A method of crystal pulling, a single crystal ingot and a crystal pulling furnace

Abstract

The application discloses a single crystal pulling method, a single crystal rod and a single crystal pulling furnace. The method comprises the following steps: in a first stage, the pulling speed of the crystal rod is gradually reduced from a first speed to a second speed, the heating power of the single crystal pulling furnace is gradually increased from a first power to a second power and maintained at the second power; in the first stage, the final radial size gradually decreases along the direction close to the solid-liquid interface; in a second stage, the second power is maintained, and the pulling speed is gradually reduced from the second speed to a third speed; the change rate of the pulling speed in the first stage is smaller than that in the second stage; in the second stage, the final radial size gradually increases along the direction close to the solid-liquid interface, and the maximum final radial size is smaller than the radial size of the constant diameter part of the crystal rod; and in a third stage, the second power and the third speed are maintained; in the third stage, the final radial size gradually decreases along the direction close to the solid-liquid interface; and the final end face forms a convex surface. The application can improve the yield of the crystal rod.

Description

Technical Field

[0001] This application relates to the field of monocrystalline silicon preparation, and in particular to a crystal pulling method, a monocrystalline rod, and a crystal pulling furnace. Background Technology

[0002] Single-crystal silicon is a fundamental core material in high-end technology fields such as semiconductors, optics, and electronics. The Czochralski method, as the mainstream process for preparing single-crystal silicon, involves key stages including crystal seeding, shoulder formation, constant-diameter growth, and finishing. Among these, finishing is the core step to ensure the integrity of the crystal structure and the yield of subsequent processing.

[0003] In existing technologies, the common finishing operation is to rapidly increase the pulling speed and simultaneously increase the heating power, so that the crystal diameter shrinks rapidly to achieve separation from the melt.

[0004] However, the rapid pulling speed and increased heating power directly result in a steep tail shape, meaning the crystal cross-sectional area changes drastically within a very short axial distance. This geometric abrupt change leads to a high concentration of thermal stress at the tail, making it highly susceptible to inducing a large number of dislocations at the moment of detachment from the melt. Once dislocations are generated, the stress concentration causes them to propagate rapidly towards the constant-diameter portion of the crystal rod, even initiating cracks. This severely reduces the internal quality of the crystal and increases the scrap rate in subsequent processing, affecting the yield. Summary of the Invention

[0005] In view of the above-mentioned defects or deficiencies in the prior art, it is desirable to provide a crystal pulling method, monocrystalline silicon and crystal pulling furnace, which can improve the yield while shortening the finishing time by controlling the finishing process.

[0006] In a first aspect, this application provides a crystal pulling method, including a finishing process, which includes the following stages: In the first stage, the pulling speed of the crystal rod is gradually reduced from the first speed to the second speed, and the heating power of the crystal pulling furnace is gradually increased from the first power to the second power and maintained at the second power. In the first stage, the radial dimension of the tail gradually decreases along the direction close to the solid-liquid interface, forming the first diameter reduction section. In the second stage, the second power is maintained, and the pulling speed is gradually reduced from the second speed to the third speed; wherein, the rate of change of the pulling speed in the first stage is less than the rate of change of the pulling speed in the second stage; in the second stage, the radial dimension of the tail gradually increases along the direction close to the solid-liquid interface, and the maximum radial dimension of the tail corresponding to the second stage is less than the radial dimension of the constant diameter portion of the crystal rod, forming an expanded diameter section; and In the third stage, the second power and the third speed are maintained. In the third stage, the radial dimension of the tail gradually decreases along the direction close to the solid-liquid interface, forming the second diameter reduction section. The second diameter reduction section, together with the diameter expansion section and the first diameter reduction section, forms the tail of the crystal rod, and a convex surface is formed on the end face of the equal diameter section away from the crystal rod at the tail.

[0007] One possible implementation is that the first speed is 1.08 to 1.15 times the reference speed, and the second speed is 0.92 to 1.06 times the reference speed; the reference speed is the lifting speed at the end of the constant diameter growth process. The first power is the heating power at the end of the constant diameter growth process, and the second power is obtained by increasing the first power by 12% to 14%. The third speed is 0.08 to 0.1 times the base speed.

[0008] One possible implementation, the first phase includes: The process of the tail section growing from 0 to the first length involves: the first power gradually increasing to the second power, and the first speed decreasing to the fourth speed; wherein the first length is 30%~50% of the total tail length, the fourth speed is 1~1.08 times the base speed, and the fourth speed is greater than the second speed; and The process of the tail section growing from the first length to the second length involves: the fourth speed gradually decreasing to the second speed while maintaining the second power unchanged; and the second length being 60% to 70% of the total tail length.

[0009] One possible implementation method also includes: In the fourth stage, the crucible level is lowered to separate the melt from the crystal rod, and the crystal rod is cooled.

[0010] One possible implementation involves cooling the crystal ingot, specifically as follows: The crystal rod is suspended at a preset distance from the surface of the melt for a preset time; The crystal rod is pulled into the cooling chamber.

[0011] One possible implementation is to preset the distance to 20–85 mm; The preset time is 5~10 minutes.

[0012] Secondly, this application also provides a single-crystal rod, including a constant-diameter portion and a terminal portion connected to the growth end of the constant-diameter portion. The terminal portion includes a first diameter-reducing section, a diameter-expanding section, and a second diameter-reducing section sequentially connected along the axial direction of the rod and away from the constant-diameter portion, wherein: The radial dimension of the first diameter reduction section decreases in the direction away from the constant diameter section, the radial dimension of the diameter expansion section increases in the direction away from the constant diameter section, and the radial dimension of the second diameter reduction section decreases in the direction away from the constant diameter section. The maximum radial dimension of the expanded section is smaller than that of the constant diameter section; and the end face of the tail section is convex.

[0013] One possible implementation is that the first tapered section accounts for 60% to 70% of the total length of the tail section; The expanded diameter section accounts for 25% to 37.5% of the total length of the tail section; The second narrowing section accounts for 2.5% to 5% of the total length of the tail section.

[0014] One possible implementation is that the total length of the tail section is 40mm to 80mm.

[0015] One possible implementation is that at least part of the convex surface is free of growth lines.

[0016] One possible implementation is that at least part of the convex surface makes an angle α with the crystal rod axis, and 20°≤α≤35.26°.

[0017] Thirdly, this application also provides a crystal pulling furnace, including: a processor, a communication interface, a memory, and a communication bus. The processor, communication interface, and memory communicate with each other via a communication bus. Memory, used to store computer programs; The processor, when executing a program stored in memory, implements the steps of any of the above crystal pulling methods.

[0018] Compared with the prior art, this application has the following beneficial effects: In the first stage, this application controls the heating power to first increase and then maintain it, while simultaneously slowing down the pulling speed, resulting in a gradual contraction of the radial dimension at the tail of the crystal ingot. In the second stage, by continuing to decrease the pulling speed while maintaining the heating power, the radial dimension at the tail of the crystal ingot gradually increases, and the maximum radial dimension in the second stage is smaller than that of the constant-diameter section. This reduces dislocation generation while shortening the tail length and diameter, thereby reducing the mass of the tail. In the third stage, by continuing to decrease the pulling speed while maintaining the heating power, the radial dimension at the tail of the crystal ingot is controlled to contract gradually, and the end face grows into a convex surface. This allows dislocations generated during crystal pulling to slide from the center of the convex surface towards the edge until they disappear. This application also improves the tailing efficiency by using a higher rate of change of pulling speed in the second stage compared to the first stage. Attached Figure Description

[0019] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings: Figure 1 A schematic flowchart of the crystal pulling method provided in this application; Figure 2 This is a schematic diagram showing the trends of the crystal rod pulling speed and heating power as a function of the length of the crystal rod tail section at each stage of this application. Figure 3 As one embodiment, this application provides a process diagram of each stage of the final process. Figure 4 This is a schematic diagram of the geometric shape of the tail end of a crystal rod obtained based on an existing crystal pulling method in one embodiment; Figure 5 In another embodiment, a schematic diagram of the geometric shape of the end of a crystal rod obtained based on an existing crystal pulling method is provided. Figure 6 This is a schematic diagram of the geometric shape of the tail end of a crystal rod obtained based on the crystal pulling method of this application in one embodiment; Figure 7 This is a schematic diagram of the architecture of the single crystal furnace of this application in one embodiment.

[0020] In the diagram, 61 is the constant diameter section; 62 is the first diameter reduction section; 63 is the diameter expansion section; 64 is the second diameter reduction section; 7 is the crystal pulling furnace; 71 is the processor; 72 is the communication interface; 73 is the memory; and 74 is the communication bus. Detailed Implementation

[0021] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0022] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. The present application will now be described in detail with reference to the accompanying drawings and embodiments. Furthermore, the term "and / or" in this document is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The terms "first" and "second," etc., in the specification and claims of the embodiments of this application are used to distinguish different objects, not to describe a specific order of objects.

[0023] In the Czochralski crystal growth process, the final stage after the crystal rod is pulled from the melt is a crucial step affecting the quality and integrity of the rod. Traditional final stages often involve rapidly increasing the pulling speed and power, resulting in a steep tail structure at the rod end. This causes thermal stress concentration, making it prone to cracking or dislocation propagation, especially due to thermal shock caused by sudden temperature changes when the rod exits the melt. These methods have limited effectiveness in suppressing dislocation slip during the final stage, and the resulting dislocations can easily propagate in the opposite direction to the constant-diameter section under thermal stress, introducing dislocation defects into this section and reducing the yield.

[0024] Based on this, embodiments of this application provide a crystal pulling method, a single crystal rod, and a crystal pulling furnace. By controlling the finishing process, the yield can be increased while shortening the finishing time.

[0025] like Figure 1 As shown, this application provides a crystal pulling method, including a finishing process, which includes the following stages: S101, In the first stage, the pulling speed of the crystal rod is gradually reduced from the first speed to the second speed, and the heating power of the crystal pulling furnace is gradually increased from the first power to the second power and maintained at the second power; In the first stage, the radial dimension of the tail gradually decreases along the direction close to the solid-liquid interface, forming the first diameter reduction section; Optionally, both the first speed and the second speed are determined based on a reference speed. For example, the first speed is 1.08 to 1.15 times the reference speed, and the second speed is 0.92 to 1.06 times the reference speed. The reference speed can be the lifting speed at the end of the constant diameter growth process. The first power is the heating power at the end of the constant diameter growth process, and the second power is obtained by increasing the first power by 12% to 14%. For ease of description, the first power is denoted as P0. For example, for a 10-inch crystal ingot, P0 is 5-6 kW; for an 11-inch crystal ingot, P0 is 6-8 kW; and for a 12-inch crystal ingot, P0 is 7-10 kW.

[0026] In the first stage, this application employs a "slow-tightening" process to optimize the shape and speed of the crystal rod's tail section: that is, gradually and slowly reducing the pulling speed while simultaneously and slowly increasing the heating power and then keeping it constant, so that the tail section of the crystal rod forms a long and gently sloping conical structure. This conical structure helps to distribute thermal stress evenly along the axial direction of the crystal rod, avoiding localized stress concentration.

[0027] Furthermore, the main objective of this application in the first stage is to shrink the tail of the crystal rod. To achieve this objective, the art typically employs techniques such as increasing the pulling speed and increasing the heating power. However, the inventors have discovered that if the heating power increases too rapidly in the first stage, it can easily lead to excessive thermal radiation at the melt surface, which, under thermal shock, can easily generate a large number of dislocations; on the other hand, it can inhibit the growth of the convex surface in the third stage. Therefore, this application first controls the heating power to increase and then maintain it in the first stage, so that the tail of the crystal rod shrinks smoothly; furthermore, moderately increasing the heating power also helps to prevent crystallization at the melt surface.

[0028] Optionally, the total length of the tail section in this application can be determined based on the diameter of the constant diameter section of the crystal rod and the conditions of the crystal pulling equipment; this invention does not impose any limitations.

[0029] Optionally, the total length of the ingot's tail section is 40mm to 80mm. In the first stage, the length of the ingot's tail section accounts for 60% to 70% of the total length.

[0030] like Figure 2As shown, the curves depicting the changes in pulling speed and power at each stage of the finishing process in this application, respectively, are as follows: Specifically, Figure 2 In the diagram, curve A represents the change in lifting speed with the length of the tail section, and curve B represents the change in heating power with the length of the tail section; from Figure 2 As can be seen, this application primarily determines the lifting speed and heating power for each stage based on the current length of the corresponding tail section. In other words, in each stage of this application, the lifting speed and heating power are controlled in segments based on the current length of the tail section.

[0031] The current length can be characterized by a specific value of the tail length or the proportion of the tail length to the total tail length; the lifting speed and heating power can be set by specific values, or by a proportion relative to a reference value, or by the amount or rate of change relative to the previous adjacent growth length. This application does not limit these settings.

[0032] Optionally, such as Figure 2 As shown, the first stage includes: The process of the tail section growing from 0 to the first length includes: when the first power is gradually increased to the second power, the first speed is reduced to the fourth speed; wherein, the first length is 30% to 50% of the total tail length, the fourth speed is 1 to 1.08 times the base speed, and the fourth speed is greater than the second speed; The process of the tail section growing from the first length to the second length involves the fourth speed gradually decreasing to the second speed while maintaining the second power unchanged; the second length is 60% to 70% of the total tail length.

[0033] S102, such as Figure 2 As shown, in the second stage, the second power is maintained and the pulling speed is gradually reduced from the second speed to the third speed; wherein, the rate of change of the pulling speed in the first stage is less than the rate of change of the pulling speed in the second stage; in the second stage, the radial dimension of the tail gradually increases along the direction close to the solid-liquid interface, and the maximum radial dimension of the tail corresponding to the second stage is less than the radial dimension of the constant diameter part of the crystal rod, forming an expanded diameter section. Optionally, the length of the finishing section corresponding to the second stage accounts for 25% to 37.5% of the total finishing length, and the third speed is also determined according to the reference speed. For example, the third speed is 0.08 to 0.1 times the reference speed.

[0034] In the second stage, this application controls the radial dimension of the tail section to gradually increase along the end toward the solid-liquid interface, and controls the maximum radial dimension of this stage to be smaller than the radial dimension of the constant diameter section, so as to save tailing time and avoid the longer time required for the larger the growth diameter.

[0035] In the second stage, the heating power remains constant, and the lifting speed is still decreased in a stepwise manner based on the current length. Figure 2 It can be seen that the lifting speed decrease curve in the first stage is significantly flatter than that in the second stage; The following explains why the rate of change of pulling speed in the first stage of this application is less than that in the second stage. Specifically, regarding the first stage, the inventors discovered that if the first speed decreases too quickly, it easily leads to an excessively low pulling speed, inhibiting the shrinkage of the diameter at the tail end. In the second stage, when the pulling speed decreases too slowly, on the one hand, the ingot may detach from the melt before the convex surface is formed; on the other hand, it also leads to an excessively long tail end time, reducing the tail end efficiency. Therefore, compared to the second stage, the pulling speed decrease process in the first stage of this application is much slower.

[0036] Optionally, in each stage, the lifting speed of this application changes in a stepwise manner. That is, as the current length of the tail gradually increases, the lifting speed is adjusted once a preset length node is reached. This application reduces the lifting speed in stages, so as to make the solid-liquid interface transition smoothly and avoid interface instability and thermal stress impact caused by sudden changes in lifting speed.

[0037] S103, such as Figure 2 As shown, in the third stage, the second power and the third speed are maintained; in the third stage, the radial dimension of the tail gradually decreases along the direction close to the solid-liquid interface, forming the second diameter reduction section; the second diameter reduction section, together with the diameter expansion section and the first diameter reduction section, forms the tail of the crystal rod, and a convex surface is formed on the end face of the constant diameter section away from the crystal rod at the tail.

[0038] The length of the finishing section in the third stage accounts for 2.5% to 5% of the total finishing length; The following describes the process of change at the end face of the crystal rod in this application: During crystal pulling, the interface where the melt transitions from the liquid phase to the solid phase is called the solid-liquid interface. Generally, the end of the crystal rod that contacts the solid-liquid interface can be categorized by shape as planar, convex, or concave. Specifically, the center point of a convex surface is closer to the melt; conversely, the center of a concave surface is closer to the side of the crystal rod. At the end of the constant diameter growth process, the end of the crystal rod that contacts the solid-liquid interface is usually concave.

[0039] Crystal growth is essentially a process in which the solid-liquid interface continuously advances towards the liquid phase. In the final stages of this application, from the first to the third stage, the solid-liquid interface morphology undergoes a continuous transformation from a concave surface towards the melt to a planar surface, and then to a convex surface towards the melt. This transformation is not instantaneous. In the concave state, dislocations slide from the edges to the center and accumulate and entangle in the central region, forming a high-density dislocation region. Simultaneously, during the crystal rod's liquid removal, the final solidification of the concave central region leads to thermal stress concentration, further increasing the risk of dislocation generation. Therefore, this application controls the final end face to gradually grow from a concave surface to a convex surface, causing dislocations generated during crystal pulling to slide from the center of the convex surface towards the edge until they disappear, thereby effectively improving the quality of the finished product.

[0040] Furthermore, the crystal pulling method of this application can also shorten the finishing length and improve the finishing efficiency. Specifically, the curvature of the convex surface causes the dislocation line to extend along the normal direction during growth, thereby obtaining a radial velocity component towards the edge. Since the normal direction of the convex surface changes with the radial position, the closer to the edge of the convex surface, the larger the angle between the normal and the crystal rod extension axis, and the greater the distance the dislocation is pushed outward within the same growth height, forming a positive feedback process: the closer the dislocation is to the edge of the convex surface, the faster it is pushed away from the center of the convex surface. Therefore, even in micro-convex interfaces with extremely low central convex height and a curvature radius of hundreds of millimeters, significant radial displacement can accumulate within a relatively low growth height, causing the dislocation to gradually migrate from the center to the edge, and eventually slide off the crystal rod surface and disappear.

[0041] The following examples illustrate the changes in heating power and pulling speed at the tail section of the crystal rod with varying tail section lengths for total lengths of 41mm, 60mm, and 80mm, based on Tables 1-3: Table 1. The total length of the crystal rod's tail section is 41 mm.

[0042] As shown in Table 1, when the total length of the crystal rod's tail section is 41 mm, the tail section length, pulling speed, and heating power at each stage are as follows: In the first stage, the length of the crystal rod's tail section is between 1 and 25 mm. The pulling speed gradually decreases from 1.15 times the reference speed to 0.95 times the reference speed, and the heating power increases from the first power to 14% and then remains constant, resulting in the second power. In the second stage, the current length of the crystal rod's tail section is greater than 25 mm and less than or equal to 38 mm. The pulling speed gradually decreases from 0.95 times the reference speed to 0.1 times the reference speed, and the second power remains unchanged. In the third stage, the current length of the crystal rod's tail section is greater than 38 mm and less than or equal to 41 mm. The pulling speed is maintained at 0.1 times the reference speed, and the second power remains unchanged.

[0043] Table 2 shows that the total length of the crystal rod's tail section is 60mm.

[0044] Table 2 shows that when the total length of the ingot's tail section is 60mm, the tail section length, pulling speed, and heating power at each stage are as follows: In the first stage, the current length of the ingot's tail section is 0~40mm, the pulling speed gradually decreases from 1.12 times the reference speed to 0.98 times the reference speed, and the heating power first increases to 12% and then remains constant to obtain the second power; In the second stage, the current length of the ingot's tail section is greater than 40mm and less than or equal to 58mm, the pulling speed gradually decreases from 0.98 times the reference speed to 0.1 times the reference speed, and the second power remains unchanged; In the third stage, the current length of the ingot's tail section is greater than 58mm and less than or equal to 60mm, the pulling speed is maintained at 0.1 times the reference speed, and the second power remains unchanged.

[0045] Table 3 shows the total length of the crystal rod's tail section is 80mm.

[0046] Table 3 shows that when the total length of the crystal rod's tail section is 80mm, the tail section length, pulling speed, and heating power at each stage are as follows: In the first stage, the current length of the crystal rod's tail section is 0~50mm, the pulling speed gradually decreases from 1.08 times the reference speed to 0.92 times the reference speed, and the heating power increases from the first power to 12% and then remains constant, resulting in the second power; In the second stage, the current length of the crystal rod's tail section is greater than 50mm and less than or equal to 78mm, the pulling speed gradually decreases from 0.92 times the reference speed to 0.08 times the reference speed, and the second power remains unchanged; In the third stage, the current length of the crystal rod's tail section is greater than 78mm and less than or equal to 80mm, the pulling speed remains unchanged at 0.08 times the reference speed, and the second power remains unchanged.

[0047] As can be seen from Tables 1 to 3, when the total length of the crystal rod's tail section is shorter, it is necessary to divide the total length into finer-grained stages to ensure that the growth process from concave to convex surfaces is completed within a limited tail section length. Optionally, within the range of 40mm to 80mm, the tail section length of the crystal rod in the second and third stages is about 20mm.

[0048] It is worth noting that the stage divisions and related textual descriptions in Tables 1 to 3 are only illustrative examples of the changing trends of control parameters and do not represent that the start and end points of these stages in the actual process must be strictly based on these data.

[0049] Preferably, at least some of the convex surfaces are free of growth lines.

[0050] The following describes the key control parameters in the formation process of convex surfaces with no growth lines or only partial growth lines: 1. Maintain the second power constant to avoid fluctuations in heating power affecting the quality of the convex surface; Even with proper pulling speed control, dislocation density can still increase significantly at high temperatures. Therefore, to ensure the low-temperature growth conditions required for forming a convex surface without growth lines, this application adopts a strategy of maintaining stable power while reducing the pulling speed. Specifically, in the first stage, the heating power is increased by only 12% to 14% based on the reference power, and no further heating is performed in the second and third stages, thereby avoiding an increase in dislocation density due to high temperatures.

[0051] 2. The argon flow rate should remain consistent with the flow rate at the end of the constant diameter growth stage; no adjustment is required. 3. Maintaining crystal rotation speed: The crystal rotation speed should be kept consistent with the end of the constant diameter growth stage to maintain melt flow stability.

[0052] Specifically, the sharpness of the convex surface is strongly correlated with the dislocation density; the more blurred the convex surface, the greater the dislocation density. Only when the convex surface is sharp can the probability of dislocation generation be minimized. Therefore, this application uses the sharpness of the convex surface to inspect the crystal rod products obtained from each finishing process. Specifically, it judges whether the generated convex surface is smooth, has clear reflection, and has no blurred areas. If the convex surface is not sharp or appears blurred, it indicates that dislocation slip will occur.

[0053] Optionally, after continuously pulling multiple crystal rods, this application evaluates the overall yield of the crystal rods based on the clarity of the convex surface formed by each pulling. When the yield does not meet the requirements, the pulling speed or heating power corresponding to each growth length value at the end is adjusted (such as further reducing the pulling speed or stabilizing the power) until the convex surface is restored to clarity.

[0054] Through multiple experiments, it has been verified that using the crystal pulling method of this application, the yield rate of the crystal rod products can reach 90% when the total length of the tail section is about 40mm; the yield rate of the crystal rod products can reach 95% when the total length of the tail section is about 60mm; and the yield rate of the crystal rod products can reach 100% when the total length of the tail section is about 80mm.

[0055] S104, Fourth stage: Lower the crucible level to separate the melt from the crystal rod and cool the crystal rod.

[0056] Optionally, in the fourth stage, the crucible level is lowered at a rate of 5000 mm / min, the crystal rod is pulled at a rate of 320 mm / h, and the crucible descent distance should be greater than 50 mm.

[0057] The following describes the key control steps for the process of the crystal rod tail detaching from the melt surface: 1. Once the pulling speed has dropped to less than 0.1 times the constant diameter pulling speed and the convex surface is clearly formed, the system enters the disengagement preparation state; 2. By utilizing the weight of the melt itself and the surface tension, the tail of the crystal rod is detached from the liquid surface under low pulling speed conditions, and during the detachment process, a very small detachment point is formed at the center of the tail end face; 3. After the separation point is formed, the crystal rod completely separates from the liquid surface, and the finishing work is completed.

[0058] During the process of the crystal rod's tail detaching from the liquid surface, this application does not mechanically pull it apart, but rather controls the crucible to descend, causing the molten surface to sink relatively, providing a continuous and stable separation driving force for detachment. Furthermore, since the end face of the crystal rod's tail is convex, the area of ​​the bottom in contact with the molten silicon is minimized at the moment the crystal rod's tail detaches from the molten surface, thereby reducing the thermal stress impact during detachment.

[0059] The following explains the principle of dislocation multiplication and slip inhibition at the release point: Observable slip lines are macroscopic traces formed by the collective movement and multiplication of numerous dislocations on the same slip surface. Sufficient space is needed to accommodate the dislocations, and sufficient stress is required to drive their slip. In existing technologies, when a crystal rod is pulled away from the molten surface, the final solidified volume is approximately 4900 mm³. The internal thermal stress is high, and cooling is uneven, providing ample space and driving force for dislocation multiplication, slip, and accumulation, ultimately forming slip lines. For example, the release point can be a tiny region with a diameter of less than 1 mm at the center of the convex surface, with a solidified volume of approximately 5 × 10⁻⁶ mm. -4 The volume is mm³, which is seven orders of magnitude smaller than traditional methods. The tiny volume cannot accommodate a sufficient number of dislocations to form a macroscopic slip line; simultaneously, the internal temperature difference is extremely small during cooling, resulting in very low thermal stress and a lack of mechanical conditions to drive dislocation multiplication and slip. Therefore, in this application, even if a trace amount of dislocation remains, it will rapidly disappear on the surface within a very short slip distance, unable to organize into collective movement, let alone form an observable slip line.

[0060] Optionally, the crystal ingot is cooled, specifically including: The crystal rod is suspended at a preset distance from the surface of the melt for a preset time; The crystal rod is pulled into the cooling chamber.

[0061] The preset distance can be 20-85mm, preferably 65-85mm, which avoids the baking of the molten thermal radiation (when <20mm, the excessive thermal radiation causes the convex surface to be subjected to thermal shock and generate secondary dislocations), and also avoids the thermal stress dislocations caused by excessive cooling when >85mm, so that the dislocation-free structure of the convex surface can be completely maintained.

[0062] Preferably, the preset time is 5~10 minutes; Before the crystal rod is removed from the liquid surface, in the first three stages of this application, the melt has been preheated for about 1 hour by increasing the heating power, so that the heat shield inside the crystal pulling furnace 7 can absorb the heat of the melt and generate stronger thermal radiation; furthermore, this application also uses suspension treatment to continuously keep the lower end of the deliquescent crystal rod warm, thereby slowing down its cooling rate and avoiding microscopic defects caused by rapid cooling.

[0063] After the crystal rod is removed from the molten surface, this application performs a slow cooling procedure. Specifically, after the crystal rod is completely removed from the molten surface, it is not immediately lifted to the cooling chamber, but is first kept in the central region of the thermal field to maintain a relatively stable high temperature state. This allows the entire crystal rod (especially the fragile tail section) to be cooled uniformly in a controlled temperature field, thereby effectively suppressing the proliferation of microcracks and dislocations induced by thermal stress.

[0064] It is worth mentioning that the suspension-cooling technology proposed in this application is not only applicable to the crystal pulling method, but can also be used before furnace shutdown and after edge breaking to reduce dislocation slip caused by thermal shock and reduce the risk of remelting.

[0065] like Figure 3 The diagram shown is a flowchart illustrating the various stages of the finalization process in this application. The finalization process specifically includes: First stage: The radial dimension of the tail gradually decreases along the direction toward the solid-liquid interface; Second stage: The radial dimension of the tail section gradually increases in the direction toward the solid-liquid interface, and the maximum radial dimension of the tail section in this stage is smaller than the radial dimension of the equal diameter section 61. Third stage: The radial dimension of the tail gradually decreases along the direction towards the solid-liquid interface, and the end face of the tail is convex. The fourth stage includes the liquid removal stage and the cooling stage. After the above stages are completed, the finishing process is finished. In the above stages, the present invention optimizes and controls the pulling speed, heating power, and cooling program, etc., to flip the growth interface of the monocrystalline silicon from a concave surface to a convex surface, and controls the convex surface to generate a clear mirror effect. Combined with the synergistic effect of multiple means such as stepped speed reduction, controlling the pulling speed to allow the crystal rod to slowly leave the liquid surface, optimal detachment distance, and suspension, uniform cooling, thermal stress dispersion, and dislocation suppression are achieved in the finishing part of the crystal rod, avoiding thermal stress impact at the moment of detachment and during the cooling process, and achieving dislocation-free or extremely low dislocation density in the finishing part.

[0066] Furthermore, in scenarios where the length of the constant diameter section of the crystal rod is limited due to equipment limitations, this application can ensure the growth length of the crystal rod by shortening the finishing length, and it can also shorten the finishing time and improve the finishing quality of the crystal rod.

[0067] On the other hand, such as Figure 6As shown, this application also provides a single crystal rod, including a constant diameter portion 61 and a tail portion connected to the growth end of the constant diameter portion 61. The tail portion includes a first diameter reduction section 62, a diameter expansion section 63 and a second diameter reduction section 64 connected sequentially along the crystal rod axis and away from the constant diameter portion 61. Among them, the radial dimension of the first diameter-reducing section 62 decreases in the direction away from the constant diameter section 61, the radial dimension of the diameter-expanding section 63 increases in the direction away from the constant diameter section 61, and the radial dimension of the second diameter-reducing section 64 decreases in the direction away from the constant diameter section 61. The maximum radial dimension of the expanded diameter section 63 is smaller than the radial dimension of the equal diameter section 61; and the end face of the tail section is convex.

[0068] The following combination Figures 4-6 A comparative description is provided of the crystal rod tail sections prepared in this application and those prepared using prior art: Figure 4 and Figure 5 These are all schematic diagrams of the geometric shape of the end of a crystal rod in the prior art. Figure 4 The defect of the end of the middle crystal rod is that the end is relatively thick, the dislocation slip path is long, and it is easy to reverse cut into the constant diameter section under the drive of thermal stress. Figure 5 The defect at the end of the crystal rod is that the shape change is relatively steep, which can easily lead to thermal stress concentration due to abrupt changes in cross-sectional area, thereby generating dislocations and driving their slip. Figure 6 This is a schematic diagram of the geometric shape of the end of the crystal rod in this application. Figure 6 The diameter of the segment undergoes a series of phases, from gradually narrowing to gradually expanding and then back to gradually narrowing, with a convex surface formed at the end face of the terminal section. The terminal portion corresponding to the first narrowing segment 62 is a relatively gentle cone shape, which helps to control the uniformity of heat transfer and reduce thermal stress; while the convex end face can guide dislocations to slide from the center to the edge and disappear on the surface, suppressing dislocation back-tangent.

[0069] Preferably, the first tapered section 62 accounts for 60% to 70% of the total length of the tail section; The expanded diameter section (63) accounts for 25% to 37.5% of the total length of the terminal section; The second reduced diameter section 64 accounts for 2.5% to 5% of the total length of the tail section.

[0070] Preferably, the total length of the tail section is 40mm to 80mm.

[0071] Preferably, at least some of the convex surfaces are free of growth lines.

[0072] Preferably, the angle α between at least a portion of the convex surface and the crystal rod axis is 20°≤α≤35.26°.

[0073] The principle of dislocation reduction in this application is explained below: The crystal growth direction is determined by the <100> crystal orientation of the seed crystal and does not change with the process. Dislocations slip along the {111} slip plane, and the angle between the slip plane and the growth direction <100> is 35.26°. The angle between the slip plane and the horizontal plane is 54.74°. These two angles also do not change with the process.

[0074] The changes in process conditions only alter the orientation of the horizontal movement of dislocations. This orientation is determined by the distribution of decomposed shear stress on the {111} plane caused by the thermal stress generated by the radial temperature gradient. Specifically, when the thermal stress generated by the radial temperature gradient is decomposed along each <110> direction within the {111} plane, the maximum shear stress component will be generated in one of the <110> directions. This direction is the direction in which dislocations preferentially slip. When the end face is concave, the horizontal component of the maximum shear stress points towards the central axis of the crystal rod, and the dislocations converge along the slip surface towards the axis. Conversely, when the end face is convex, the radial temperature gradient is reversed, and the horizontal component of the maximum decomposed shear stress points towards the edge of the crystal rod. The dislocations then slide along the {111} plane to the surface of the crystal rod and eventually disappear. Therefore, the crystal pulling method of this application can actively guide dislocations to move towards the edge of the end face and disappear, thereby improving the yield of the crystal rod.

[0075] Furthermore, this application sets the upper limit of the included angle α to 35.26°, which is consistent with the inherent included angle of the slip surface {111} relative to the crystal rod growth direction <100>. When the included angle α is less than 35.26°, the dislocation can be guided to the end side surface and disappear when it moves along the slip surface, thus achieving effective escape. At the same time, the lower limit of α is set to 20° to ensure that the radial displacement space provided by the convex surface is large enough under the limited axial growth height, so that the dislocation can migrate smoothly from the center to the edge and eventually slide out of the crystal rod surface and disappear.

[0076] Optionally, the crystal rod in this application may be made of any one of the following: single-crystal silicon, single-crystal germanium, single-crystal gallium arsenide, single-crystal indium phosphide, or single-crystal sapphire.

[0077] The effects of the present invention will be further described in detail below with reference to specific embodiments, but these embodiments should not be construed as limiting the scope of protection of the present invention.

[0078] Example 1: A silicon single crystal with a diameter of 278mm and a total tail length of 60mm.

[0079] This embodiment focuses on a 278mm diameter sample prepared using the Czochralski method. <100> When the N-type silicon single crystal is grown in constant diameter, the specific process parameters are as follows: constant diameter pulling speed: 100 mm / h (approximately 1.67 mm / min), constant diameter power: 50~60 kW, argon flow rate: 100 L / min, constant diameter crucible rotation speed: 5~6 rpm.

[0080] Step 1 (First Stage): Starting from a constant diameter of 278mm for the crystal rod, the diameter is reduced to 244mm over a length of 40mm. During the diameter reduction process, the power is gradually increased from the constant diameter power, while the pulling speed is slowly decreased.

[0081] Step 2 (Second Stage): The diameter of the crystal rod at the end is grown from 244mm to 254mm.

[0082] Step 3 (Third Stage): The end of the crystal rod shrinks again, and the end face grows into a convex surface.

[0083] The parameters for stages one through three are shown in Table 4: Table 4. Length parameters and corresponding control parameters in Example 1

[0084] In step two, Example 1 controls the low temperature (power stabilized at equal diameter power + 12%, without additional heating) to match the low pulling speed conditions, ensuring that the formed convex surface has a clear and unblurred reflection.

[0085] Step 4 (Fourth Stage): Maintain the heating power at 60~65kW, the argon flow rate at 100L / min, and the crystal rotation speed at 5~6rpm.

[0086] In step four, the pulling speed is reduced to 0.1 times the constant diameter pulling speed (10 mm / h, approximately 0.167 mm / min), causing the tail of the crystal rod to detach from the liquid surface. During the detachment process, due to surface tension, a tiny detachment point is formed in the center of the convex surface, completing the detachment.

[0087] Step 5 (Fourth Stage): After the tail of the crystal rod detaches from the molten metal, precisely control the bottom of the tail of the crystal rod to be 75±2mm from the molten metal surface, and suspend it for 10 minutes. During the suspension period, the argon gas flow rate is maintained at 100L / min.

[0088] After suspension, the crystal rod is pulled out to the cooling chamber, completing the entire crystal pulling process.

[0089] The crystal rod with a diameter of 278 mm prepared in Example 1 has a clear convex surface at the tail.

[0090] Comparative Example 1 The difference from Example 1 is that the heating power was ultimately increased by 18% on the basis of the equal diameter power, resulting in a blurred convex surface.

[0091] Comparative Example 2 The difference from Example 1 is that the suspension distance is controlled at 45mm, and obvious dislocation slippage occurs on the convex surface.

[0092] Comparative Example 3 The difference from Example 1 is that the suspension distance is controlled at 95mm, and dislocation slippage occurs on the convex surface.

[0093] As can be seen from Example 1 and Comparative Example 1, when the heating power is too high, the resulting convex surface becomes blurry and unclear, which means that the dislocation density at the end of the crystal rod increases and obvious dislocation slip occurs. As can be seen from Example 1 and Comparative Example 2, when the suspension distance is too close, obvious dislocation slip occurs. This is because the convex area is baked by the thermal radiation of the melt, generating secondary dislocations and increasing the dislocation density. As can be seen from Example 1 and Comparative Example 3, when the suspension distance is too far, obvious dislocation slip occurs. This is because the crystal rod cooling rate is too fast, leading to thermal stress concentration.

[0094] To determine the quality of crystal rod products, this application employs two evaluation methods: firstly, visually observing the smoothness of the finished end face and the distribution of dislocation lines; secondly, comprehensively evaluating the product by combining data on crystal cracks generated during subsequent machining. Further correlation analysis of the data obtained from these two evaluation methods reveals a correlation between the dislocation distribution characteristics of the finished end face and the degree of cracking in the crystal rod. Specifically, higher dislocation density and longer dislocation slip distance result in a longer crack length during machining, showing a positive correlation. Specifically, this application achieves a finishing success rate of over 97%, and the finishing time is reduced by more than 0.5 hours compared to existing technologies.

[0095] It should be noted that although the operations of the method of the present invention are described in a specific order in the accompanying drawings, this does not require or imply that these operations must be performed in that specific order, or that all of the operations shown must be performed to achieve the desired result. On the contrary, the steps depicted in the flowchart may be performed in a different order. Additionally or alternatively, certain steps may be omitted, multiple steps may be combined into one step, and / or one step may be broken down into multiple steps.

[0096] Figure 7 A schematic diagram of the structure of a crystal pulling furnace 7 according to an embodiment of the present invention is shown. Figure 7 As shown, the crystal pulling furnace 7 includes a processor 71, a communication interface 72, a memory 73, and a communication bus 74. The processor 71, communication interface 72, and memory 73 communicate with each other through the communication bus 74. Memory 73 is used to store computer programs; When the processor 71 executes the program stored in the memory 73, it implements the steps of any of the aforementioned crystal pulling methods.

[0097] For example, when processor 71 executes the program stored in memory 73, it performs the following steps: In the first stage, the pulling speed of the crystal ingot is gradually reduced from a first speed to a second speed, and the heating power of the crystal pulling furnace is gradually increased from a first power to a second power and maintained at the second power; in the first stage, the radial dimension of the tail gradually decreases along the direction close to the solid-liquid interface, forming a first diameter-reducing section; in the second stage, the second power is maintained, and the pulling speed is gradually reduced from the second speed to a third speed; wherein the rate of change of the pulling speed in the first stage is less than the rate of change of the pulling speed in the second stage; in the second stage, the radial dimension of the tail gradually increases along the direction close to the solid-liquid interface, and the maximum radial dimension of the tail corresponding to the second stage is less than the radial dimension of the constant-diameter portion of the crystal ingot, forming a diameter-expanding section; and In the third stage, the second power and the third speed are maintained. In the third stage, the radial dimension of the tail gradually decreases along the direction close to the solid-liquid interface, forming the second diameter reduction section. The second diameter reduction section, together with the diameter expansion section and the first diameter reduction section, forms the tail of the crystal rod, and a convex surface is formed on the end face of the equal diameter section away from the crystal rod at the tail.

[0098] The communication bus 74 mentioned above for the single crystal furnace can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus 74 can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, it is represented by only one thick line in the figure, but this does not indicate that there is only one bus or one type of bus.

[0099] Communication interface 72 is used for communication between the aforementioned single crystal furnace and other equipment.

[0100] The memory 73 may include random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Optionally, the memory 73 may also be at least one storage device located remotely from the aforementioned processor.

[0101] The processor 71 mentioned above can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0102] It should be noted that, for the sake of simplicity, the method embodiments are all described as a series of actions. However, those skilled in the art should understand that the embodiments of this application are not limited to the described order of actions, because according to the embodiments of this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions involved are not necessarily essential to the embodiments of this application.

Claims

1. A crystal pulling method, comprising a finishing process, characterized in that, The finishing process includes the following stages: In the first stage, the pulling speed of the crystal rod is gradually reduced from the first speed to the second speed, and the heating power of the crystal pulling furnace is gradually increased from the first power to the second power and maintained at the second power. In the first stage, the radial dimension of the tail gradually decreases along the direction close to the solid-liquid interface, forming the first diameter reduction section. In the second stage, the second power is maintained, and the pulling speed is gradually reduced from the second speed to the third speed; wherein, the rate of change of the pulling speed in the first stage is less than the rate of change of the pulling speed in the second stage; in the second stage, the radial dimension of the tail gradually increases along the direction close to the solid-liquid interface, and the maximum radial dimension of the tail corresponding to the second stage is less than the radial dimension of the constant diameter portion of the crystal rod, forming an expanded diameter section; and In the third stage, the second power and the third speed are maintained; in the third stage, the radial dimension of the tail gradually decreases along the direction close to the solid-liquid interface to form a second diameter reduction section; the second diameter reduction section, together with the diameter expansion section and the first diameter reduction section, forms the tail portion of the crystal rod, and a convex surface is formed on the end face of the tail portion away from the constant diameter portion of the crystal rod.

2. The method according to claim 1, characterized in that, The first speed is 1.08 to 1.15 times the reference speed, and the second speed is 0.92 to 1.06 times the reference speed; the reference speed is the lifting speed at the end of the equal diameter growth process. The first power is the heating power at the end of the constant diameter growth process, and the second power is obtained by increasing the first power by 12% to 14%. The third speed is 0.08 to 0.1 times the reference speed.

3. The method according to claim 1, characterized in that, The first stage includes: The process of the tail section growing from 0 to a first length includes: the first power gradually increasing to a second power, and the first speed decreasing to a fourth speed; wherein the first length is 30% to 50% of the total tail length, the fourth speed is 1 to 1.08 times the reference speed, and the fourth speed is greater than the second speed; and The process of the tail section growing from the first length to the second length includes: the fourth speed gradually decreasing to the second speed while maintaining the second power unchanged; wherein the second length is 60% to 70% of the total tail length.

4. The method according to claim 1, characterized in that, The method further includes: In the fourth stage, the crucible level is lowered to separate the melt from the crystal rod, and the crystal rod is cooled.

5. The method according to claim 4, characterized in that, The cooling process for the crystal rod specifically includes: The crystal rod is suspended at a predetermined distance from the surface of the melt for a predetermined time; and The crystal rod is pulled into the cooling chamber.

6. The method according to claim 5, characterized in that, The preset distance is 20–85 mm; The preset time is 5 to 10 minutes.

7. A single-crystal rod, characterized in that, The growth end of the equal-diameter section includes a constant-diameter section and a terminal section connected to the growth end of the constant-diameter section. The terminal section includes a first diameter-reducing section, a diameter-expanding section, and a second diameter-reducing section connected sequentially along the crystal rod axis and away from the constant-diameter section, wherein: The radial dimension of the first diameter-reducing section decreases in the direction away from the constant diameter portion, the radial dimension of the diameter-expanding section increases in the direction away from the constant diameter portion, and the radial dimension of the second diameter-reducing section decreases in the direction away from the constant diameter portion. The maximum radial dimension of the expanded diameter section is smaller than the radial dimension of the constant diameter section; and the end face of the tapering section is convex.

8. The single crystal rod according to claim 7, characterized in that, The first tapered section accounts for 60% to 70% of the total length of the tapered section; The enlarged diameter section accounts for 25% to 37.5% of the total length of the tapering section; The second tapered section accounts for 2.5% to 5% of the total length of the tail section.

9. The single crystal rod according to claim 8, characterized in that, The total length of the tail section is 40mm~80mm.

10. The single-crystal rod according to claim 8, characterized in that, At least some of the convex surfaces are free of growth lines.

11. The single-crystal rod according to claim 7, characterized in that, At least a portion of the convex surface forms an angle α with the crystal rod axis, and 20°≤α≤35.26°.

12. A crystal pulling furnace, characterized in that, include: Processor, communication interface, memory, and communication bus. The processor, communication interface, and memory communicate with each other via a communication bus. Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the steps of the crystal pulling method according to any one of claims 1 to 6.

Images