Microbubble characterization method for transparent layer of quartz crucible

Abstract

The invention belongs to the technical field of quartz crucibles, and particularly relates to a microbubble characterization method for a transparent layer of a quartz crucible. According to the method, a seventh sample is obtained through cutting sampling, precise linear cutting separation of a bubble layer and a transparent layer, coarse grinding thinning, accurate grinding, chemical mechanical polishing and multi-step cleaning; the seventh sample is subjected to haze testing, haze data of the transparent layer of the quartz crucible is obtained, the bubble number and grade classification of the transparent layer of the quartz crucible are obtained according to the haze data of the transparent layer of the quartz crucible and a corresponding model, and the corresponding model is a model corresponding to the haze of the transparent layer of the quartz crucible and the number of microbubbles. The method provided by the invention is high in data objectivity, simple in operation process, high in detection efficiency and low in equipment cost; meanwhile, the scattering capacity of bubbles to light can be directly quantified, the correlation with optical interference in actual use is tighter, and microscope counting only reflects the number and cannot directly correspond to the function influence.

Description

Technical Field

[0001] This invention belongs to the field of quartz crucible technology, specifically relating to a microbubble characterization method for the transparent layer of a quartz crucible. Background Technology

[0002] The Czochralski process is a common method for producing monocrystalline silicon ingots in the photovoltaic and semiconductor industries. The quartz crucible is a crucial consumable in the Czochralski process, primarily used to hold the molten silicon. Because the inner layer of the quartz crucible needs to be in prolonged contact with the molten silicon at a high temperature of 1450℃ during the crystal pulling process, the quality of the inner layer is paramount. The inner layer of the quartz crucible is generally designed as a double-layer structure, consisting of a bubble-free transparent layer and a bubble layer containing a large number of bubbles. The number of microbubbles in the transparent layer is one of the important parameters of the crucible, affecting not only the quality of the monocrystalline silicon ingot but also determining the crucible's lifespan. Relevant standards in the photovoltaic and semiconductor industries have strict requirements regarding the number of microbubbles in the transparent layer.

[0003] In the actual production process of crucibles, they are usually broken after production, and the number of microbubbles in the transparent layer is determined by visually observing the transparency of the cross-section of the crucible wall. Although this method is fast and efficient, it relies on the observer's experience, is highly subjective, and the crucible is completely destroyed, which is not conducive to subsequent related tests. GB / T5949—2014, "Test Methods for Bubbles and Gas Lines in Transparent Quartz Glass," provides a method for counting microbubbles in the transparent layer using a microscope. This involves gradually adjusting the focal length under the microscope to count the number of bubbles at different depths, and then taking the largest number as the final count. For microscopes without focal length recording, the adjustment of the focal length and the identification of bubbles (optical microscopes generally have a small depth of field, and bubbles at different depths have different levels of clarity in the same field of view) are highly subjective. Therefore, the largest number of bubbles may be missed during the count, causing the statistical results to deviate from the actual situation. Different operators may also obtain significantly different results due to different criteria for judging bubbles with different levels of clarity. Summary of the Invention

[0004] The purpose of this invention is to provide a microbubble characterization method for the transparent layer of a quartz crucible. The method provided by this invention has strong data objectivity, simple operation process, high detection efficiency, and low equipment cost. At the same time, it can directly quantify the light scattering ability of bubbles, which is more closely related to "optical interference" in actual use. In contrast, microscope counting only reflects the quantity and cannot directly correspond to the functional impact.

[0005] To achieve the above objectives, the present invention provides the following technical solution: This invention provides a method for characterizing microbubbles in the transparent layer of a quartz crucible, comprising the following steps: (1) A sample is taken from the straight wall area of ​​the quartz crucible to obtain the first sample, wherein the transparent layer of the quartz crucible does not contain air bubbles with a size greater than 0.5 mm; (2) Grind the edges of the first sample to obtain the second sample; (3) Remove the bubble layer from the second sample to obtain the third sample; (4) The third sample is coarsely ground after removing the bubble layer from the side surface to obtain the fourth sample; (5) The coarse grinding side surface of the fourth sample is finely ground to obtain the fifth sample; (6) Perform chemical mechanical polishing on both sides of the fifth sample to obtain the sixth sample; (7) The sixth sample is washed and dried to obtain the seventh sample; (8) The seventh sample is subjected to a haze test to obtain haze data of the transparent layer of the quartz crucible. The number and grade classification of bubbles in the transparent layer of the quartz crucible are obtained from the haze data of the transparent layer of the quartz crucible and the corresponding model. The corresponding model is the correspondence model between the haze of the transparent layer of the quartz crucible and the number of microbubbles.

[0006] Preferably, in step (1), the outer diameter of the quartz crucible is ≥330 mm; the sampling is performed by cutting and sampling with a 2-wire parallel diamond wire saw with a diameter of 0.35 mm, and the conditions for cutting and sampling include: the dual-wire synchronous linear speed is 10~15 m / s, the feed speed is 2~3 mm / min; the size of the first sample is 26×26 mm; and the number of the first samples is 4.

[0007] Preferably, in step (2), the grinding is performed using a #600 mesh resin-bonded diamond grinding wheel, and the grinding conditions include: a linear speed of 15~18 m / s, a feed speed of 0.4~0.8 mm / min, and a single cutting depth of 0.01~0.03 mm; the size of the second sample is 25×25 mm.

[0008] Preferably, in step (3), the bubble layer is removed by wire cutting, which is performed using diamond wire with a diameter of 0.2 mm. The conditions for wire cutting include: a feed rate of 0.4~0.6 mm / min and a cutting speed of 10~14 m / s; and the thickness of the third sample is 3.50±0.05 mm.

[0009] Preferably, in step (4), the rough grinding is performed using a CNC surface grinder with a #150 mesh metal-bonded diamond grinding wheel. The conditions for rough grinding include: a grinding wheel linear speed of 22~26 m / s, a feed rate of 2.3~2.7 mm / min, and a single cutting depth of 0.04~0.08 mm; the thickness of the fourth sample is 3.14±0.03 mm.

[0010] Preferably, in step (5), the fine grinding is performed using a #1500 resin-bonded diamond grinding wheel surface grinder, and the fine grinding conditions include: a grinding wheel linear speed of 16~20 m / s, a feed speed of 0.6~1.0 mm / min, and a single cutting depth of 0.003~0.005 mm; the thickness of the fifth sample is 3.02±0.015 mm.

[0011] Preferably, in step (6), the polishing solution used in the chemical mechanical polishing includes CeO2, KOH and water, wherein the content of CeO2 in the polishing solution is 1~3wt%, the content of KOH is 0.2~0.4wt%, and the pH value of the polishing solution is 10~11; the polishing pad used in the chemical mechanical polishing is made of polyurethane; the conditions of the chemical mechanical polishing include: the polishing disc rotation speed is 20~40 rpm, the workpiece rotates in the opposite direction at a speed of 10~15 rpm, the downward pressure is 5~8 kPa, and the polishing time is 10~15 min; the thickness of the sixth sample is 3.00±0.005 mm.

[0012] Preferably, in step (7), the cleaning process includes: pre-washing, first ultrasonic washing, rinsing, and second ultrasonic washing. The pre-washing is performed sequentially using water and ammonia, with the ammonia having a pH of 9-11. The reagent used for the first ultrasonic washing is an EDTA solution, with the EDTA content in the EDTA solution being 1-2 wt%. The frequency of the first ultrasonic washing is 30-45 kHz, and the time is 8-10 min. The reagent used for the rinsing is a citric acid solution, with the citric acid content in the citric acid solution being 0.5-0.8 wt%. The rinsing time is 15-20 min. The reagent used for the second ultrasonic washing is water, with the frequency of the second ultrasonic washing being 700-900 kHz, and the time being 3-5 min. The drying process is centrifugal drying, with the centrifugal drying speed being 3000-4000 rpm, and the time being 3-5 min.

[0013] Preferably, in step (8), the conditions for the haze test include: an ambient temperature of 23±2℃, a relative humidity of 50±5%RH, a measuring aperture of ≤21 mm for the haze meter used in the test, an integrating sphere diameter of 150±10 mm for the haze meter, a test wavelength of 380~780 nm (visible light range), a spot diameter of ≤2 cm, and a haze measurement accuracy of ±0.01%.

[0014] Preferably, the corresponding model is: when the haze value is ≤3%, the number of bubbles in the transparent layer is ≤8 per mm. 2 When the rating is A+ and the haze value is ≤4%, the number of bubbles in the transparent layer is ≤10 per mm. 2 For grade A, when the haze value is ≤6%, the number of bubbles in the transparent layer is ≤15 per mm. 2 The grade is B.

[0015] This invention provides a method for characterizing microbubbles in the transparent layer of a quartz crucible, comprising the following steps: (1) sampling the straight wall area of ​​the quartz crucible to obtain a first sample, wherein the transparent layer of the quartz crucible does not contain bubbles larger than 0.5 mm; (2) grinding the edge of the first sample to obtain a second sample; (3) removing the bubble layer from the second sample to obtain a third sample; (4) coarsely grinding the side surface of the third sample after removing the bubble layer to obtain a fourth sample; (5) finely grinding the side surface of the fourth sample to obtain a fifth sample; (6) chemically and mechanically polishing both sides of the fifth sample to obtain a sixth sample; (7) cleaning and drying the sixth sample to obtain a seventh sample; (8) performing a haze test on the seventh sample to obtain haze data of the transparent layer of the quartz crucible, and obtaining the number and grade classification of bubbles in the transparent layer of the quartz crucible from the haze data and the corresponding model, wherein the corresponding model is a model of the haze of the transparent layer of the quartz crucible corresponding to the number of microbubbles. Haze refers to the degree to which light deviates from its original direction when passing through a transparent or translucent material due to scattering caused by defects (such as bubbles, impurities, and grain boundaries) within or on the material's surface. It is usually expressed as a percentage. Haze is equal to the ratio of scattered light flux to total transmitted light flux, reflecting the material's optical uniformity. A higher haze value indicates more severe light scattering and poorer transparency of the material. Since the transparent layer of a crucible is made of quartz glass, which lacks grain boundaries, and the impurity content in the transparent layer of quartz crucibles used in crystal pulling in the photovoltaic and semiconductor industries generally needs to be below 20 ppm, the factors that scatter light in the transparent layer are mainly microbubbles and a small amount of scattering impurities. Among these, the bubbles are mainly 10-100 μm microbubbles, which have a strong scattering effect on visible light. Therefore, the haze of the crucible's transparent layer is closely related to the number of microbubbles in the transparent layer. Compared with the currently widely used microscopic counting method, the microbubble characterization method of the transparent layer of the quartz crucible provided by this invention, which uses the haze of the transparent layer to characterize the number of microbubbles, has the following advantages: First, with sample pretreatment completed, the haze method only requires a haze meter to complete the detection, resulting in low equipment cost. Each sample only requires 3-5 minutes, enabling "instant testing," and the data is highly subjective, allowing for direct quantitative comparison. If automated equipment is used for sample grinding and other pretreatment processes, the detection efficiency is high, suitable for online or offline quality monitoring in production lines. Second, one of the hazards of microbubbles in the transparent layer of the quartz crucible to crystal pulling is the scattering of infrared radiation in the melt, leading to local temperature unevenness and affecting silicon crystal growth. The haze value directly quantifies the ability of bubbles to scatter light, and is more closely related to "optical interference" in actual use, while microscopic counting only reflects quantity and cannot directly correspond to functional impact. Attached Figure Description

[0016] Figure 1 This is a schematic diagram of the crucible transparency layer haze testing process provided by the present invention. Detailed Implementation

[0017] This invention provides a method for characterizing microbubbles in the transparent layer of a quartz crucible, comprising the following steps: (1) A sample is taken from the straight wall area of ​​the quartz crucible to obtain the first sample, wherein the transparent layer of the quartz crucible does not contain air bubbles with a size greater than 0.5 mm; (2) Grind the edges of the first sample to obtain the second sample; (3) Remove the bubble layer from the second sample to obtain the third sample; (4) The third sample is coarsely ground after removing the bubble layer from the side surface to obtain the fourth sample; (5) The coarse grinding side surface of the fourth sample is finely ground to obtain the fifth sample; (6) Perform chemical mechanical polishing on both sides of the fifth sample to obtain the sixth sample; (7) The sixth sample is washed and dried to obtain the seventh sample; (8) The seventh sample is subjected to a haze test to obtain haze data of the transparent layer of the quartz crucible. The number and grade classification of bubbles in the transparent layer of the quartz crucible are obtained from the haze data of the transparent layer of the quartz crucible and the corresponding model. The corresponding model is the correspondence model between the haze of the transparent layer of the quartz crucible and the number of microbubbles.

[0018] In this invention, unless otherwise specified, all raw materials / components used in the preparation are commercially available products well known to those skilled in the art.

[0019] Step (1): In this invention, a sample is taken from the straight-walled area of ​​the quartz crucible to obtain a first sample. The transparent layer of the quartz crucible does not contain air bubbles larger than 0.5 mm. In this invention, the sampling point is preferably located at the intersection of any two vertical diameters in the middle of the straight wall of the crucible with the crucible wall. The outer diameter of the quartz crucible is preferably ≥330 mm. By controlling the outer diameter of the quartz crucible, this invention can ensure that the inner surface of the first sample is flat. The inner surface of the quartz crucible should not have obvious protrusions or depressions.

[0020] In this invention, the sampling is preferably performed using a 2-wire parallel diamond wire saw with a diameter of 0.35 mm. The preferred conditions for the cutting and sampling include: a dual-wire synchronous linear speed of 10-15 m / s (12 m / s in this embodiment); and a feed rate of 2-3 mm / min (2 mm / min in this embodiment). During the cutting and sampling process, this invention preferably uses a water-based synthetic coolant for cooling to reduce damage to the sample during processing.

[0021] In this invention, the size of the first sample is preferably 26 × 26 mm. The quantity of the first sample is preferably 4.

[0022] Step (2): After obtaining the first sample, the edges of the first sample are polished to obtain the second sample. In this invention, the polishing is preferably performed using a #600 resin-bonded diamond grinding wheel. The preferred polishing conditions include: a linear speed of 15~18 m / s, which can be 16 m / s in the embodiment; a feed rate of 0.4~0.8 mm / min, which can be 0.6 mm / min in the embodiment; and a single cutting depth of 0.01~0.03 mm, which can be 0.02 mm in the embodiment. During the polishing process, this invention preferably uses a water-based synthetic coolant for cooling.

[0023] In this invention, the size of the second sample is preferably 25 × 25 mm. The quantity of the second sample is preferably four.

[0024] Step (3): After obtaining the second sample, the present invention removes the bubble layer from the second sample to obtain the third sample. In the present invention, the removal of the bubble layer is preferably performed by wire cutting. The wire cutting is preferably performed using diamond wire, and the diameter of the diamond wire is preferably 0.2 mm. The wire cutting conditions preferably include: a feed rate of 0.4~0.6 mm / min, which can be 0.5 mm / min in the embodiment; and a cutting speed of 10~14 m / s, which can be 12 m / s in the embodiment. During the wire cutting process, the present invention preferably uses a water-based synthetic coolant for cooling. In this invention, the thickness of the third sample is preferably 3.50 ± 0.05 mm. The number of third samples is preferably four.

[0025] Step (4): After obtaining the third sample, the third sample is rough-ground to remove the bubble layer side surface, resulting in the fourth sample. In this invention, the rough grinding is preferably performed using a CNC surface grinder with a #150 mesh metal-bonded diamond grinding wheel. The preferred conditions for rough grinding include: a wheel linear speed of 22~26 m / s (24 m / s in this example); a feed rate of 2.3~2.7 mm / min (2.5 mm / min in this example); and a single-pass depth of cut of 0.04~0.08 mm (0.06 mm in this example). During the rough grinding process, this invention preferably uses a water-based synthetic coolant for continuous cooling. During the rough grinding process, this invention preferably uses a laser thickness gauge to control the thickness of the transparent layer of the sample. In this invention, the thickness of the fourth sample is preferably 3.14 ± 0.03 mm. The number of fourth samples is preferably four.

[0026] Step (5): After obtaining the fourth sample, the rough-ground side surface of the fourth sample is fine-ground to obtain the fifth sample. In this invention, the fine grinding is preferably performed using a #1500 resin-bonded diamond wheel surface grinder. The preferred conditions for fine grinding include: a wheel linear speed of 16~20 m / s, which can be 18 m / s in the embodiment; a feed rate of 0.6~1.0 mm / min, which can be 0.8 mm / min in the embodiment; and a single cutting depth of 0.003~0.005 mm, which can be 0.004 mm in the embodiment. In the fine grinding process, deionized water is preferably used as the coolant. In the fine grinding process, a laser thickness gauge is preferably used to control the thickness of the transparent layer.

[0027] In this invention, the thickness of the fifth sample is preferably 3.02 ± 0.015 mm. The number of fifth samples is preferably four.

[0028] Step (6): After obtaining the fifth sample, the present invention performs chemical mechanical polishing on both sides of the fifth sample to obtain the sixth sample. In the present invention, the polishing solution used for chemical mechanical polishing preferably includes CeO2, KOH and water. The content of CeO2 in the polishing solution is preferably 1~3wt%, and can be 2wt% in the example; the content of KOH is preferably 0.2~0.4wt%, and can be 0.3wt% in the example. The pH value of the polishing solution is preferably 10~11. The polishing pad used for chemical mechanical polishing is preferably a polyurethane pad. The conditions for chemical mechanical polishing preferably include: the polishing disc speed is preferably 20~40 rpm, and can be 30 rpm in the example; the workpiece is preferably rotated in the opposite direction at a speed of 10~15 rpm, and can be 10 rpm in the example; the downward pressure is preferably 5~8 kPa, and the polishing time is preferably 10~15 min. In the process of chemical mechanical polishing, the present invention preferably uses a laser thickness gauge to control the thickness of the transparent layer of the sample. In this invention, the thickness of the sixth sample is preferably 3.00 ± 0.005 mm. The number of the sixth samples is preferably four.

[0029] Step (7): After obtaining the sixth sample, the present invention cleans and dries the sixth sample to obtain the seventh sample. In the present invention, the cleaning preferably includes, in sequence: pre-washing, first ultrasonic washing, rinsing, and second ultrasonic washing. In the present invention, the pre-washing preferably uses water and ammonia water in sequence, and the water is preferably deionized water. The pH value of the ammonia water is preferably 9-11, and can be 10 in the embodiment. The pre-washing is preferably rinsing. During the pre-washing process, the present invention preferably rinses the loose particles and residual polishing liquid on the surface of the sixth sample with deionized water. In the present invention, the reagent used for the first ultrasonic washing is preferably an EDTA solution, and the EDTA content in the EDTA solution is preferably 1-2 wt%. The frequency of the first ultrasonic washing is preferably 30-45 kHz, and can be 40 kHz in the embodiment. The time of the first ultrasonic washing is preferably 8-10 min. The present invention preferably removes residual organic matter and residual CeO2 through the first ultrasonic washing. In the present invention, the reagent used for the rinsing is preferably a citric acid solution. The citric acid content in the citric acid solution is preferably 0.5-0.8 wt%. The rinsing time is preferably 15-20 min. This invention preferably dissolves Ce-O-Si bonds and byproducts generated during the rinsing and polishing process. In this invention, the reagent used for the second ultrasonic cleaning is preferably water, and the water is preferably deionized water. The frequency of the second ultrasonic cleaning is preferably 700-900 kHz, and in the embodiments, it can be 800 kHz. The time for the second ultrasonic cleaning is preferably 3-5 min. The drying is preferably centrifugal drying, and the centrifugal drying speed is preferably 3000-4000 rpm, and the time is preferably 3-5 min.

[0030] Step (8): After obtaining the seventh sample, the present invention performs a haze test on the seventh sample to obtain haze data of the transparent layer of the quartz crucible. The number and grade classification of bubbles in the transparent layer of the quartz crucible are obtained from the haze data and the corresponding model. The corresponding model is a model relating the haze of the transparent layer of the quartz crucible to the number of microbubbles. In the present invention, the preferred conditions for the haze test include: an ambient temperature preferably of 23±2℃; a relative humidity preferably of 50±5%RH; a measuring aperture of ≤21 mm for the haze meter used in the test; an integrating sphere diameter of 150±10 mm for the haze meter; a test wavelength preferably of 380~780 nm; a spot diameter preferably of ≤2 cm; and a haze measurement accuracy preferably of ±0.01%. Since most of the microbubbles in the transparent layer of the quartz crucible have a diameter of 10~100 micrometers, light in this wavelength range scatters strongly from microbubbles of this diameter. Therefore, the preferred test wavelength for the haze test in the present invention is 380~780 nm. In this invention, the remaining specifications of the haze meter shall comply with the GB / T 2410-2008 standard.

[0031] In this invention, the number of the seventh samples is preferably four. Preferably, the arithmetic mean of the haze data of the four seventh samples is used as the haze data of the transparent layer of the quartz crucible, and the haze data is accurate to 0.1%.

[0032] The present invention preferably establishes a correspondence model between the haze of the transparent layer and the number of microbubbles. During the establishment of this correspondence model, the haze data of the quartz crucible measured according to the above method is optimized.

[0033] This invention uses the haze of the transparent layer of a quartz crucible as a characterization parameter for the number of microbubbles in the transparent layer and classifies them into grades, as shown in Table 1, as reference data for the quality evaluation of the transparent layer of the crucible.

[0034] Table 1. Determination of Haze Level in the Transparent Layer of the Crucible

[0035] In this invention, the corresponding model is: when the haze value is ≤3%, the number of bubbles in the transparent layer is ≤8 per mm. 2 When the rating is A+ and the haze value is ≤4%, the number of bubbles in the transparent layer is ≤10 per mm. 2 For grade A, when the haze value is ≤6%, the number of bubbles in the transparent layer is ≤15 per mm. 2 The grade is B.

[0036] In summary, this invention employs a sample preparation process for obtaining haze test samples of the transparent layer of a quartz crucible: "two-wire parallel wire cutting → precision wire cutting to separate the bubble layer and transparent layer → rough grinding for thinning → fine grinding → chemical mechanical polishing → multi-step cleaning." This process is time-efficient and minimizes direct damage to the transparent layer during processing, as well as microcracks caused by thermal stress. It also reduces the surface roughness of the transparent layer, minimizing interference with haze testing. This invention uses a haze meter to measure the haze of the transparent layer of the quartz crucible and uses haze to characterize the number of microbubbles in the transparent layer. This method offers high detection efficiency, low equipment cost, and strong data objectivity. The obtained values ​​are closely correlated with "optical interference" in actual use, providing valuable practical guidance for production.

[0037] To further illustrate the present invention, the technical solutions provided by the present invention will be described in detail below with reference to the embodiments, but they should not be construed as limiting the scope of protection of the present invention.

[0038] Example 1 This embodiment provides a method for characterizing microbubbles in the transparent layer of a quartz crucible, the flowchart of which is shown below. Figure 1 As shown, the specific steps include: (1) The sampling point is the intersection of any two vertical diameters in the middle of the straight wall of the crucible with the crucible wall (i.e., draw a cross line with the diameter length in the middle of the straight wall, and the intersection of the cross line with the crucible wall is the center of the sample to reduce sampling error). Four 26×26 mm samples are cut from the middle of the straight wall of the quartz crucible using a 0.35 mm diameter parallel diamond wire saw (to increase cutting efficiency). These samples are named Sample 1. During the cutting process, the dual-wire synchronous linear speed is 12 m / s and the feed rate is 2 mm / min. Water-based synthetic coolant is used for cooling during the wire cutting process to reduce damage to the sample during processing. To ensure the flatness of the inner surface of the sample, the outer diameter of the sampling crucible must be greater than or equal to 330 mm, the inner surface must not have obvious protrusions or depressions, and there must be no obvious impurity points or bubbles larger than 0.5 mm in the transparent layer.

[0039] (2) The edge of sample 1 was ground by CNC surface grinder. A #600 resin-bonded diamond grinding wheel was selected, with a linear speed of 16 m / s, a feed rate of 0.6 mm / min, and a single cutting depth of 0.02 mm. Water-based synthetic coolant was used for cooling during the grinding process, and finally sample 2 with a side length of 25×25 mm was obtained.

[0040] (3) The bubble layer of sample 2 was removed by precision wire cutting. The cutting wire was a diamond cutting wire with a diameter of 0.2 mm, the feed rate was 0.5 mm / min, the cutting speed was 12 m / s, and water-based synthetic coolant was used for cooling during the cutting process. Finally, sample 3 with a thickness of 3.50±0.05 mm was obtained.

[0041] (4) The bubble layer side surface of sample 3 was rough ground by a CNC surface grinder with a #150 mesh metal bond diamond grinding wheel. The grinding wheel linear speed was 24 m / s, the feed speed was 2.5 mm / min, and the single cutting depth was 0.06 mm. During the grinding process, water-based synthetic coolant was continuously used for cooling, and the thickness of the transparent layer was controlled by a laser thickness gauge. Finally, sample 4 with a thickness of 3.14±0.03 mm was obtained.

[0042] (5) The bubble layer side surface of sample 4 was finely ground using a #1500 resin-bonded diamond grinding wheel surface grinder. During the grinding process, the grinding wheel linear speed was 18 m / s, the feed speed was 0.8 mm / min, and the single cutting depth was 0.004 mm. Deionized water was used as the coolant during the grinding process, and the thickness of the transparent layer was controlled by a laser thickness gauge. Finally, sample 5 with a thickness of 3.02±0.015 mm was obtained.

[0043] (6) The outer surfaces of both sides of sample 5 were chemically and mechanically polished. The polishing solution was a mixed solution of CeO2 with a concentration of 2 wt% and KOH with a concentration of 0.3 wt%. The pH of the polishing solution was 10-11. The polishing pad was a polyurethane pad. The polishing disc rotated at 30 rpm, and the workpiece rotated in the opposite direction at 10 rpm. The downward pressure was 6.5±1.5 kPa, and the polishing time was 12.5±2.5 min. During the polishing process, the thickness of the transparent layer was controlled by a laser thickness gauge. Finally, sample 6 with a thickness of 3.00±0.005 mm was obtained.

[0044] (7) Sample 6 was cleaned by rinsing with deionized water to remove loose particles and residual polishing liquid on the surface → ammonia water with pH 10 + 1 wt% EDTA + 40 kHz sonication for 8 minutes (to remove residual organic matter and residual CeO2) → rinsing with 0.5% citric acid solution for 15 seconds (to dissolve Ce-O-Si bonds and byproducts generated during polishing) → sonication with deionized water at 800 kHz for 3 minutes → centrifugation at 3000 rpm for 3 minutes. After drying, it was recorded as sample 7.

[0045] (8) Perform a haze test on sample 7. The test environment shall be a clean environment of 23±2℃ and 50±5%RH. The sample shall be allowed to stand in this environment for 30 minutes before the test. The measuring aperture of the haze meter used for the test shall be less than or equal to 21 mm, the integrating sphere diameter of the haze meter shall be 150±10 mm, the test wavelength shall be 380~780 nm (most of the microbubbles in the transparent layer of the crucible are 10~100 micrometers in diameter, and light of this wavelength range will scatter strongly from microbubbles of this diameter), the spot diameter shall be ≤2 cm, the haze measurement accuracy shall be ±0.01%, and the other indicators of the haze meter shall comply with GB / T 2410-2008 standard.

[0046] The arithmetic mean of the haze of the four samples was taken as the haze of the transparent layer of the crucible (accurate to 0.1%).

[0047] A model was established to correspond to the haze of the transparent layer and the number of microbubbles. The haze of the transparent layer of the crucible was used as a characterization parameter for the number of microbubbles in the transparent layer and classified into levels, as shown in Table 1, which serves as reference data for the quality evaluation of the transparent layer of the crucible.

[0048] As can be seen from the above embodiments, the method provided by the present invention has a short sample preparation time and can reduce the direct damage to the transparent layer during processing and the microcracks caused by thermal stress, reduce the surface roughness of the transparent layer, and reduce the interference to haze testing. At the same time, the present invention uses the haze of the transparent layer of the quartz crucible to characterize the number of microbubbles, which has high detection efficiency, low equipment cost, strong data objectivity, and the obtained values ​​are closely related to the "optical interference" in actual use, and have more practical production guidance value.

[0049] Although the above embodiments have provided a detailed description of the present invention, they are only some embodiments of the present invention, and not all embodiments. Other embodiments can be obtained based on these embodiments without creative effort, and these embodiments all fall within the protection scope of the present invention.

Claims

1. A method for characterizing microbubbles in the transparent layer of a quartz crucible, characterized in that, Includes the following steps: (1) A sample is taken from the straight wall area of ​​the quartz crucible to obtain the first sample, wherein the transparent layer of the quartz crucible does not contain air bubbles with a size greater than 0.5 mm; (2) Grind the edges of the first sample to obtain the second sample; (3) Remove the bubble layer from the second sample to obtain the third sample; (4) The third sample is coarsely ground after removing the bubble layer from the side surface to obtain the fourth sample; (5) The coarse grinding side surface of the fourth sample is finely ground to obtain the fifth sample; (6) Perform chemical mechanical polishing on both sides of the fifth sample to obtain the sixth sample; (7) The sixth sample is washed and dried to obtain the seventh sample; (8) The seventh sample is subjected to a haze test to obtain haze data of the transparent layer of the quartz crucible. The number and grade classification of bubbles in the transparent layer of the quartz crucible are obtained from the haze data of the transparent layer of the quartz crucible and the corresponding model. The corresponding model is the correspondence model between the haze of the transparent layer of the quartz crucible and the number of microbubbles.

2. The microbubble characterization method for the transparent layer of a quartz crucible according to claim 1, characterized in that, In step (1), the outer diameter of the quartz crucible is ≥330 mm; the sampling is performed by cutting and sampling with a 2-wire parallel diamond wire saw with a diameter of 0.35 mm. The conditions for cutting and sampling include: the dual-wire synchronous linear speed is 10~15 m / s and the feed speed is 2~3 mm / min; the size of the first sample is 26×26 mm; and the number of the first samples is 4.

3. The microbubble characterization method for the transparent layer of a quartz crucible according to claim 1, characterized in that, In step (2), the grinding is carried out using a #600 mesh resin-bonded diamond grinding wheel. The grinding conditions include: a linear speed of 15~18 m / s, a feed speed of 0.4~0.8 mm / min, and a single cutting depth of 0.01~0.03 mm; the size of the second sample is 25×25 mm.

4. The microbubble characterization method for the transparent layer of a quartz crucible according to claim 1, characterized in that, In step (3), the bubble layer is removed by wire cutting, which is performed using diamond wire with a diameter of 0.2 mm. The conditions for wire cutting include: feed rate of 0.4~0.6 mm / min and cutting speed of 10~14 m / s; the thickness of the third sample is 3.50±0.05 mm.

5. The microbubble characterization method for the transparent layer of a quartz crucible according to claim 1, characterized in that, In step (4), the rough grinding is performed using a CNC surface grinder with a #150 mesh metal-bonded diamond grinding wheel. The conditions for rough grinding include: a grinding wheel linear speed of 22~26 m / s, a feed rate of 2.3~2.7 mm / min, and a single cutting depth of 0.04~0.08 mm; the thickness of the fourth sample is 3.14±0.03 mm.

6. The microbubble characterization method for the transparent layer of a quartz crucible according to claim 1, characterized in that, In step (5), the fine grinding is carried out using a #1500 resin-bonded diamond grinding wheel surface grinder. The fine grinding conditions include: a grinding wheel linear speed of 16~20 m / s, a feed rate of 0.6~1.0 mm / min, and a single cutting depth of 0.003~0.005 mm; the thickness of the fifth sample is 3.02±0.015 mm.

7. The microbubble characterization method for the transparent layer of a quartz crucible according to claim 1, characterized in that, In step (6), the polishing solution used in the chemical mechanical polishing includes CeO2, KOH and water. The content of CeO2 in the polishing solution is 1~3wt%, the content of KOH is 0.2~0.4wt%, and the pH value of the polishing solution is 10~11. The polishing pad used in the chemical mechanical polishing is made of polyurethane. The conditions of the chemical mechanical polishing include: the polishing disc speed is 20~40 rpm, the workpiece rotates in the opposite direction at a speed of 10~15 rpm, the downward pressure is 5~8 kPa, and the polishing time is 10~15 min. The thickness of the sixth sample is 3.00±0.005 mm.

8. The microbubble characterization method for the transparent layer of a quartz crucible according to claim 1, characterized in that, In step (7), the cleaning process includes: pre-washing, first ultrasonic washing, rinsing, and second ultrasonic washing. The pre-washing is performed sequentially using water and ammonia, with the ammonia having a pH of 9-11. The reagent used for the first ultrasonic washing is an EDTA solution, with an EDTA content of 1-2 wt%. The frequency of the first ultrasonic washing is 30-45 kHz, and the time is 8-10 min. The reagent used for the rinsing is a citric acid solution, with a citric acid content of 0.5-0.8 wt%. The rinsing time is 15-20 min. The reagent used for the second ultrasonic washing is water, with a frequency of 700-900 kHz, and the time is 3-5 min. The drying process is centrifugal drying, with a centrifugal speed of 3000-4000 rpm and a time of 3-5 min.

9. The microbubble characterization method for the transparent layer of a quartz crucible according to claim 1, characterized in that, In step (8), the conditions for the haze test include: ambient temperature of 23±2℃, relative humidity of 50±5%RH, the measuring aperture of the haze meter used for the test is ≤21 mm, the diameter of the integrating sphere of the haze meter is 150±10 mm, the test wavelength is 380~780 nm, the spot diameter is ≤2 cm, and the haze measurement accuracy is ±0.01%.

10. The microbubble characterization method for the transparent layer of a quartz crucible according to claim 1, characterized in that, The corresponding model is as follows: when the haze value is ≤3%, the number of bubbles in the transparent layer is ≤8 per mm. 2 When the rating is A+ and the haze value is ≤4%, the number of bubbles in the transparent layer is ≤10 per mm. 2 For grade A, when the haze value is ≤6%, the number of bubbles in the transparent layer is ≤15 per mm. 2 The grade is B.

Images