Manufacturing method of ingot with reduced distortion in the latter half of the body length

Abstract

A method for growing single crystal silicon ingots with reduced diameter deviation by the Czochralski method is disclosed.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of priority to U.S. Provisional Application No. 62 / 832,561, filed April 11, 2019, the disclosure of which is incorporated by reference as if set forth in its entirety. [Technical Field]

[0002] The field of the disclosure relates to methods for growing single crystal silicon ingots using the Czochralski method. [Background technology]

[0003] Single-crystal silicon, the starting material for most processes for the fabrication of semiconductor electronic components, is typically produced by the Czochralski ("Cz") method. In this method, polycrystalline silicon (hereafter "polysilicon") is introduced into a crucible and melted. A seed crystal is then slowly pulled up to grow a single crystal. Once the neck is formed, the crystal diameter is increased by decreasing the pulling rate or melt temperature until the desired diameter is reached. The pulling rate and melt temperature are then controlled to compensate for the decrease in melt level, resulting in the growth of a cylindrical crystal body with a nearly constant diameter. Near the end of the growth process, the crystal diameter is typically gradually reduced to form an end-cone-shaped tail end before the crucible is emptied of molten silicon. The end-cone is typically formed by increasing the crystal pulling rate and the heat input to the crucible. Once the crystal reaches a sufficiently small diameter, it is separated from the melt.

[0004] Czochralski growth methods include batch Czochralski and continuous Czochralski. In batch CZ, a single polycrystalline charge is loaded into a crucible, and that single charge is sufficient to grow a single crystalline silicon ingot, after which the crucible is essentially depleted of silicon melt. In continuous Czochralski (CCZ) growth, polycrystalline silicon can be added continuously or periodically to replenish the silicon melt during the growth process, allowing multiple ingots to be removed from a single crucible during the growth process.

[0005] To implement the CCZ process, conventional batch-type Czochralski growth chambers and equipment are modified to include a means for continuously or semi-continuously supplying additional polycrystalline silicon to the melt without adversely affecting the properties of the growing ingot. As the seed crystal continuously grows from the melt, solid polycrystalline silicon, such as granular polycrystalline silicon, is added to the melt to replenish the melt. The rate at which the solid polycrystalline silicon is added to the melt is typically controlled to maintain process parameters. To mitigate the adverse effects of this replenishment on the simultaneous crystal growth, conventional quartz crucibles are often modified to include an outer or annular melt zone where additional material is introduced and an inner growth zone from which the silicon ingot is withdrawn. These zones are in fluid communication with each other.

[0006] The continuous shrinking of the size of microelectronic devices in recent years has imposed difficult limits on the quality of silicon substrates, which is fundamentally determined by the size and distribution of grown microdefects. Most of the microdefects formed in silicon crystals grown by the Czochralski (CZ) and float-zone (FZ) processes are point defects inherent in silicon, namely agglomerates of vacancies and self-lattices (or simply interstitials).

[0007] A series of studies identified two forms of interstitial aggregates: spherical interstitial clusters called B-vortex defects (or B defects) and dislocation loops called A-vortex defects (or A defects). Later, clusters of vacancies called D defects were identified as octahedral voids. Voronkov successfully explained the distribution of microdefects in silicon crystals based on the crystal growth conditions. According to Voronkov's model, the temperature field near the melt / crystal interface promotes point defect recombination and drives point defect diffusion from the melt / crystal interface, where equilibrium concentrations exist, into the bulk of the crystal. The interplay between point defect transport by both diffusion and convection and recombination determines the point defect concentration beyond a short distance from the interface, known as the recombination length. Typically, the difference between the vacancy and interstitial concentrations beyond the recombination length, known as the excess point defect concentration, is essentially constant away from the sides of the crystal. In rapidly pulled crystals, spatial redistribution due to point defect diffusion beyond the recombination length is generally insignificant. However, regions near the sides of the crystal act as either a sink or a source of point defects. Therefore, if the excess point defect concentration beyond the recombination length is positive, vacancies remain in excess and, at low temperatures, agglomerate to form D defects. If the excess point defect concentration is negative, interstitials remain the dominant point defect species, agglomerating to form A and B defects. If the excess point defect concentration is below a certain detection threshold, no detectable microdefects form. Thus, the type of microdefect grown is usually determined simply by the excess point defect concentration established beyond the recombination length. The process of establishing the excess point defect concentration is called initial incorporation, and the dominant point defect species is called the incorporated dominant point defect. The type of incorporated point defect is determined by the ratio of the crystal pulling rate (v) to the magnitude of the axial temperature gradient (G) near the interface. When v / G is large, point defect convection dominates diffusion, and the vacancy concentration at the interface is higher than the interstitial concentration, so vacancies remain the incorporated dominant point defect. At low v / G, diffusion dominates convection, allowing the faster-diffusing interstitials to be incorporated as the dominant point defects. At v / G close to the critical value, both types of point defects are incorporated at very low and comparable concentrations, and they annihilate each other, suppressing the formation of microdefects at low temperatures.The observed spatial microdefect distribution can typically be explained by the variation of v / G caused by the radial nonuniformity of G and the axial variation of v. A notable feature of the radial microdefect distribution is oxide particles formed by oxygen-vacancy interactions in regions with relatively low incorporated vacancy concentrations, i.e., in a small range of v / G slightly above the critical v / G. These particles form narrow spatial bands that are revealed by thermal oxidation as OSF (oxidation-induced stacking fault) rings. In many cases, the OSF rings mark the adjacent boundary between vacancy-dominated and interstitial-dominated crystalline regions, known as the V / I boundary.

[0008] However, in many modern processes, the microdefect distribution in CZ crystals grown at slow rates is affected by point defect diffusion in the crystal bulk, including diffusion induced from the crystal's lateral sides. Therefore, to accurately quantify the microdefect distribution in CZ crystals, it is preferable to consider two-dimensional point defect diffusion in both the axial and radial directions. The microdefect distribution in CZ crystals can be qualitatively understood by quantifying only the point defect concentration field, since the type of microdefect formed is directly determined. However, to quantify the microdefect distribution more accurately, it is necessary to capture the point defect agglomeration state. Traditionally, quantification of the microdefect distribution has been performed by separating the initial point defect incorporation from the subsequent microdefect formation. This method ignores the diffusion of point defects, which dominate near the nucleation region, from high-temperature regions (where the microdefect density is negligible) to low-temperature regions (where the microdefects are present in high density and consume the point defects). Alternatively, rigorous numerical simulations based on predicting the size distribution of the microdefect population at every location in the crystal are numerically expensive.

[0009] The transition between vacancy and interstitial dominance in the material occurs at a critical value of v / G, which is currently about 2.5×10 -5 cm 2 / sK. When v / G exceeds a critical value, vacancies become the predominant intrinsic point defect, and their concentration increases with increasing v / G. When v / G is below a critical value, silicon self-lattice defects become the predominant intrinsic point defect, and their concentration increases as v / G decreases. Therefore, process conditions such as growth rate (which affects v) and hot zone configuration (which affects G) can be controlled to determine whether vacancies (v / G generally above the critical value) or silicon self-lattice defects (v / G generally below the critical value) will predominate as intrinsic point defects in single crystal silicon.

[0010] The formation of agglomerated defects generally occurs in two steps: first, defect "nucleation" occurs, which is the result of intrinsic point defects becoming supersaturated at a certain temperature, and above this "nucleation threshold" temperature, the intrinsic point defects remain dissolved in the silicon lattice. The nucleation temperature for agglomerated intrinsic point defects is approximately 1000 °C or higher.

[0011] Once this "nucleation threshold" temperature is reached, the intrinsic point defects agglomerate, i.e., precipitate out of the "solid solution" of the silicon lattice. The intrinsic point defects continue to diffuse through the silicon lattice as long as the temperature of the portion of the ingot in which they reside remains above a second threshold temperature (i.e., the "diffusivity threshold"). Below this "diffusivity threshold" temperature, the intrinsic point defects do not migrate within commercially practical time periods.

[0012] While the ingot is held above the "diffusivity threshold" temperature, vacancies or interstitials diffuse through the silicon lattice, reaching sites where agglomerated vacancies or interstitials, respectively, already exist, increasing the size of the given agglomerated defect. Growth occurs because these agglomerated defect sites essentially act as "sinks," attracting and collecting intrinsic point defects due to their more favorable energy conditions for agglomeration.

[0013] Vacancy-type defects are recognized as the source of observable crystalline defects, such as D defects, flow pattern defects (FPDs), gate oxide integrity (GOI) defects, crystal-originating particle (COP) defects, and light point defects (LPDs), as well as certain classes of bulk defects observed with infrared light scattering techniques such as scanning infrared microscopy and laser scanning tomography. Oxygen and silicon dioxide clusters also exist in regions of excess vacancies. While some of these clusters are small and relatively strain-free, essentially harmless to the majority of devices fabricated from such silicon, some clusters are large enough to nucleate ring-shaped oxidation-induced stacking faults (OISFs). The formation of such defects is speculated to be promoted by the presence of excess vacancies, which act as catalysts for the agglomeration of previously nucleated oxygen. These oxide clusters primarily form in the presence of moderate vacancy concentrations during CZ growth below 1000 °C.

[0014] Self-interstitial defects have not been studied extensively. They are generally considered to be low-density interstitial-type dislocation loops or networks. While such defects do not cause gate oxide integrity failures, a critical wafer performance criterion, they are widely recognized as the cause of other types of device failures, typically related to current leakage issues.

[0015] In this regard, oxygen in the form of interstitials within the silicon lattice is generally considered to be a point defect in silicon, but not an intrinsic point defect. On the other hand, silicon lattice vacancies and silicon self-interstitials (or simply interstitials) are generally considered to be intrinsic point defects. Thus, essentially all microdefects are generally described as agglomerated point defects, while D-defects (or voids), A-defects, and B-defects (i.e., interstitial defects) are more specifically described as agglomerated intrinsic point defects. Also, because oxygen clusters are formed by absorbing vacancies, they can be considered to be agglomerations of both vacancies and oxygen.

[0016] Historically, the density of such vacancies and self-lattice agglomerated point defects in Czochralski silicon has been estimated to be approximately 1 × 10 3 / cm 3 to approximately 1 x 10 7 / cm 3 , whereas the density of oxygen clusters is about 1×10 8 / cm 3 From 1×10 10 / cm 3 Agglomerated intrinsic point defects are rapidly becoming of importance to device manufacturers because they severely impact the yield of single-crystal silicon materials in the fabrication of complex, highly integrated circuits.

[0017] For these reasons, in many applications, it is preferable that some or all of the silicon crystals that are subsequently sliced ​​into silicon wafers be substantially free of these agglomerated intrinsic point defects. Several approaches for growing substantially defect-free silicon crystals have been reported. Generally, these approaches all involve controlling the v / G ratio to determine the initial type and concentration of intrinsic point defects present in the growing CZ single crystal silicon crystal. Furthermore, these approaches involve controlling the subsequent thermal history of the crystal to allow for a longer diffusion time to suppress the concentration of intrinsic point defects therein, thereby substantially limiting or avoiding the formation of agglomerated intrinsic point defects in some or all of the crystal (see, e.g., U.S. Pat. Nos. 6,287,380, 6,254,672, 5,919,302, 6,312,516, and 6,328,795, the entire contents of which are incorporated herein by reference). Alternatively, a rapid cooled silicon (RCS) growth process can be used, with subsequent control of the thermal history of the crystal to rapidly cool at least a portion of the crystal to a target nucleation temperature and suppress the formation of agglomerated intrinsic point defects in that portion. One or both of these approaches also involve holding at least a portion of the grown crystal above the nucleation temperature for an extended period of time to reduce the concentration of intrinsic point defects before rapidly cooling that portion of the crystal to the target nucleation temperature, thereby substantially limiting or eliminating the formation of agglomerated intrinsic point defects therein (see, e.g., U.S. Patent Application Publication No. 2003 / 0196587, the entire disclosure of which is incorporated herein by reference). Furthermore, methods have been developed to simultaneously control the cooling rate of the solidified ingot and the radial variation of the axial temperature gradient near the interface (G) to reduce or eliminate agglomerated point defects from the center to the edge of the ingot (see, e.g., U.S. Patent No. 8,673,248, the entire disclosure of which is incorporated herein by reference).

[0018] This section is intended to introduce the reader to various aspects of art that may be related to various aspects of the present disclosure, which are described and / or claimed below. This discussion is believed to be helpful in providing the reader with background information to facilitate a better understanding of the various aspects of the present disclosure. Accordingly, it should be understood that these statements are not intended as admissions of prior art, and are to be read in this light. Summary of the Invention

[0019] One aspect of the present disclosure relates to a method for preparing a single crystal silicon ingot by the Czochralski method. The method includes the steps of adding an initial charge of polycrystalline silicon to a crucible in a growth chamber, the crucible having a bottom wall and a side wall, the growth chamber further having a bottom heater disposed adjacent the bottom wall of the crucible, a side heater disposed adjacent the side wall, and a reflector; applying power to the bottom heater, the side heater, or both the bottom and side heaters to heat the crucible containing the initial charge of polycrystalline silicon and form a silicon melt in the crucible, the power applied to the side heater being greater than the power applied to the bottom heater, and the silicon melt having a free melt height level; contacting a silicon seed crystal with the silicon melt in the crucible; pulling the silicon seed crystal from the silicon melt in a direction perpendicular to the melt height level at an initial pull rate to form a solid neck of a single crystal silicon ingot; and modifying the initial pull rate to achieve an outwardly flared seed cone pull rate to form a solid outwardly flared silicon melt adjacent the neck of the single crystal silicon ingot. and pulling a solid body of a monocrystalline silicon ingot adjacent to the flared seed cone from the silicon melt by modifying a pull rate of the flared seed cone to achieve a body pull rate, wherein the solid body of the monocrystalline silicon ingot has a radial diameter and an axial length, and surface tension forces generated as the solid body of the monocrystalline silicon ingot is pulled from the molten silicon form a melt-solid interface located above a free melt height level, and further a meniscus of molten silicon exists between the melt-solid interface and the free melt height level, wherein a cusp magnetic field is applied to the silicon melt during growth of the single crystal silicon ingot body, and an axial heat flux between the melt-solid interface and the free melt height level during growth of at least 40% of the total axial length of the solid body of the monocrystalline silicon ingot is at least about 20,000 W / m 2 has an absolute value of

[0020] Various refinements of the features described in connection with the above-described aspects of the present disclosure exist. Additional features may also be incorporated into the above-described aspects of the present disclosure as well. These refinements and additional features may exist individually or in any combination. For example, the various features discussed below in connection with any of the illustrated embodiments of the present disclosure may be incorporated into any of the above-described aspects of the present disclosure, alone or in any combination. [Brief explanation of the drawings]

[0021] [Figure 1A] The melt temperature profile near the crystal / melt interface for a typical hot zone is shown. [Figure 1B] 1 shows the melt temperature profile near the crystal / melt interface for a low-power hot zone. [Figure 2A] The melt temperature profile near the meniscus in a typical hot zone is shown, taken after 400 mm and 800 mm of growth. [Figure 2B] The temperature profile of the melt near the meniscus in the low-power hot zone is shown, taken after 400 mm and 800 mm of growth. [Figure 3] 1 shows a hot zone suitable for carrying out the method of the present invention. [Figure 4] 1A-1C are a series of diagrams illustrating crucible movement during batch Czochralski ingot growth. [Figure 5] FIG. 1 is a simplified diagram of a hot zone configuration suitable for use in the method of the present invention. [Figure 6] FIG. 2 is an explanatory diagram showing a meniscus curve. [Figure 7A] 1 is a graph illustrating relative crucible height (HR) protocols during ingot growth according to some embodiments of the method of the present invention. [Figure 7B]1 is a graph illustrating relative crucible height (HR) protocols during ingot growth according to some embodiments of the method of the present invention. [Figure 8A] 1 is a graph illustrating seed rotation speed protocols during ingot growth according to some embodiments of the method of the present invention. [Figure 8B] 1 is a graph illustrating seed rotation speed protocols during ingot growth according to some embodiments of the method of the present invention. [Figure 9] The positions of the magnetic coils suitable for generating a cusp magnetic field are shown. [Figure 10A] 1 shows the magnetic field strength and cusp position during conventional Czochralski crystal growth. [Figure 10B] 1 shows the magnetic field strength and cusp position during Czochralski crystal growth in the method of the present invention. [Figure 10C] 1 shows the magnetic field strength and cusp position during Czochralski crystal growth in the method of the present invention. [Figure 11A] Figure 1 shows a meniscus temperature profile according to an embodiment of the present invention. Temperature and heat flux data were obtained after 400 mm of growth and after 800 mm of growth. [Figure 11B] Figure 1 shows melt-side heat flux according to an embodiment of the present invention. Temperature and heat flux data were obtained after 400 mm of growth and after 800 mm of growth. [Figure 12A] Figure 1 shows a meniscus temperature profile according to an embodiment of the present invention. Temperature and heat flux data were obtained after 400 mm of growth and after 800 mm of growth. [Figure 12B] Figure 1 shows melt-side heat flux according to an embodiment of the present invention. Temperature and heat flux data were obtained after 400 mm of growth and after 800 mm of growth. [Figure 13A] FIG. 10 illustrates a meniscus temperature profile according to an embodiment of the present invention. [Figure 13B] 1 illustrates melt-side heat flux according to an embodiment of the present invention. [Figure 14A] 1 shows the diameter of a single crystal silicon ingot during bulk growth in a conventional low-power hot zone process. [Figure 14B] 1 illustrates the diameter of a single crystal silicon ingot during body growth in a process according to an embodiment of the present invention.

[0022] Corresponding reference characters indicate corresponding parts throughout the drawings. DETAILED DESCRIPTION OF THE INVENTION

[0023] According to the method of the present invention, a single crystal silicon ingot is grown by the Czochralski (CZ) method under process conditions suitable for suppressing or preventing distortion of the ingot during the latter half of its length (after 600 mm) in a low-power hot zone growth chamber. It has been observed that during ingot growth in a low-power hot zone growth chamber, the cross-section of the crystal changes from the initial desired circular shape to a star-like shape during the later growth of the single crystal silicon ingot's main body. Unfortunately, this distortion causes cameras monitoring the ingot's growth status to lose track of the shape of the growing ingot and meniscus. According to some embodiments, appropriate growth conditions are selected during ingot growth to suppress a decrease in melt temperature near the crystal / melt interface and meniscus, thereby reducing the possibility of overcooling and reducing distortion. That is, conditions are selected to minimize temperature changes at the free melt surface level, as well as the melt below the melt / crystal interface, during growth of the entire length of the ingot, thereby minimizing or eliminating distortion so that the ingot's cross-section maintains the desired circular shape along the entire length of the ingot.

[0024] The method of the present invention can be applied to a Czochralski puller that has an applied magnetic field, e.g., a cusp magnetic field, and a low-power hot zone (LPHZ) with added insulation at the bottom of the hot zone. Conventional hot zone pullers typically require high bottom heater power during growth. The low-power hot zone is an improvement over the conventional hot zone, where additional insulation is added to the bottom to prevent heat loss from the bottom, thereby reducing the required bottom heater power and contributing to reduced power consumption and manufacturing costs.

[0025] Crystal growth in a low-power hot zone is prone to ingot distortion in the latter half of the body length (after 600 mm), a problem not commonly seen in conventional hot-zone pullers with high-power bottom heater settings. Current understanding suggests that distortion in the latter half of the body length is due, at least in part, to the melt temperature near the melt / crystal interface. Figures 1A and 1B compare the thermal fields in the melt at the initial body length (400 mm) and later body length (800 mm) of ingots grown in a conventional hot zone (Figure 1A) and a low-power hot zone (Figure 1B). The figures show the temperatures in the melt near the crystal / melt interface. As shown in Figure 1B, these temperatures were significantly reduced in the low-power hot zone at the end of the body length, whereas the temperature difference was much smaller in the conventional hot zone (Figure 1A). For illustrative purposes, the contour at T = 1690 K is highlighted in each thermal field for comparison between Figures 1A and 1B.

[0026] Figures 2A (conventional hot zone) and 2B (low-power hot zone) show the temperature profiles along the meniscus curve at the beginning and end of the body length for both hot zone types. As shown in Figure 2A, for the conventional hot zone, the meniscus edge temperature at the end of the body length of 800 mm, i.e., the temperature at the melt free interface, was approximately 1694 K (no distortion was observed at this stage), which is very close to the temperature of the meniscus at the beginning of the body length, i.e., 1694.4 K at the melt free interface of the meniscus at 400 mm. See Figure 6 for a depiction of the meniscus curve. The distance 0 mm in Figure 2A corresponds to the solid-melt interface, and the distance 35 mm corresponds to the melt free interface. These distances may vary depending on the hot zone design and pulling conditions. As depicted in Figure 2B, during crystal growth in the low-power hot zone, the meniscus temperature at the melt free interface at 800 mm of growth decreased to a value of 1689.9 K, a significant difference from the temperature of 1691.4 K at 400 mm of body length. Figures 2A and 2B also reveal that the temperature gradient at the interface as well as at the meniscus in the low-power hot zone decreases from 400 mm to 800 mm of body length. These changes may be responsible for the higher distortion at 800 mm of body length in the low-power hot zone configuration.

[0027] According to the method of the present invention, hot zone conditions are selected to increase the melt-side heat flux and gradient along the meniscus during growth of the entire body length of the single crystal silicon ingot. Suitable process conditions selected according to some embodiments of the present invention include bottom heater power, relative crucible height, seed rotation speed, crucible rotation rate, and magnetic field strength, and suitable conditions have been found. Certain variables, including relative crucible height (RH), seed rotation speed, bottom heater power, and magnetic field position, have been found to increase the melt-side heat flux, thereby increasing the temperature near the melt and the temperature gradient along the meniscus. Furthermore, the temperature profile achieved by the method of the present invention resulted in ingots grown with reduced or eliminated strain throughout the entire body length of the ingot.

[0028] FIG. 3 illustrates a hot zone suitable for carrying out the method of the present invention. The hot zone configuration includes a quartz crucible 10 having a diameter suitable for holding a silicon melt and pulling an ingot having a diameter of 450 mm or more. A graphite crucible 20 surrounds and supports the quartz crucible 10. Other configurations for holding the quartz crucible 10 are within the scope of the method of the present invention, such as a configuration without a graphite crucible 20. The hot zone configuration includes a side heater 30 located near the side wall of the crucible 10 and, optionally, a bottom heater 40 located below the crucible 10. The hot zone configuration includes insulation 50 configured to retain heat within the hot zone. A low-power hot zone according to the present invention includes additional insulation 50 near the bottom of the crucible 10. The diameter of the growing crystal and the shape and height of the meniscus are monitored by a camera 60 (e.g., a CCD camera) positioned in a top window 70. Data obtained from the cameras provides feedback to the side heaters 30 and bottom heater 40. During crystal growth, the power distribution between the heaters may be adjusted to allow for uniformity of the melt / solid interface, i.e., to maintain the desired meniscus shape and height. A heat shield or reflector 80 reflects heat flux from the hot section of the furnace, including the heaters 30, 40 and crucible 10, to the melt 90. The reflector 80 reduces heat transfer from the hot section to the cold section of the furnace, thereby maintaining separation between these two regions of the furnace. The reflector 80 helps control the axial and radial temperature gradients that promote solidification and crystallization of the molten silicon 90 into the growing ingot 100.

[0029] Referring to FIG. 3, the Czochralski process begins with loading polycrystalline silicon into a quartz crucible 10. The solid polysilicon added to the crucible 10 is typically granular polysilicon, although chunk polysilicon may also be used and is delivered to the crucible using a polysilicon delivery system optimized for use with granular polysilicon. Chunk polysilicon typically has a size of 3 to 45 millimeters (e.g., largest dimension), while granular polysilicon typically has a size of 400 to 1400 microns. Granular polysilicon has the advantage of being small in size, allowing for easy and accurate control of the delivery rate. However, granular polysilicon requires manufacturing methods such as chemical vapor deposition, which results in higher costs compared to chunk polysilicon. Bulk polysilicon has the advantages of lower cost and larger size, allowing for larger delivery rates.

[0030] Typically, the melt 90 from which the ingot 100 is pulled is formed by loading polycrystalline silicon into a crucible 10 to form an initial charge of silicon. Typically, the initial charge is between about 100 kilograms and about 1000 kilograms, or between about 100 kilograms and about 800 kilograms, or between about 100 kilograms and about 500 kilograms of polycrystalline silicon, and may be granular, chunk, or a combination of granular and chunk. The mass of the initial charge depends on the desired crystal diameter and HZ design. In some embodiments, the initial polycrystalline silicon charge is sufficient to grow one single crystal silicon ingot, i.e., batchwise. Typically, the total axial length of the solid body of the single crystal silicon ingot is at least about 1100 mm, e.g., between about 1200 mm and about 1300 mm, e.g., between about 1200 mm and about 1250 mm. In the continuous Czochralski process, polycrystalline silicon is continuously supplied during crystal growth, so the initial charge does not reflect the crystal length. Therefore, the initial charge amount can be smaller, such as between about 100 kg and about 200 kg. Furthermore, with a continuous supply of polycrystalline silicon and sufficient chamber height, the crystal length can be extended to 2000 mm, 3000 mm, or even 4000 mm. Various polycrystalline silicon sources can be used, including, for example, granular polycrystalline silicon produced by pyrolysis of silane or halosilane in a fluidized-bed reactor or polycrystalline silicon produced in a Siemens reactor. Once the polycrystalline silicon is added to the crucible to form a charge, the charge is heated to a temperature above the melting point of silicon (e.g., about 1412°C) to melt the charge, thereby forming a silicon melt consisting of molten silicon. The silicon melt has an initial volume of molten silicon and an initial melt height level, both of which are determined by the size of the initial charge. In some embodiments, the crucible containing the silicon melt is heated to a temperature of at least about 1425° C., at least about 1450° C., or even at least about 1500° C. The initial polycrystalline silicon charge is heated by supplying power to the bottom heater 40, the side heater 30, or both the bottom heater 40 and the side heater 30.According to some embodiments, the power supplied to the side heater 30 is greater than the power supplied to the bottom heater 40. In some embodiments, the bottom heater power 40 is 2 kW or less, such as 1 kW or less. In some embodiments, the bottom heater 40 has no power supplied to it, such that the hot zone configuration is a low-power hot zone configuration, i.e., the power supplied is 0 kW. A low-power hot zone according to the present invention includes additional insulation 50 near the bottom of the crucible 10.

[0031] Referring to FIG. 4 , as the solid polycrystalline silicon charge 150 liquefies to form the silicon melt 120 of molten silicon, the silicon seed crystal 160 is lowered into contact with the melt. The silicon seed crystal 160, with the silicon attached, is then pulled out of the melt, thereby forming a melt-solid interface near or at the surface of the melt. Generally, the initial pull rate to form the neck is fast. In some embodiments, the silicon seed crystal and neck are pulled at a neck pull rate of at least about 1.0 mm / min, e.g., between about 1.5 mm / min and about 6 mm / min, e.g., between about 3 mm / min and about 5 mm / min. In some embodiments, the silicon seed crystal and the crucible are rotated in opposite directions, i.e., counter-rotated. Counter-rotation facilitates convection of the silicon melt. Crystal rotation is primarily used to provide a symmetrical temperature profile, reduce the angular distribution of impurities, and control the shape of the crystal-melt interface. In some embodiments, the silicon seed rotates at a speed of between about 5 rpm and about 30 rpm, or between about 5 rpm and about 20 rpm, or between about 5 rpm and about 15 rpm, e.g., about 8 rpm, about 9 rpm, or about 10 rpm. In some embodiments, the seed rotation speed may vary during growth of the main body of the single crystal silicon ingot. In some embodiments, the crucible rotates at a speed of between about 0.5 rpm and about 10 rpm, or between about 1 rpm and about 10 rpm, or between about 4 rpm and about 10 rpm, or between about 5 rpm and about 10 rpm. In some embodiments, the seed rotates at a faster speed than the crucible. In some embodiments, the seed rotates at a speed at least 1 rpm faster than the rotation speed of the crucible, e.g., at least about 3 rpm faster, or at least about 5 rpm faster. Typically, the neck has a length of between about 300 millimeters and about 700 millimeters, e.g., between about 450 millimeters and about 550 millimeters. However, the neck length may vary outside these ranges.

[0032] After the neck is formed, referring to FIG. 4 , an outwardly flared seed cone portion 170 adjacent to the neck is grown. Typically, the pull rate is reduced from the neck pull rate to a rate suitable for growing the outwardly flared seed cone portion. For example, the seed cone pull rate during growth of the outwardly flared seed cone 170 is between about 0.5 mm / min and about 2.0 mm / min, e.g., about 1.0 mm / min. In some embodiments, the outwardly flared seed cone 170 has a length between about 100 mm and about 400 mm, e.g., between about 150 mm and about 250 mm. The length of the outwardly flared seed cone 170 may vary outside these ranges. In some embodiments, the outwardly flared seed cone 170 grows to a terminal diameter of about 150 mm, at least about 150 mm, about 200 mm, at least about 200 mm, about 300 mm, about 450 mm, or even at least about 450 mm. The terminal diameter of the flared seed cone 170 is generally equal to the constant diameter of the main ingot body 180 of the single crystal silicon ingot.

[0033] After the formation of the neck portion and the outwardly flared seed cone 170 adjacent to the neck portion, a main ingot body 180 is then grown having a constant diameter adjacent to the outwardly flared seed cone 170. The constant diameter portion of the main ingot body 180 has a periphery, a central axis parallel to the periphery, and a radius extending from the central axis toward the periphery. The central axis also passes through the cone portion and the neck portion. The diameter of the main ingot body 180 may vary, and in some embodiments, the diameter may be about 150 millimeters, at least about 150 millimeters, about 200 millimeters, at least about 200 millimeters, about 300 millimeters, at least about 300 millimeters, about 450 millimeters, or even at least about 450 millimeters. Stated another way, the radial length of the solid main ingot body 180 of the single crystal silicon ingot is about 75 millimeters, at least about 75 millimeters, about 100 millimeters, at least about 100 millimeters, about 150 millimeters, at least about 150 millimeters, about 225 millimeters, or even at least about 225 millimeters. The single crystal silicon ingot main ingot body 180 is eventually grown to a length of at least about 1000 millimeters, such as at least 1200 millimeters, such as at least 1250 millimeters, such as at least 1400 millimeters, such as at least 1500 millimeters, or at least 2000 millimeters, or at least 2200 millimeters, or at least about 3000 millimeters, or at least about 4000 millimeters. In some preferred embodiments, the total axial length of the solid main ingot body 180 of the single crystal silicon ingot is at least about 1100 millimeters, such as between about 1200 millimeters and about 1300 millimeters, such as between about 1200 millimeters and about 1250 millimeters.

[0034] In some embodiments, the main ingot body 180 may be pulled according to a pull rate protocol. The pull rate may be reduced from a relatively fast pull rate to a minimum pull rate and then increased to a constant pull rate for a significant portion of the growth of the single crystal silicon ingot body. The initial fast pull rate may be between about 0.5 mm / min and about 2.0 mm / min, e.g., about 1.0 mm / min, and then reduced to a slower pull rate, such as about 0.4 mm / min or even about 0.3 mm / min, before increasing to a constant pull rate of between about 0.4 mm / min and about 0.8 mm / min, between about 0.4 mm / min and about 0.7 mm / min, or between about 0.4 mm / min and about 0.65 mm / min.

[0035] In the continuous Czochralski process, polycrystalline silicon, i.e., granular, chunk, or a combination of granular and chunk, is added to the molten silicon during the growth of the main ingot body 180 of the single crystal silicon ingot, thereby achieving a constant molten silicon volume and a constant melt height level. According to the method of the present invention, by maintaining a substantially constant melt volume during the growth of a substantial portion of the axial length of the single crystal silicon ingot body, high ingot quality can be achieved at a constant pull rate along a substantial portion of the axial length of the single crystal silicon ingot body. The constant melt volume regardless of the length of the crystal allows the crystal / melt interface to remain constant, thereby maintaining uniform crystal quality along a substantial portion of the ingot body. Thus, in some embodiments, the volume of the molten silicon varies by no more than about 1.0% by volume during the growth of at least about 90% of the body of the single crystal silicon ingot, or by no more than about 0.5% by volume during the growth of at least about 90% of the body of the single crystal silicon ingot, or even by no more than about 0.1% by volume during the growth of at least about 90% of the body of the single crystal silicon ingot. Stated differently, in some embodiments, the melt height level varies by less than about ±0.5 millimeters during the growth of at least about 90% of the body of the single crystal silicon ingot.

[0036] In the batch Czochralski process, an initial charge of polycrystalline silicon is sufficient to grow the entire length of the ingot. Rather than maintaining a constant melt height level, the volume of silicon melt decreases as the ingot grows. Thus, referring to FIG. 4 (inset), a crucible 110 holds molten silicon 120 from which a crystalline ingot 140 is pulled. As shown in sections a) through j) of FIG. 4, an initial charge of solid polycrystalline silicon 150 is melted by application of heat from a heater 130 to form the melt 120. A seed crystal 160 is contacted with the molten silicon 120 and slowly extracted to grow a single crystal ingot 140. As can be seen, as the length of the single crystal silicon ingot 140 increases, the volume of the molten silicon 120 decreases, requiring the crucible 110 to be moved up and down in the same direction as the ingot is being pulled.

[0037] Regardless of whether the method is batch or continuous, growth conditions are selected to achieve optimized melt-side heat flux, temperature near the melt, and temperature gradient near the meniscus to grow an ingot with reduced or eliminated strain along the entire length of the ingot body. Among these growth conditions, referring to FIG. 5, is monitoring the relative height of the bottom of the reflector 200 above the surface of the melt 210. The relative height "HR" can be measured directly as the distance between the bottom of the reflector 200 and the melt level 210, as shown in FIG. 3, as the vector labeled "HR." If the hot zone configuration does not allow for direct measurement, i.e., if the camera 220 cannot identify the bottom of the reflector 200, the relative height "HR" can be measured indirectly, for example, using a reference point located below the crucible. The distance between the melt level and reference point 230 is measured, and the distance between the bottom of the reflector 200 and reference point 230, labeled "RZ," is measured. The relative height, "HR," is calculated by subtracting the distance between the melt level 210 and the reference point 230 from "RZ." In a batch process, the melt level changes as the ingot grows; therefore, the crucible must be moved, i.e., raised, to maintain the desired "HR."

[0038] According to some embodiments, the "HR" begins at a relatively high value, such as between about 60 mm and 120 mm, or between about 70 mm and 100 mm. In some embodiments, the relatively high value of HR occurs during the growth of the neck and crown of the monocrystalline silicon ingot, i.e., the outward-spreading seed cone. The HR distance is initially large so that the camera can capture the meniscus during the neck and crown stages. The HR value is initially large to keep the meniscus within the viewing window of the camera tracking the meniscus. In some embodiments, the relatively high value of HR may continue during the growth of the initial portion of the ingot body, for example, during the first 200 mm to 400 mm of the ingot body. In some embodiments, after the growth of the neck and crown, the distance between the bottom of the reflector and the melt level is rapidly reduced by moving the crucible closer to the bottom of the reflector. In some embodiments, the distance between the bottom of the reflector and the melt level decreases at a rate of at least −0.05 millimeters per millimeter of ingot growth, or at least about −0.06 millimeters per millimeter of ingot growth, e.g., about −0.065 millimeters per millimeter of ingot growth, preferably less than about −0.1 millimeters per millimeter of ingot growth, or less than about −0.08 millimeters per millimeter of ingot growth. The values ​​are listed as negative because the relative height decreases from a higher value to a lower value. In some embodiments, the HR is between 40 mm and 50 mm, e.g., between 45 mm and 50 mm, or between 45 mm and 48 mm, during the growth of a significant portion of the ingot's main body. In some embodiments, the HR is 45 mm during the growth of a significant portion of the ingot's main body. In some embodiments, the HR is 47 mm during the growth of the majority of the ingot's main body. These HR distances apply to growth of at least about 50%, at least about 60%, or at least about 70% of the length of the main body of the ingot. "HR" profiles according to exemplary embodiments of the present invention are shown in Figures 7A and 7B. HR values ​​according to the present invention alter the slope of the crystal edge, which in turn controls the defect profile within the crystal.Essentially, the defect profile determines to what value HR should be ramped and when it should begin. For example, in the initial main body, the crystal is primarily heated from the sides, so it is initially pulled at a slow rate to achieve the desired diameter, and then the seed drift is ramped. This seed drift varies with the length of the main body, so a temperature gradient must be varied to control defects, which is controlled by varying HR during crystal growth. As shown there, HR is maintained at a constant low value for most of the growth of the main body of the ingot, then rapidly increases as the ingot is completed and pulled from the remaining melt.

[0039] In some embodiments, a low HR is combined with a low seed rotation speed during the growth of a significant length of the single crystal silicon ingot body. In some embodiments, the seed rotation speed may begin at a high initial speed. Because melt temperatures and heat fluxes are typically high during the growth of the first 200 to 600 mm of the body length, a high seed rotation speed is appropriate for growing the initial portion of the ingot. In some embodiments, the initial seed rotation speed may be at least about 10 rpm, e.g., at least about 11 rpm, or at least about 12 rpm. In some preferred embodiments, the initial seed rotation speed may be at least about 11 rpm. After growing the initial portion of the ingot, the seed rotation speed is decreased. In some embodiments, the seed rotation speed may be decreased from about −0.005 rpm per millimeter of ingot growth to about −0.020 rpm per millimeter of ingot growth, e.g., from about −0.005 rpm per millimeter of ingot growth to about −0.014 rpm per millimeter of ingot growth. Because the seed rotation speed decreases from a high value to a low value, the values ​​are listed as negative. When the seed rotation rate is reduced within this range, the seed rotation rate can be between about 5 rpm and about 10 rpm, e.g., about 8 rpm, about 9 rpm, or about 10 rpm, for ingot body lengths between about 600 mm and about 900 mm. In some preferred embodiments, the seed rotation rate can be about 9 rpm for ingot body lengths between about 600 mm and about 900 mm. In some embodiments, the rotation rate can be further reduced to between about 5 rpm and about 8 rpm, e.g., about 7 rpm or about 8 rpm, during later growth of the main body, such as after about 1200 mm of growth. In some preferred embodiments, the rotation rate can be about 7 rpm during later growth of the main body, such as after about 1200 mm of growth. In some preferred embodiments, the rotation rate can be about 8 rpm during later growth of the main body, such as after about 1200 mm of growth. Selecting these seed rotation values ​​can increase the temperature gradient in the melt region near the solid-liquid interface and may also increase the temperature of the meniscus. Very low seed rotation values ​​can result in poor oxygen radial gradients.Seed rotation speed profiles according to exemplary embodiments of the present invention are shown in Figures 8A and 8B.

[0040] Furthermore, according to the process of the present invention, a magnetic field may be applied to the crucible containing the silicon melt. Either a cusp magnetic field or a horizontal magnetic field may be applied to set the appropriate crystal / melt interface, i.e., the meniscus, shape and height. The magnetic field is primarily used to fix the desired crystal / melt interface shape and height, with control of the oxygen content (Oi) being a secondary objective.

[0041] Applying a magnetic field to the silicon melt during growth of the monocrystalline silicon ingot body can control the melt flow and the shape of the melt / solid interface, improving ingot quality. In some embodiments, the applied magnetic field maintains a substantially constant melt / solid interface profile for at least about 70% of the growth of the monocrystalline silicon ingot body, or for about 70% to about 90% of the growth of the monocrystalline silicon ingot body. The magnetic field exerts an electromagnetic force that affects the flow of the silicon melt, thereby affecting heat transfer in the melt, changing the shape of the crystal / melt interface and the temperature of the growing crystal.

[0042] The magnetic field affects the oxygen content and uniformity within the ingot. The oxygen sources within the ingot are the dissolution of the quartz crucible wall, evaporation of SiOx(g) at the free melt surface (controlled by the melt flow kinetics), and incorporation into the crystal front during growth. The magnetic field affects the convective melt flow during growth, which in turn affects oxygen evaporation and incorporation. The time evolution of the oxygen incorporation into a single crystal silicon ingot is controlled by the diffusion and convection of oxygen in the melt, which is expressed by the following equation: TIFF0007672989000001.tif1050

[0043] where C is the oxygen concentration of the solidified silicon, t is time, v is the convection velocity (melt flow rate), ρ is the density of the silicon melt, and V is the gradient (d / dx). The applied magnetic field affects the melt velocity (v) and the oxygen concentration gradient in the melt (dC / dx = VC). The magnetic field forces the melt flow to a steady state, resulting in a time-constant oxygen uptake (Oi) into the ingot, improving the radial and axial oxygen concentration uniformity. The SOURCE term is derived from two parameters: the dissolution of the quartz (SiO2) crucible (Si(l) + SiO2(s) → SiOx(g)), which is the evolution of oxygen, and evaporation, which is the removal (loss) of oxygen (SiOx(g)) from the melt. In the batch Cz process, this SOURCE term is not constant; instead, it depends on the length of the crystal because the mass of the melt decreases as the crystal grows. As the ingot grows to a significant portion of its length, the remaining melt volume decreases, resulting in a decrease in the amount of silicon melt in contact with the crucible, resulting in a decrease in the concentration of oxygen absorbed into the melt from the crucible. Therefore, assuming other conditions (diffusion, convection, and evaporation) are constant, less oxygen is absorbed into the solidifying silicon crystal. The area of ​​the melt's free surface (the interface between the melt and the gas) affects the rate of SiOx(g) evaporation. Less SiOx(g) evaporation means more oxygen is absorbed into the silicon crystal during solidification. According to the method of the present invention, polysilicon is added as the crystal ingot grows, so the melt mass remains constant. Therefore, all source terms (oxygen generation from the dissolution of SiO2 in the melt from the crucible and evaporation of SiOx(g) gas from the free melt surface) remain constant. Therefore, the diffusion and convection terms affect the oxygen content of the solidifying silicon crystal. The applied magnetic field makes the melt flow more stable (i.e., the melt flow is constant, like a time-independent steady state), so that oxygen uptake is uniform and stable axially and radially throughout the growth of the ingot. In some embodiments, interstitial oxygen can be incorporated into the ingot at a concentration between about 4 PPMA and about 18 PPMA.In some embodiments, interstitial oxygen may be incorporated into the ingot at a concentration between about 10 PPMA and about 35 PPMA. In some embodiments, the ingot contains oxygen at a concentration not exceeding about 15 PPMA, or not exceeding about 10 PPMA. Interstitial oxygen can be measured according to SEMI MF 1188-1105.

[0044] In some embodiments, a horizontal magnetic field is applied to the silicon melt during growth of the main body of a single crystal silicon ingot. Crystal growth in the presence of a horizontal magnetic field is achieved by placing the crucible holding the silicon melt between the poles of a conventional electromagnet. In some embodiments, the horizontal magnetic field may have a magnetic flux density of between about 0.2 Tesla and about 0.4 Tesla in the melt region. The magnetic field fluctuations in the melt are less than about ±0.03 Tesla at a given strength. The application of the horizontal magnetic field generates a Lorentz force in the axial direction, opposite to the fluid movement, which opposes the forces promoting melt convection. As a result, melt convection is suppressed and the axial temperature gradient in the crystal near the interface increases. The melt-crystal interface then moves upward toward the crystal in response to the increased axial temperature gradient in the crystal near the interface, reducing the contribution from melt convection within the crucible.

[0045] In some embodiments, a cusp magnetic field is applied to the silicon melt during growth of the main body of a single-crystal silicon ingot. Suitable magnetic coil positions for achieving a cusp magnetic field are shown in FIG. 9. The cusp magnetic field has two control parameters: magnetic flux density and field shape. The cusp magnetic field combines a horizontal (radial) magnetic field component at the immediate surface of the melt with a vertical (axial) magnetic field deeper within the melt near the axis of the ingot. This cusp magnetic field is generated using a pair of Helmholtz coils 300, 310 carrying currents in opposite directions. As a result, at the midpoint between the two magnetic fields, in the vertical direction along the ingot axis, the magnetic fields cancel each other, resulting in a zero or near-zero vertical magnetic field component. For example, the magnetic flux density of the cusp is typically about 0 to about 0.2 Tesla in the axial direction. Furthermore, the radial magnetic flux density is typically higher than the vertical magnetic flux density. For example, the cusp magnetic flux density is typically between about 0 and about 0.6 Tesla, e.g., between about 0.2 and about 0.5 Tesla, depending on the radial position. The radial cusp magnetic field suppresses the melt flow, thereby stabilizing the melt. In other words, applying a radial cusp magnetic field induces convection near the solid-liquid interface where crystal growth occurs and suppresses convection in the rest of the melt, providing an effective means for achieving uniform oxygen distribution. Furthermore, the cusp magnetic fields at the free surface of the melt and the melt-crucible interface can independently control the thermal convection in the melt locally. This allows the oxygen concentration in the growing crystal to be controlled solely by the magnetic flux density, regardless of the crystal rotation speed. When an axial or radial magnetic field is present, control of the oxygen concentration is achieved by controlling the crystal rotation speed. Applying a cusp magnetic field can grow ingots with lower oxygen contents than ingots grown without a magnetic field, e.g., less than about 15 ppma, or less than about 10 ppma. Interstitial oxygen may be measured according to SEMI MF 1188-1105.

[0046] According to the method of the present invention, the cusp magnetic field applied to the silicon melt during growth of the body of the single crystal silicon ingot is derived from upper magnetic coil 300 and lower magnetic coil 310, with the upper magnetic field strength derived from upper magnetic coil 300 being greater than the lower magnetic field strength derived from lower magnetic coil 310. In some embodiments, the upper magnetic field strength derived from upper magnetic coil 300 exceeds the lower magnetic field strength derived from lower magnetic coil 310 by at least 10%, or at least 15%. By applying a greater magnetic field strength to the upper magnetic coil compared to the lower magnetic coil, the cusp position can be moved lower in the melt, as shown in FIGS. 10B and 10C, compared to conventional methods such as that shown in FIG. 10A. Lowering the cusp position has been found to increase the temperature in the meniscus region.

[0047] According to the method of the present invention, the conditions disclosed herein minimize temperature changes not only in the melt below the melt / crystal interface but also at the surface level of the free melt in the later body, since no distortion is observed in the initial body. The conditions of the method of the present invention help prevent a decrease in the melt temperature near the crystal / melt interface and meniscus, thereby reducing the possibility of undercooling and the resulting distortion. Favorable factors for achieving the desired results are the melt-side heat flux, the temperature near the melt, and the temperature gradient near the meniscus, which can be increased by selecting the HR, seed rotation, and magnetic cusp conditions.

[0048] The melt side heat flux is calculated using the following formula: TIFF0007672989000002.tif1547 where k m is the thermal conductivity of the melt, (dT / dx) mis the axial temperature gradient. Heat flux depends on the axial temperature gradient of the melt. Thus, a higher heat flux results in a higher axial temperature gradient, which means a higher temperature in the melt below the melt / crystal interface, a higher temperature gradient near the meniscus, and reduced distortion of the ingot along the axial length of the ingot. According to some embodiments, the axial heat flux between the growing melt-solid interface and the free melt height level for at least 40% of the total axial length of the solid body of the single crystal silicon ingot is at least about 20,000 W / m² over at least about 85% of the radial length of the solid body of the single crystal silicon ingot. 2 , at least about 21,000 W / m 2 , at least about 22,000 W / m 2 , at least about 23,000 W / m 2 , or at least about 24,000 W / m 2 According to some embodiments, during growth of at least 60% of the axial length of the solid body of the single crystal silicon ingot, the axial heat flux between the melt-solid interface and the free melt height level is at least about 20,000 W / m over at least about 85% of the radial length of the solid body of the single crystal silicon ingot. 2 , at least about 21,000 W / m 2 , at least about 22,000 W / m 2 , at least about 23,000 W / m 2 , or at least about 24,000 W / m 2 According to some embodiments, the axial heat flux between the growing melt-solid interface and the free melt height level over at least 80% of the axial length of the solid main body of the single crystal silicon ingot has an absolute value of at least about 20,000 W / m across at least about 80% of the diameter of the solid main body of the single crystal silicon ingot. 2 , at least about 21,000 W / m 2 , at least about 22,000 W / m 2 , at least about 23,000 W / m 2 , or at least about 24,000 W / m 2According to some embodiments, the axial heat flux between the growing melt-solid interface and the free melt height level over at least 90% of the total length of the solid main body of the single crystal silicon ingot has an absolute value of at least about 20,000 W / m over at least about 85% of the radial length of the solid main body of the single crystal silicon ingot. 2 , at least about 21,000 W / m 2 , at least about 22,000 W / m 2 , at least about 23,000 W / m 2 , or at least about 24,000 W / m 2 According to some embodiments, the axial heat flux between the growing melt-solid interface and the free melt height level over at least 95% of the total length of the solid body of the single crystal silicon ingot has an absolute value of at least about 20,000 W / m over at least about 85% of the radial length of the solid body of the single crystal silicon ingot. 2 , at least about 21,000 W / m 2 , at least about 22,000 W / m 2 , at least about 23,000 W / m 2 , or at least about 24,000 W / m 2 has an absolute value of

[0049] Furthermore, the conditions are selected to obtain a temperature gradient along the meniscus curve. See FIG. 6 . According to some embodiments, during the growth of at least 40% of the total length of the solid body of the single crystal silicon ingot, the temperature gradient along the meniscus curve between the melt-solid interface and the height level of the free melt has an average value of at least about 0.16° / mm or at least about 0.18° / mm. According to some embodiments, during the growth of at least 60% of the total length of the solid body of the single crystal silicon ingot, the temperature gradient along the meniscus curve between the melt-solid interface and the height level of the free melt has an average value of at least about 0.16° / mm or at least about 0.18° / mm. According to some embodiments, during the growth of at least 80% of the total length of the solid body of the single crystal silicon ingot, the temperature gradient along the meniscus curve between the melt-solid interface and the height level of the free melt has an average value of at least about 0.16° / mm or at least about 0.18° / mm. According to some embodiments, during growth of at least 90% of the total length of the solid body of the single crystal silicon ingot, the temperature gradient along the meniscus curve between the melt-solid interface and the height level of the free melt has an average value of at least about 0.16° / mm or at least about 0.18° / mm.

[0050] By achieving these melt fluxes and temperature gradients, the temperature of the molten silicon at the meniscus is higher than can conventionally be achieved with a low-power hot zone. According to some embodiments, the temperature of the molten silicon at the meniscus is at least 1691 K, or at least 1692 K, as measured at the free melt height level during growth of at least 40% of the total length of the solid body of the single crystal silicon ingot. According to some embodiments, the temperature of the molten silicon in the meniscus is at least 1691 K, or at least 1692 K, as measured at the free melt height level during growth of at least 60% of the total length of the solid body of the single crystal silicon ingot. According to some embodiments, the temperature of the molten silicon in the meniscus is at least 1691 K, or at least 1692 K, as measured at the free melt height level during growth of at least 80% of the total length of the solid body of the single crystal silicon ingot. According to some embodiments, the temperature of the molten silicon in the meniscus is at least 1691 K, or at least 1692 K, measured at the free melt height level during growth of at least 85% of the total length of the solid body of the single crystal silicon ingot. According to some embodiments, the temperature of the molten silicon in the meniscus is at least 1691 K, or at least 1692 K, measured at the free melt height level during growth of at least 90% of the total length of the solid body of the single crystal silicon ingot.

[0051] In some embodiments, suitable process conditions for achieving the melt-side meniscus temperature profile and melt flux depicted in Figures 11A and 11B include bottom heater power (BH) = 0 kW, seed rotation speed (SR) = 9 rpm, and relative height (HR) = 47 mm (Test Condition 1). The heat flux (Qmelt) value is a measure of the heat flux entering the melt across the crystallization front. Negative values ​​indicate heat transfer from the melt to the crystal through the melt / solid interface. As shown in Figure 11A, the meniscus temperature is at least 1692 K along the meniscus curve for both 400 mm and 800 mm of axial growth. Furthermore, as shown in Figure 11B, the heat flux is at least 20,000 W / m² across the radial length of the interface for both 400 mm and 800 mm of axial growth. 2 has a larger absolute value than

[0052] In some embodiments, suitable process conditions for achieving the melt-side meniscus temperature profile and melt flux depicted in Figures 12A and 12B include bottom heater power (BH) = 0 kW, seed rotation speed (SR) = 9 rpm, and relative height (HR) = 45 mm (Test Condition 1). The heat flux (Qmelt) value is a measure of the heat flux entering the melt across the crystallization front. Negative values ​​indicate heat transfer from the melt to the crystal through the melt / solid interface. As shown in Figure 12A, the meniscus temperature is at least 1692 K along the meniscus curve for both 400 mm and 800 mm of axial growth. Furthermore, as shown in Figure 12B, the heat flux is at least 20,000 W / m² across the radial length of the interface for both 400 mm and 800 mm of axial growth. 2 has a larger absolute value than

[0053] In some embodiments, lowering the magnetic cusp position was found to increase the temperature of the meniscus region. To lower the magnetic cusp position, a 15% gap was added between the upper and lower magnets, and the upper magnets were made stronger. A comparison of the meniscus region temperature profile and melt-side heat flux was performed as shown in Figures 13A and 13B below. The lower temperature and gradient shown in Figure 13A pertains to a conventional low-power hot zone (conventional LPHZ) with the magnetic cusp position illustrated in Figure 10A. The higher temperature and gradient shown in Figure 13A pertains to hot zones according to embodiments of the present invention (Test Condition 1 and Test Condition 2) with the cusp positions illustrated in Figures 10B and 10C. Thus, the magnetic field position of the cusp, along with the seed rotation speed and relative height, may be combined to increase the meniscus temperature profile.

[0054] The process of the present invention enables the growth of monocrystalline silicon ingots with minimized deviation from the set diameter. By minimizing / reducing distortion, the actual diameter profile of the grown crystal is closer to the reference point value. Referring to FIG. 14A, a conventional low-power hot-zone process can result in significant deviation from the desired crystal diameter. The set diameter is represented by a solid line, and FIG. 14A shows that the actual diameter can significantly deviate from the set diameter. Referring to FIG. 14B, although some oscillations and deviations may still occur, the process of the present invention allows for the growth of crystals whose diameter values ​​remain close to the set value. From this perspective, the process of the present invention results in crystals with diameters that vary little around the set value, thereby reducing distortion of the ingot throughout the entire length of the ingot body.

[0055] As used herein, the terms "about," "substantially," "essentially," and "about" when used in connection with a range of dimensions, concentrations, temperatures, or other physical or chemical properties or characteristics are meant to cover variations that may exist at the upper and / or lower limits of the range of the property or characteristic, including, for example, variations due to rounding, measurement method, or other statistical variations.

[0056] When introducing elements of the disclosure or embodiments thereof, the articles "a," "an," "the," and "said" are intended to mean that there are one or more elements. The terms "comprising," "including," "containing," and "having" are intended to be inclusive and mean that there may be additional elements other than the listed elements. The use of specific orientation terms (e.g., "top," "bottom," "side," etc.) is for convenience of description and does not require a particular orientation of the described items.

[0057] Because various changes can be made in the structures and methods described above without departing from the scope of the present disclosure, it is intended that all matter contained in the above description and shown in the accompanying drawings be interpreted as illustrative and not in a limiting sense.

Claims

1. 1. A method for preparing a single crystal silicon ingot by the Czochralski process, comprising: adding an initial charge of polycrystalline silicon to a crucible within a growth chamber, the crucible having a bottom wall and a side wall, the growth chamber further having a bottom heater disposed adjacent the bottom wall of the crucible, a side heater disposed adjacent the side wall, and a reflector; applying power to a bottom heater, a side heater, or both the bottom heater and the side heater to heat a crucible containing an initial charge of polycrystalline silicon and form a silicon melt in the crucible, wherein the power applied to the side heater is greater than the power applied to the bottom heater, the power applied to the bottom heater being 2 kW or less, the crucible rotating at a speed between about 0.5 rpm and about 10 rpm, and the silicon melt having a free melt height level; contacting a silicon seed crystal with the silicon melt in the crucible; pulling the silicon seed crystal from the silicon melt in a direction perpendicular to the height level of the free melt at an initial pull rate of about 1.5 mm / min to about 6 mm / min to form a solid neck of a single crystal silicon ingot, wherein during growth of the solid neck, a relative height between the height level of the free melt and a bottom of a reflector is about 60 mm to about 120 mm, and further wherein during growth of the solid neck, the silicon seed crystal is rotated at a speed of about 5 rpm to about 30 rpm; pulling a solid outwardly extending seed cone from the silicon melt adjacent to a solid neck of the single crystal silicon ingot by modifying an initial pull rate to about 0.5 mm / min to about 2.0 mm / min to achieve a pull rate of the outwardly extending seed cone, wherein a relative height between the height level of the free melt and the bottom of the reflector during growth of the solid outwardly extending seed cone is about 60 mm to about 120 mm; pulling a solid body of the single crystal silicon ingot adjacent the solid outwardly flared seed cone out of the silicon melt by modifying a pull rate of the outwardly flared seed cone to achieve a body pull rate of about 0.4 mm / min to about 0.8 mm / min, wherein the solid body of the single crystal silicon ingot has a radius and an axial length, a melt-solid interface located above a free melt height level is formed due to surface tension forces generated as the solid body of the single crystal silicon ingot is pulled out of the silicon melt, and further, a meniscus containing the silicon melt exists between the melt-solid interface and the free melt height level, and during at least 50% of the growth of the solid body of the single crystal silicon ingot, a relative height between the free melt height level and a bottom of the reflector is about 40 mm to about 50 mm, and further, during the growth of the solid body of the single crystal silicon ingot, the silicon seed is rotated at a speed of about 5 rpm to about 10 rpm; During growth of a solid body of a single crystal silicon ingot, a cusp magnetic field is applied to the silicon melt; The cusp magnetic field applied to the silicon melt during the growth of the solid body of the single crystal silicon ingot is derived from an upper magnetic coil and a lower magnetic coil, further, the upper magnetic field strength derived from the upper magnetic coil is greater than the lower magnetic field strength derived from the lower magnetic coil; The axial heat flux between the melt-solid interface and the free melt height level during growth of at least 40% of the total axial length of the solid body of the single crystal silicon ingot is at least 20,000 W / m2 over at least about 85% of the radius of the solid body of the single crystal silicon ingot. ２ A method having the absolute value of

2. 10. The method of claim 1, wherein the total axial length of the solid body of the single crystal silicon ingot is at least about 1100 mm.

3. 10. The method of claim 1, wherein the total axial length of the solid body of the single crystal silicon ingot is between about 1200 mm and about 1300 mm.

4. The method of any one of claims 1 to 3, wherein the radius of the solid body of the single crystal silicon ingot is about 75 mm, at least about 75 mm, about 100 mm, or at least about 100 mm.

5. The method of any one of claims 1 to 3, wherein the radius of the solid body of the single crystal silicon ingot is about, or at least about, 150 mm.

6. The axial heat flux between the melt-solid interface and the free melt height level during growth of at least 60% of the axial length of the solid body of the single crystal silicon ingot is at least 20,000 W / m2 over at least about 85% of the radius of the solid body of the single crystal silicon ingot. ２ The method according to any one of claims 1 to 5, having an absolute value of

7. The axial heat flux between the melt-solid interface and the free melt height level during growth of at least 80% of the axial length of the solid body of the single crystal silicon ingot is at least 20,000 W / m2 over at least about 80% of the diameter of the solid body of the single crystal silicon ingot. ２ The method according to any one of claims 1 to 6, having an absolute value of

8. The axial heat flux between the melt-solid interface and the free melt height level during growth of at least 90% of the total length of the solid body of the single crystal silicon ingot is at least 20,000 W / m2 over at least about 85% of the radius of the solid body of the single crystal silicon ingot. ２ The method according to any one of claims 1 to 6, having an absolute value of

9. 9. The method according to any one of claims 1 to 8, wherein the temperature gradient along the meniscus curve between the melt-solid interface and the height level of the free melt has an average value of at least 0.16 K / mm during growth of at least 40% of the total length of the solid body of the single crystal silicon ingot.

10. 9. The method according to any one of claims 1 to 8, wherein the temperature gradient along the meniscus curve between the melt-solid interface and the height level of the free melt has an average value of at least 0.18 K / mm during growth of at least 40% of the total length of the solid body of the single crystal silicon ingot.

11. 9. The method according to any one of claims 1 to 8, wherein the temperature gradient along the meniscus curve between the melt-solid interface and the height level of the free melt has an average value of at least 0.16 K / mm while growing at least 60% of the total length of the solid body of the single crystal silicon ingot.

12. 9. The method according to any one of claims 1 to 8, wherein the temperature gradient along the meniscus curve between the melt-solid interface and the height level of the free melt has an average value of at least 0.18 K / mm while growing at least 60% of the total length of the solid body of the single crystal silicon ingot.

13. 9. The method according to any one of claims 1 to 8, wherein the temperature gradient along the meniscus curve between the melt-solid interface and the height level of the free melt has an average value of at least 0.16 K / mm during growth of at least 80% of the total length of the solid body of the single crystal silicon ingot.

14. 9. The method according to any one of claims 1 to 8, wherein the temperature gradient along the meniscus curve between the melt-solid interface and the height level of the free melt has an average value of at least 0.18 K / mm during growth of at least 80% of the total length of the solid body of the single crystal silicon ingot.

15. The method according to any one of claims 1 to 14, wherein the temperature of the silicon melt in the meniscus is at least 1691 K, measured at the free melt high level, during growth of at least 40% of the total length of the solid body of the single crystal silicon ingot.

16. The method according to any one of claims 1 to 14, wherein the temperature of the silicon melt in the meniscus is at least 1692 K, measured at the free melt height level, during growth of at least 40% of the total length of the solid body of the single crystal silicon ingot.

17. The method according to any one of claims 1 to 14, wherein the temperature of the silicon melt in the meniscus is at least 1691 K, measured at the free melt height level, during growth of at least 60% of the total length of the solid body of the single crystal silicon ingot.

18. The method according to any one of claims 1 to 14, wherein the temperature of the silicon melt in the meniscus is at least 1692 K, measured at the free melt height level, during growth of at least 60% of the total length of the solid body of the single crystal silicon ingot.

19. The method according to any one of claims 1 to 14, wherein the temperature of the silicon melt in the meniscus is at least 1691 K, measured at the free melt height level, during growth of at least 85% of the total length of the solid body of the single crystal silicon ingot.

20. 15. The method according to any one of claims 1 to 14, wherein the temperature of the silicon melt in the meniscus is at least 1692 K, measured at the free melt height level, during growth of at least 85% of the axial length of the solid body of the single crystal silicon ingot.

21. A method described in any one of claims 1 to 20, wherein the upper magnetic field strength resulting from the upper magnetic coil is at least 10% higher than the lower magnetic field strength resulting from the lower magnetic coil.

22. A method described in any one of claims 1 to 21, wherein the upper magnetic field strength resulting from the upper magnetic coil is at least 15% greater than the lower magnetic field strength resulting from the lower magnetic coil.

23. The method according to any one of claims 1 to 22, wherein the bottom wall of the crucible is insulated.

Images