number of generated brown ring cores abruptly decreases after peaking at some point and, after that, a growth stage of brown ring cores is entered. Thus, afterwards, the number of generated brown ring cores does not significantly increase even when the total concentration of Na, K, and Ca is somewhat high. In the latter half (II) of the raw material melting step, the core gradually grows to generate brown rings. However, when the peak value of the total concentration of Na, K, and Ca within a depth range of 16 to 32 μm from the inner surface 10i is set to 2 to 19 times the average value (reference concentration) of the total concentration of Na, K, and Ca within a depth range of 0 to 8 μm from the inner surface 10i, brown ring cores can be eliminated by making the dissolution rate of the inner surface 10i higher than the growth rate of brown ring cores.
[0056] It is known that Na, K, and Ca contained in the silica glass promote dissolution of the silica glass. In the present embodiment, the total concentration of Na, K, and Ca within a depth range of 16 to 32 μm from the inner surface 10i is made relatively high, so that the dissolution rate of the inner surface 10i can be made higher than the growth rate of brown rings. This allows brown ring cores to be eliminated before they grow larger, whereby the brown ring core can be removed.
[0057] As described above, by making the total concentration of Na, K, and Ca in the first surface layer portion Z.sub.1 relatively low and making the total concentration of Na, K, and Ca in the second surface layer portion Z.sub.2 relatively high, the number of brown rings can be reduced to some degree. However, it is difficult to completely eliminate brown rings, and some brown rings are generated on the inner surface 10i of the crucible. During a silicon single crystal pulling-up process (III), brown rings grow larger, increasing a risk of peeling-off of the brown rings. In the present embodiment, the average value of the total concentration of Na, K, and Ca within a depth range (32 to 1000 μm) of deeper than 32 μm from the inner surface 10i is 0.6 to 1 times the reference concentration, so that dissolution of the inner surface 10i during the single crystal pulling-up process can be suppressed. Further, the total concentration of Na, K, and Ca within a depth range of 32 to 1000 μm from the inner surface 10i of the crucible has a negative concentration gradient of 8.2×10.sup.10 atoms/cc/μm or less with the depth direction as the positive direction, so that it is possible to stably grow brown rings while suppressing an abrupt change in the total concentration of Na, K, and Ca, whereby peeling-off of brown rings can be suppressed.
[0058]
[0059] In the manufacture of the quartz glass crucible 1, the quartz glass crucible 1 is first produced by a rotational molding method (step S11) as illustrated in
[0060] Then, arc electrodes 15 are placed in the mold 14, and the deposited layer 16 of the raw material quartz powder is arc-melted from the inner surface 14i side of the mold 14. Specific conditions such as heating time and heating temperature need to be determined in consideration of conditions including raw material and size of the crucible. At this time, the deposited layer 16 of the raw material quartz powder is subjected to suction through many vent holes 14a formed in the inner surface 14i of the mold 14 to control the amount of bubbles in melted glass. Specifically, suction force through the many vent holes 14a formed in the inner surface 14i of the mold 14 is increased at the start time of the arc melting to form the transparent layer 11, and then the suction force is reduced after formation of the transparent layer 11 to form the bubble layer 12.
[0061] The arc heat is gradually transmitted outward from the inner side of the deposited layer 16 of the raw material quartz powder to melt the raw material quartz powder, so that by changing decompression conditions at the time at which the raw material quartz powder starts to be melted, the transparent layer 11 and the bubble layer 12 can be separately formed. By conducting decompression melting for increasing decompression at the time at which the silica powder melts, the arc atmosphere gas is not confined in the glass, and silica glass containing no bubbles is formed. In addition, when normal melting (atmospheric pressure melting) is performed to weaken decompression at the time at which the raw material quartz powder is melted, the arc atmosphere gas is confined in the glass, and silica glass containing a large number of bubbles is formed. By changing, for example, the arrangement of the arc electrodes 15 or current applied thereto during the decompression melting or normal melting to partly change the degree of melting, the thickness of the transparent layer 11 or bubble layer 12 can be adjusted at each location.
[0062] Thereafter, the arc heating is ended, and the crucible is cooled down, whereby the quartz glass crucible 1 in which the transparent layer 11 and bubble layer 12 are sequentially provided from the inside toward the outside of the crucible wall is completed. In the thus obtained quartz glass crucible 1 after the arc melting (before washing), impurity concentration distribution in the depth direction from the inner surface 10i is as illustrated in
[0063] Then, the inner surface 10i of the quartz glass crucible 1 is washed with pure water (step S12). The specific resistance of the pure water used at this time is preferably 17 MΩ cm or more, and the flow rate thereof is preferably 50 to 60 L/min. The amount of water to be used per one quartz glass crucible 1 is preferably 125 liter or more (125 liter per crucible), and the water temperature is preferably 45 to 99° C. Under these conditions, Li, Na, K, and Ca, which are easily dissolved into pure water, are eluted from the vicinity of the inner surface 10i of the crucible to reduce the impurity concentration in the silica glass as compared to that before washing. As a result, the impurity concentration distribution in the depth direction from the inner surface 10i of the quartz glass crucible 1 becomes as illustrated in
[0064] The water temperature of the pure water used in the pure water washing is preferably 45 to 99° C. and, particularly preferably 55 to 65° C. in consideration of ease of handling and safety. When the temperature of the pure water is about 25 to 35° C., the effect of reducing the total concentration of Li, Na, K, and Ca on the inner surface 10i of the crucible cannot be obtained. When the pure water is intentionally heated to 45° C. or more, it is possible to elute Li, Na, K, and Ca from the inner surface 10i of the crucible to thereby reduce the concentration of impurities in the silica glass. Thus, by using high-temperature pure water, impurities in the crucible inner surface can be dissolved into the pure water to be washed out, thus allowing the peak of the impurity concentration to be formed in an area slightly deeper than the outermost surface.
[0065] Then, the inner surface 10i of the quartz glass crucible 1 is etched using washing liquid containing hydrofluoric acid to remove the surface layer portion of the crucible inner surface 10i (step S13). The degree of etching at the inner surface at this time is preferably 5 to 10 μm. With this etching, the inner surface 10i of the crucible can be cleaned, and the impurity concentration distribution in the depth direction from the inner surface 10i of the quartz glass crucible 1 is as illustrated in
[0066] Finally, the quartz glass crucible 1 is entirely finish-washed with pure water (step 14). In this finish washing, it is necessary to carry out pure water washing under the condition that the peak position of the total concentration of Na, K, and Ca does not significantly change. This involves a reduction in the washing time and water temperature to suppress elution of Na, K, and Ca. Through the above-described steps, the quartz glass crucible 1 according to the present embodiment is completed.
[0067] Typically, in the quartz glass crucible 1 produced by arc melting, impurity concentration is highest on the inner surface 10i due to the surface condensation effect. Thus, by the action of the impurities, brown ring cores are likely to be generated at the crucible inner surface, and the generated brown ring cores are likely to grow. However, in the present embodiment, the peak position of the total concentration of Na, K, and Ca contributing to generation of brown ring cores and dissolution of the crucible is shifted to a deeper position than the crucible inner surface, so that the number of generated brown ring cores can be reduced, and the generated brown ring cores can be removed. Thus, a probability of peeling-off of the brown ring during the silicon single crystal pulling-up process can be reduced to improve the single crystal yield.
[0068]
[0069] As illustrated in
[0070] In the first half of the raw material melting step S21, brown ring cores are generated on the inner surface 10i of the quartz glass crucible 1 that contacts the silicon melt. However, in the quartz glass crucible 1 according to the present embodiment, the total concentration of Na, K, and Ca within a depth range (in the first surface layer portion Z.sub.1) of 0 to 8 μm from the crucible inner surface 10i is low, which can reduce the number of brown ring cores being generated in the first half of the raw material dissolution step S21.
[0071] In the latter half of the raw material melting step S21, the brown ring core grows larger. However, in the quartz glass crucible 1 according to the present embodiment, the total concentration of Na, K, and Ca within a depth range (in the second surface layer portion Z.sub.2) of 16 to 32 μm from the crucible inner surface 10i is high, and the peak value of the total concentration of Na, K, and Ca in the second surface layer portion Z.sub.2 is 2 to 19 times the average value of the total concentration of Na, K, and Ca in the first surface layer portion Z.sub.1, so that it is possible to promote dissolution of the crucible inner surface 10i to make the dissolution rate of the crucible inner surface 10i higher than the growth rate of brown rings. Thus, the brown ring cores generated in the first half of the raw material melting step can be removed to reduce the number of brown rings being generated.
[0072] During the growing step of the silicon single crystal from the necking step S23 to the tail section growing step S26, brown rings may not only grow but also partly be peeled off. When some brown rings are transported by melt convection to the solid/liquid interface, dislocations may be generated in the silicon single crystal. However, the total concentration of Na, K, and Ca within a depth range (in the third surface layer portion Z.sub.3) of 32 to 1000 μm from the crucible inner surface 10i is low, and the average value of the total concentration of Na, K, and Ca in the third surface layer portion Z.sub.3 is 0.6 to 1 times the average value of the total concentration of Na, K, and Ca in the first surface layer portion Z.sub.1, so that excessive dissolution of the crucible inner surface 10i can be suppressed. Further, the total concentration of Na, K, and Ca in the third surface layer portion Z.sub.3 exhibits a small change and, in particular, has a gradient of −8.2×10.sup.10 atoms/cc/μm or more and less than 0 atoms/cc/μm, so that it is possible to suppress peeling-off of brown ring due to an abrupt change in the state of the crucible inner surface 10i.
[0073] While the preferred embodiment of the present invention has been described, the present invention is not limited to the above embodiment, and various modifications may be made within the scope of the present invention, and all such modifications are included in the present invention.
EXAMPLES
[0074] Samples A1 to A8 of 32-inch quartz glass crucibles were prepared. The quartz glass crucibles were produced by a rotational molding method as described above, followed by pure water washing, hydrofluoric acid washing, and finish washing in this order. As described above, in the pure water washing, the specific resistance of the pure water was set to 17 MΩ cm or more, and the flow rate thereof was to 50 to 60 L/min. The amount of water to be used per quartz glass crucible was set to 150 liters, and the water temperature was to 55 to 65° C. In the hydrofluoric acid washing, the degree of etching at the inner surface was set to about 8 μm. The distribution (peak position and concentration ratio of the total concentration) in the depth direction of the total concentration of Na, K, and Ca around the crucible inner surface was adjusted by changing the hydrofluoric acid washing time (etching amount).
[0075] Then, the arithmetic average roughness Ra (μm) of the inner surface was measured for each of the quartz glass crucible samples A1 to A8. Further, to evaluate the impurity concentration of a top surface layer portion of the quartz glass crucible inner surface, a sample of a silica glass piece in the top surface layer portion was taken by an SAICAS (Surface And Interfacial Cutting Analysis System) method.
[0076]
[0077] As illustrated in
[0078]
[0079] Then, the total concentration of Na, K, and Ca contained in the silica glass piece of each of the quartz glass crucible samples A1 to A8 was measured according to D-SIMS (Dynamic-Secondary Ion Mass Spectrometry). The impurity concentration was measured, using the sample illustrated in
[0080] Then, other quartz glass crucible samples produced under the same conditions as the samples A1 to A8 were prepared, and a silicon single crystal was actually pulled up therein. After that, the number density (the number of brown rings/cm.sup.2) of brown rings generated on the inner surface of the quartz glass crucible was evaluated. In addition, a lag time (hr) for the silicon single crystal and a single crystal yield (%) were evaluated. It should be noted that the lag time (hr) refers to a time difference between the start time of the necking step for the first time and the start time of the necking step for the last time for producing one silicon single crystal. If the single crystal is produced normally, without doing the single crystal pulling-up over again, the lag time is 0 hr. The single crystal yield refers to a value calculated by (single crystal weight after cylindrical grinding)/(silicon raw material weight)×100%.
[0081]
[0082] As illustrated in
[0083] In comparative example 1 (sample A5), the concentration ratio ([II/I]) in the second surface layer portion Z.sub.2 was 1, the concentration ratio ([III/I]) in the third surface layer portion Z.sub.3 was 0.1, and the concentration gradient in the third surface layer portion Z.sub.3 was −2.9×10.sup.10 atoms/cc/μm. Also the arithmetic average roughness of the outermost surface was 0.03 μm. Further, in the used crucibles after pulling-up of the silicon single crystal, the number density of brown rings on the crucible inner surface was 5/cm.sup.2, and peeling area ratio was 26%. Further, in the results of the silicon single crystal pulling-up, the lag time was 15.3 hrs, and the single crystal yield was 35.1%. In the sample A5, the total concentration of Na, K, and Ca in the second surface layer portion Z.sub.2 is relatively low, and the total concentration of Na, K, and Ca in the second surface layer portion Z.sub.2 was very low, and thus, it is considered that the number of generated brown rings and peeling area ratio of brown rings increased, leading to dislocation of the single crystal.
[0084] In comparative example 2 (sample A6), the concentration ratio ([II/I]) in the second surface layer portion Z.sub.2 was 1, the concentration ratio ([III/I]) in the third surface layer portion Z.sub.3 was 0.1, and the concentration gradient in the third surface layer portion Z.sub.3 was −8.6×10.sup.11 atoms/cc/μm. Also the arithmetic average roughness of the outermost surface was 0.02 μm. Further, in the used crucibles after pulling-up of the silicon single crystal, the number density of brown rings on the crucible inner surface of was 6/cm.sup.2, and peeling area ratio was 31% or less. Further, in the results of the silicon single crystal pulling-up, the lag time was 5.3 hrs, and the single crystal yield was 51.5%. In the sample A6 as well, the total concentration of Na, K, and Ca in the second surface layer portion Z.sub.2 was relatively low, and the total concentration of Na, K, and Ca in the second surface layer portion Z.sub.2 was very low, and thus, it is considered that the number of generated brown rings and peeling area ratio of brown rings increased, leading to dislocation of the single crystal.
[0085] In comparative example 3 (sample A7), the concentration ratio ([II/I]) in the second surface layer portion Z.sub.2 was 2, the concentration ratio ([III/I]) in the third surface layer portion Z.sub.3 was 0.1, and the concentration gradient in the third surface layer portion Z.sub.3 was −5.9×10.sup.10 atoms/cc/μm. Also the arithmetic average roughness of the outermost surface was 0.04 μm. Further, in the used crucibles after pulling-up of the silicon single crystal, the number density of the brown rings on the crucible inner surface was 2/cm.sup.2, and peeling area ratio was 42%. Further, in the results of the silicon single crystal pulling-up, no lag time occurred, but the single crystal yield was 60.5%. In the sample A7, the total concentration of Na, K, and Ca in the second surface layer portion Z.sub.2 was very low, and thus, it is considered that the peeling area ratio increased, leading to dislocation of the single crystal.
[0086] In comparative example 4 (sample A8), the concentration ratio ([II/I]) in the second surface layer portion Z.sub.2 was 21, the concentration ratio ([III/I]) in the third surface layer portion Z.sub.3 was 0.8, and the concentration gradient in the third surface layer portion Z.sub.3 was −4.3×10.sup.10 atoms/cc/μm. Also the arithmetic average roughness of the outermost surface was 0.05 μm. Further, in the used crucibles after pulling-up of the silicon single crystal, the number density of brown rings on the crucible inner surface was 1/cm.sup.2, and peeling area ratio was 36%. Further, in the results of the silicon single crystal pulling-up, no lag time occurred, but the single crystal yield was 45.9%. In the sample A8, the total concentration of Na, K, and Ca in the second surface layer portion Z.sub.2 was relatively excessively high, and thus, it is considered that the number of generated brown rings and peeling area ratio increased, leading to dislocation of the single crystal.
REFERENCE SIGNS LIST
[0087] 1: Quartz glass crucible [0088] 10a: Side wall portion [0089] 10b: Bottom portion [0090] 10c: Corner portion [0091] 10i: Inner surface [0092] 10o: Outer surface [0093] 11: Transparent layer [0094] 12: Bubble layer [0095] 14: Mold [0096] 14i: Mold inner surface [0097] 14a: Vent hole [0098] 15: Arc electrode [0099] 16: Deposited layer [0100] 16A: Synthetic quartz powder [0101] 16B Natural quartz powder [0102] 20: Oblique cutting apparatus [0103] 21: Cutting blade [0104] S11: Crucible producing step [0105] S12: Pure water washing step [0106] S13: Hydrofluoric acid washing (etching) step [0107] S14: Finish washing step S21: Raw material melting step [0108] S22: Dipping step [0109] S23: Necking step [0110] S24: Shoulder portion growing step [0111] S25: Straight body portion growing step [0112] S26: Tail portion growing step [0113] S27: Cooling step [0114] X: Surface layer portion [0115] Z.sub.1: First surface layer portion [0116] Z.sub.2: Second surface layer portion [0117] Z.sub.3: Third surface layer portion