Method for systematic control of heavy doped head resistivity

Abstract

The application provides a method for systemically regulating heavy-doped head resistivity, and belongs to the technical field of monocrystalline silicon preparation. The method comprises the following steps: S1, systemically setting pressure, argon, first-doping multiple, total-doping multiple, supplementary-doping coefficient and test temperature time based on the classification of heavy-doped head resistivity; wherein, the systemically setting comprises uniformly setting the pressure corresponding to each classification, and adjusting the parameter values of argon, first-doping multiple, total-doping multiple, supplementary-doping coefficient and test temperature time based on the parameter values of pressure; when setting the first-doping multiple, total-doping multiple and supplementary-doping coefficient, the first-doping multiple, total-doping multiple and supplementary-doping coefficient of one classification are selected and set first, and then the first-doping multiple, total-doping multiple and supplementary-doping coefficient of the processes of the remaining classifications are set in series based on the first-doping multiple, total-doping multiple and supplementary-doping coefficient of the process; S2, pulling a crystal based on the parameter values set in step S1, and obtaining heavy-doped monocrystalline silicon corresponding to the head resistivity.

Description

Technical Field

[0001] This invention relates to the field of single-crystal silicon preparation technology, and specifically to a method for systematically controlling the resistivity of heavily doped head. Background Technology

[0002] The resistivity distribution of heavily doped single crystals generally exhibits a trend of high at the beginning and low at the end. There are two methods for controlling resistivity: one is controlling the resistivity at the beginning, and the other is controlling the decreasing resistivity trend. Stable and effective control of these two methods is a prerequisite for ensuring the pass rate of resistivity in heavily doped crystals. Taking heavily doped arsenic as an example, the required resistivity is generally between 1-5 mΩ·cm. The crystal growth process of heavily doped crystals is usually differentiated by controlling the magnitude of the head resistivity, classifying the head resistivity into three categories: <2 mΩ·cm; <3 mΩ·cm; and <4 mΩ·cm. Controlling the head resistivity within ±0.15 mΩ·cm is considered effective control. For processes with <4 mΩ·cm, the head resistivity needs to be controlled between 3.65-3.95 mΩ·cm; for processes with <3 mΩ·cm, the head resistivity needs to be controlled between 2.65-2.95 mΩ·cm; and for processes with <2 mΩ·cm, the head resistivity needs to be controlled between 2.35-2.65 mΩ·cm. The stability of resistivity control is not only reflected in process control, but also requires efficient control in process switching management. In traditional control methods, the core control elements of different resistivity levels are set independently, lacking systematic correlation. This leads to the head resistivity easily exceeding the control window when switching between processes at different resistivity levels, resulting in poor stability. Summary of the Invention

[0003] In view of this, the present invention provides a method for systematically regulating the resistivity of heavily doped head to solve the technical problem of poor head resistivity control stability when switching between processes with different resistivity levels in traditional control methods.

[0004] The technical solution adopted by this invention to solve its technical problem is:

[0005] A method for systematically controlling the resistivity of heavily doped head includes the following steps:

[0006] S1. Based on the classification of heavily doped head resistivity, systematically set the parameter values ​​for pressure, argon, initial doping multiple, total doping multiple, supplementary doping coefficient, and test temperature time. The systematic setting includes uniformly setting the pressure for each classification, and adjusting the parameter values ​​for argon, initial doping multiple, total doping multiple, supplementary doping coefficient, and test temperature time accordingly based on the pressure parameter values ​​to compensate for the volatilization loss caused by the uniform pressure setting. When setting the initial doping multiple, total doping multiple, and supplementary doping coefficient, first select one classification of the process to set the initial doping multiple, total doping multiple, and supplementary doping coefficient, and then use the initial doping multiple, total doping multiple, and supplementary doping coefficient of that process as a benchmark to set the initial doping multiple, total doping multiple, and supplementary doping coefficient of the other classifications in series.

[0007] S2. Based on the parameter values ​​systematically set in step S1, crystal pulling is performed to obtain heavily doped monocrystalline silicon with the corresponding head resistivity.

[0008] Preferably, based on the pressure control level of the release stage controlled by the large-diameter butterfly valve, the control range is determined to be (Min1, Max1), and the pressure values ​​of various processes are uniformly set as Min1 + (20%-40%) × (Max1-Min1); where Min1 refers to the minimum pressure value of the release stage controlled by the large-diameter butterfly valve, and Max1 refers to the maximum pressure value of the release stage controlled by the large-diameter butterfly valve.

[0009] Preferably, the classification of heavily doped head resistivity includes heavily doped head resistivity <2mΩ.cm, heavily doped head resistivity <3mΩ.cm, and heavily doped head resistivity <4mΩ.cm; based on the argon control level of the equipment during the ignition stage, the control range is determined as (Min2, Max2), and the argon for the <4mΩ.cm and <3mΩ.cm processes is uniformly set to Min2 + (20%-40%) × (Max2-Min2), and the argon for the <2mΩ.cm process is set to Max2; where Min2 refers to the minimum argon value of the equipment during the ignition stage, and Max2 refers to the maximum argon value of the equipment during the ignition stage.

[0010] Preferably, the initial doping ratio for the <3mΩ·cm process is set to 6.5-8.0, the total doping ratio is set to 8.0-9.5, and the supplementary doping coefficient is set to 7.0-9.0.

[0011] Preferably, the initial doping multiple for the <4mΩ.cm process is set to be reduced by (20-35)% from the initial doping multiple for the <3mΩ.cm process, and the initial doping multiple for the <2mΩ.cm process is set to be increased by (25-40)% from the initial doping multiple for the <3mΩ.cm process.

[0012] Preferably, the total doping factor for the <4mΩ.cm process is set to be reduced by (15-25)% from the total doping factor for the <3mΩ.cm process, and the total doping factor for the <2mΩ.cm process is set to be increased by (15-20)% from the total doping factor for the <3mΩ.cm process.

[0013] Preferably, the doping coefficient for the <4mΩ.cm process is set to be reduced by (30-40)% from the doping coefficient for the <3mΩ.cm process, and the doping coefficient for the <2mΩ.cm process is set to be increased by (25-30)% from the doping coefficient for the <3mΩ.cm process.

[0014] Preferably, the temperature test time for all types of processes is uniformly set to 4±0.5h.

[0015] Preferably, when the quality of the crystallizer is not up to standard, the crystallizer is melted back and re-induced, and no additional doping is added this time. The next addition is stopped when the total amount of additional doping is greater than or equal to the total doping multiple × the amount of material fed.

[0016] Preferably, the liquid outlet distance for all processes is uniformly set to 25-30mm.

[0017] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0018] This invention proposes a systematic control method to address the control of resistivity at different levels under the same conditions. By setting and compensating parameters such as pressure, argon gas, initial doping multiple, total doping multiple, supplementary doping coefficient, and test temperature time within a unified framework, systematic control ensures that the head resistivity can be stably controlled under any process switching or production fluctuations. Attached Figure Description

[0019] Figure 1 A head resistivity comparison diagram for the comparative example and the target setting 1 of the embodiment.

[0020] Figure 2 A head resistivity comparison diagram for the target setting 2 of the comparative example and embodiment.

[0021] Figure 3 A head resistivity comparison diagram for the target setting 3 of the comparative example and embodiment.

[0022] Figure 4 A head resistivity comparison diagram for the target setting 2 of the example and the embodiment under other preconditions. Detailed Implementation

[0023] The technical solutions and effects of the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0024] This invention provides a method for systematically controlling the resistivity of heavily doped head, comprising the following steps:

[0025] S1. Based on the classification of heavily doped head resistivity, systematically set the parameter values ​​for pressure, argon, initial doping multiple, total doping multiple, supplementary doping coefficient, and test temperature time. The systematic setting includes uniformly setting the pressure for each classification, and adjusting the parameter values ​​for argon, initial doping multiple, total doping multiple, supplementary doping coefficient, and test temperature time accordingly based on the pressure parameter values ​​to compensate for the volatilization loss caused by the uniform pressure setting. When setting the initial doping multiple, total doping multiple, and supplementary doping coefficient, first select one classification of the process to set the initial doping multiple, total doping multiple, and supplementary doping coefficient, and then use the initial doping multiple, total doping multiple, and supplementary doping coefficient of that process as a benchmark to set the initial doping multiple, total doping multiple, and supplementary doping coefficient of the other classifications in series.

[0026] S2. Based on the parameter values ​​systematically set in step S1, crystal pulling is performed to obtain heavily doped monocrystalline silicon with the corresponding head resistivity.

[0027] When switching from one process to another, the independent changes in each parameter, such as pressure, argon gas, initial doping multiple, total doping multiple, supplementary doping coefficient, and testing time, will have a cumulative and coupled effect on the final result of "effective doping concentration in the melt," making the final head resistivity difficult to predict and control. This invention proposes a systematic control method to address the control of resistivity at different levels under the same conditions. By setting and compensating for parameters such as pressure, argon gas, initial doping multiple, total doping multiple, supplementary doping coefficient, and testing time within a unified framework, systematic control ensures stable control of the head resistivity under any process switch or production fluctuations.

[0028] In some embodiments, the control method of the present invention is applicable to heat fields with the same height-to-diameter ratio, single-type heat shields, integral heaters, and down-draft heat fields.

[0029] For traditional control methods, lower resistivity and higher pressure result in a smaller butterfly valve opening angle, which is challenging for production. Conversely, higher resistivity and lower pressure allow for a larger butterfly valve opening angle, making production relatively easier. Therefore, switching between these conditions can lead to production imbalances. Furthermore, in heavily arsenic-doped processes, the biggest uncontrollable factor is the volatility of arsenic. With independent settings, changes in P (pressure) and A (argon gas) alter the furnace draft and gas flow rate, causing drastic fluctuations in the arsenic volatilization rate and resulting in inconsistent total arsenic content in the melt. Therefore, this invention uses a uniform pressure setting to address production imbalances and subsequent parameter adjustments to compensate for losses caused by consistent pressure. In some embodiments, the present invention, based on the pressure control level of the release stage controlled by a large-diameter butterfly valve, obtains a stably controllable range of (Min1, Max1), and uniformly sets the pressure values ​​for various processes as Min1 + (20%-40%) × (Max1-Min1); where Min1 refers to the minimum pressure value of the release stage controlled by the large-diameter butterfly valve, and Max1 refers to the maximum pressure value of the release stage controlled by the large-diameter butterfly valve. Within the range of (Min1, Max1), the opening angle of the butterfly valve is controlled within the range of 30-40°, which can provide stable suction to carry away volatiles without forming strong eddies due to excessive suction, causing oxides to accumulate in the eddy region and fall back into the melt, interfering with growth. At the same time, it eliminates the sudden change in evaporation rate caused by suction fluctuations, providing a stable "evaporation background field" for subsequent doping calculations.

[0030] The classification of heavily doped head resistivity includes heavily doped head resistivity <2 mΩ·cm, heavily doped head resistivity <3 mΩ·cm, and heavily doped head resistivity <4 mΩ·cm. Both pressure P and argon gas A jointly determine the evaporation rate of the melt surface and the transport of impurities within the furnace. However, pressure has a greater impact on resistivity than argon gas. Therefore, this invention classifies argon gas into two types: a low-argon gas design is used for head resistivity <4 mΩ·cm and <3 mΩ·cm, while a high-argon gas design is used for head resistivity <2 mΩ·cm. Specifically, based on the argon gas control level during the ignition phase, the control range is determined as (Min2, Max2). The argon gas for the <4 mΩ·cm and <3 mΩ·cm processes is uniformly set to Min2 + (20% - 40%) × (Max2 - Min2), and the argon gas for the <2 mΩ·cm process is set to Max2. Here, Min2 refers to the minimum argon gas value during the ignition phase, and Max2 refers to the maximum argon gas value during the ignition phase. For extremely high doping processes (<2 mΩ·cm), the arsenic concentration in the melt is extremely high. During crystallization, a large amount of arsenic is displaced to the solid-liquid interface, forming an extremely thick, impurity-enriched boundary layer on the melt side of the interface. This "barrier" hinders the smooth entry of dopants into the crystal, leading to high head resistance. By setting a high argon gas flow, the strong argon gas flow washes over the liquid surface, forcibly thinning the enriched boundary layer thickness at the solid-liquid interface. This balances the boundary layer thickness at different impurity concentrations, allowing arsenic atoms to smoothly enter the crystal. This compensates for the increase in resistivity caused by the high-concentration boundary layer, ensuring that the head resistance of the high-doping process can still be kept within the target window.

[0031] Furthermore, after impurities are incorporated, some of them enter the melt, while others evaporate and are lost over time. When the loss reaches a certain level, re-doping is performed to compensate for the lost impurities. This invention uses the initial doping multiple S1, total doping multiple S2, and supplementary doping coefficient S3 for processes <3 mΩ·cm as benchmarks, and correspondingly sets the initial doping multiple S1, total doping multiple S2, and supplementary doping coefficient S3 for processes <2 mΩ·cm and <4 mΩ·cm.

[0032] Furthermore, the initial doping factor S1 for the <3 mΩ·cm process is set to 6.5-8.0, and set in series with the S1 of the <3 mΩ·cm process as the reference. Since the <4 mΩ·cm process has high resistivity, low target doping concentration, and low volatilization loss, the doping amount is reduced, and the initial doping factor S1 for the <4 mΩ·cm process is set to be (20-35)% less than the initial doping factor S1 of the <3 mΩ·cm process. However, the <2 mΩ·cm process has low resistivity, extremely high target doping concentration, and the greatest volatilization loss, so the doping amount needs to be increased. The initial doping factor S1 for the <2 mΩ·cm process is set to be (25-40)% more than the initial doping factor S1 of the <3 mΩ·cm process to compensate for the volatilization loss caused by consistent pressure.

[0033] Furthermore, the total doping factor S2 determines the total doping amount throughout the entire crystal rod length. This invention sets the total doping factor S2 for the <3 mΩ·cm process to 8.0-9.5, using the <3 mΩ·cm process as a baseline for series settings. The total doping factor for the <4 mΩ·cm process is set to be reduced by (15-25)% from the baseline of the <3 mΩ·cm process, and the total doping factor for the <2 mΩ·cm process is set to be increased by (15-20)% from the baseline of the <3 mΩ·cm process. Thus, fine-tuning is performed based on the correlation coefficient between the total doping factor S2 and the head resistance to address the "segregation effect"—that is, during head crystallization, impurities are squeezed into the melt (segregation coefficient <1). By adjusting the total doping amount, it is ensured that the melt concentration transitions smoothly when the head is pulled into the stable region, preventing subsequent growth fluctuations caused by a break in the tail section despite the head resistance meeting the standard.

[0034] Furthermore, the compensation logic for the supplementary doping coefficient S3 is the same as that for the initial doping multiple S1. The supplementary doping coefficient for the <3mΩ.cm process is set to 7.0-9.0, and set in series with S3 of the <3mΩ.cm process as the reference. Since the target impurity concentration is low and the volatilization loss is small for high resistance, the amount of impurity doping is reduced. The supplementary doping coefficient S3 for the <4mΩ.cm process is set to be reduced by (30-40)% from the reference of the supplementary doping coefficient S3 of the <3mΩ.cm process to match its lower evaporation loss. On the other hand, the target impurity concentration is high and the volatilization loss is the greatest for low resistance. Therefore, the amount of impurity doping is increased. The supplementary doping coefficient S3 for the <2mΩ.cm process is set to be increased by (25-30)% from the reference of the supplementary doping coefficient S3 of the <3mΩ.cm process to offset the additional evaporation loss caused by its high concentration. By precisely increasing or decreasing the percentage, the loss of volatilization caused by different volatilization rates under different impurity concentrations is compensated, ensuring that no matter which setting is switched to, the actual amount of absolutely effective arsenic retained by the melt when entering constant diameter growth is exactly equal to the amount required for the target value of that setting.

[0035] During the temperature testing phase, the melt is in a high-temperature, dry-burning state, which is the period when impurities volatilize most severely. When independently set, the temperature testing time varies, resulting in random volatilization. This invention uniformly sets the temperature testing time T for all three processes to 4 ± 0.5 h, allowing the melt sufficient time to achieve uniform impurity distribution (homogenization). Furthermore, it transforms the variable "temperature testing loss" into a fixed constant, allowing the initial doping amount to be directly deducted from this fixed loss, thus facilitating subsequent adjustments to other parameters.

[0036] When setting up the process independently, the decision to add additives or when to add additives when the lead crystal fails to meet the requirements often involves human judgment. This can lead to slight but not negligible differences in the total melt volume and total arsenic content when entering the equal diameter stage. However, this invention uses a consistent setting for all three processes: when the lead crystal quality fails to meet the requirements, it is melted back and re-leaded, without adding additives this time. Furthermore, additives are stopped the next time only after the total additive amount is ≥ the total additive ratio × the feed amount. Simultaneously, the failure to meet the lead crystal quality requirements is quantitatively controlled, primarily by adjusting the drawing speed, diameter, and secondarily by adjusting the temperature, thereby reducing the additional time loss caused by no-go (NG) due to substandard lead crystal quality. Specifically, the drawing speed control is as follows: a target drawing speed is set, and the drawing speed for the last 200mm of the lead crystal length needs to be controlled within ±1mm / min of the target drawing speed, and it cannot remain at the upper or lower limit for extended periods, nor can there be any abrupt changes. The diameter control is as follows: a target diameter is set, and the diameter for the last 200mm of the lead crystal length needs to be controlled within ±0.2mm of the target diameter. The temperature control is as follows: the temperature at 50mm of the lead crystal length minus the final temperature < -4℃. By quantifying the judgment of crystal-driving quality and specifying the rules for doping replenishment, the error of human judgment is completely eliminated. The quantification standard ensures that the melt state (volume, temperature gradient) is highly consistent each time it enters the constant diameter stage, while the "no replenishment during remelting" rule avoids the fatal problem of the total melt mass increasing due to multiple crystal-driving operations, thereby diluting the doping concentration.

[0037] Furthermore, the gap value between the liquid outlet and the liquid surface is uniformly set to 25-30mm for the three types of processes to reduce eddies at the solid-liquid interface, fix the airflow eddy shape and liquid surface pressure at the solid-liquid interface, and ensure that the physical path of impurities escaping from the liquid surface is consistent under different processes.

[0038] Furthermore, when setting the aforementioned parameters independently, the lack of coordination between them means that even minor external disturbances (such as changes in raw material batches or fluctuations in ambient temperature) can be amplified, leading to drastic fluctuations in head resistance between different furnace runs under the same process and poor product consistency. Moreover, independently set processes often rely on operator experience for "fine-tuning," heavily depending on individual skills and hindering standardization and technology transfer. When developing new resistivity levels, the independently set method cannot provide a predictable adjustment path, resulting in long development cycles and high costs for new processes. This invention integrates the influence and effects of different factors on head resistivity, aiming for high standardization to compensate for time or evaporation losses caused by adjustments in the original method, thus significantly improving the stability of head resistivity control. Simultaneously, when switching between the three types of processes, operators no longer need to rely on experience to "guess" how much P and A should be adjusted; they can simply mechanically execute the adjustment method of this invention to immediately lock in the target head resistance (within ±1.5 mΩ·cm) for the new level. This allows skills that originally relied heavily on the operator's "feel" to be truly transformed into replicable, standardized industrial technologies.

[0039] Experimental Setup: The experiment was conducted in two groups (comparative and exemplary). Target setting 1: <4 mΩ·cm, Target setting 2: <3 mΩ·cm, Target setting 3: <2 mΩ·cm. Comparative Example: For Target Setting 1, Target Setting 2, and Target Setting 3, the pressure, argon gas, initial doping ratio, total doping ratio, supplementary doping coefficient, and test temperature time for each process were independently set. Exemplary Example: Using the control method of this patent, the pressure, argon gas, initial doping ratio, total doping ratio, supplementary doping coefficient, and test temperature time for each process were systematically set for Target Setting 1, Target Setting 2, and Target Setting 3. Data were selected and compared within a time difference of ±5 hours for equal diameter infeed. N=10, where N is the sample size, meaning 10 sets of experiments were conducted for both the comparative and exemplary examples.

[0040] Except for the parameters that differed, the same test parameters were: SR: 10-14 rpm; CR: 1-3 rpm.

[0041] Target setting parameter table (Table 1): Comparative Example 13 110 5.0 7.0 9.0 2.0 Example 16 100 5.2 7.0 5.5 4.5

[0042] Based on the above experimental parameters and the parameters in Table 1, crystal pulling was performed on the comparative examples and embodiments to obtain the corresponding heavily doped monocrystalline silicon. The head resistivity of the heavily doped monocrystalline silicon pulled each time was recorded, and the results were plotted as shown in the figure. Figure 1 The diagram shows a comparison of the resistivity of the head. Among them, Figure 1 The curve corresponding to the original method is the head resistivity curve of heavily doped monocrystalline silicon pulled using the control method of the embodiment, while the curve corresponding to the original method is the head resistivity curve of heavily doped monocrystalline silicon pulled using the comparative control method. Figure 1 It can be seen that the head resistivity of the comparative example is controlled between 3.55-4 mΩ·cm (U1-L1), while the head resistivity of the embodiment is controlled between 3.65-3.95 mΩ·cm (U2-L2). Therefore, the head resistivity controlled by the embodiment meets the required control range, indicating that the head resistivity of the embodiment is effectively controlled. Furthermore, from... Figure 1 The standard deviations in the data show that the standard deviation of the comparative example is greater than that of the example. The smaller the standard deviation, the more concentrated the data is, the smaller the fluctuation, and the fluctuation range is reduced by 45%.

[0043] Target setting parameter table (Table 2): Comparative Example 20 110 7.0 8.5 9.0 2.0 Example 16 100 7.3 8.5 8.3 4.5

[0044] Based on the above experimental parameters and the parameters in Table 2, crystal pulling was performed on the comparative examples and embodiments to obtain the corresponding heavily doped monocrystalline silicon. The head resistivity of the heavily doped monocrystalline silicon pulled each time was recorded, and the results were plotted as shown in the figure. Figure 2 The diagram shows a comparison of the resistivity of the head. Among them, Figure 2 The curve corresponding to the original method is the head resistivity curve of heavily doped monocrystalline silicon pulled using the control method of the embodiment, while the curve corresponding to the original method is the head resistivity curve of heavily doped monocrystalline silicon pulled using the comparative control method. Figure 2 It can be seen that the head resistivity of the comparative example is controlled between 2.6-3.05 mΩ·cm (U1-L1), while the head resistivity of the embodiment is controlled between 2.65-2.95 mΩ·cm (U2-L2). Therefore, the head resistivity controlled by the embodiment meets the required control range, indicating that the head resistivity of the embodiment is effectively controlled. Furthermore, from... Figure 2 The standard deviations in the data show that the standard deviation of the comparative example is greater than that of the example. The smaller the standard deviation, the more concentrated the data is, the smaller the fluctuation, and the fluctuation range is reduced by 52%.

[0045] Target setting 3-parameter table (Table 3): Comparative Example 30 120 8.5 10.0 9.0 4.0 Example 16 150 9.6 10.4 10.6 4.5

[0046] Based on the above experimental parameters and the parameters in Table 3, crystal pulling was performed on the comparative examples and embodiments to obtain the corresponding heavily doped monocrystalline silicon. The head resistivity of the heavily doped monocrystalline silicon pulled each time was recorded, and the results were plotted as shown in the figure. Figure 3 The diagram shows a comparison of the resistivity of the head. Among them, Figure 3 The curve corresponding to the original method is the head resistivity curve of heavily doped monocrystalline silicon pulled using the control method of the embodiment, while the curve corresponding to the original method is the head resistivity curve of heavily doped monocrystalline silicon pulled using the comparative control method. Figure 3It can be seen that the head resistivity of the comparative example is controlled between 2.31-2.8 mΩ·cm (U1-L1), while the head resistivity of the embodiment is controlled between 2.35-2.65 mΩ·cm (U2-L2). Therefore, the head resistivity controlled by the embodiment meets the required control range, indicating that the head resistivity of the embodiment is effectively controlled. Furthermore, from... Figure 3 The standard deviations in the data show that the standard deviation of the comparative example is greater than that of the embodiment. A smaller standard deviation indicates more concentrated data and less fluctuation, with the fluctuation range reduced by 55%. Based on the above experimental results, it can be concluded that the systematic control method of this invention, compared with independently controlling the parameters of each process, can effectively control the resistivity of the three types of head resistivity and reduce the fluctuation range of head resistivity.

[0047] In some implementations, this patent can be extended to conditions other than those described herein, and most of the standardized measures can be adopted. Specifically, under the premise of low aspect ratio, irregularly shaped heat shield, and integral heater, the effect of using this method for control is as follows:

[0048] Experimental setup: The experiment was conducted in two groups (comparative example and implementation example). The comparative example was controlled using the conventional method, while the implementation example used the control method of this patent. Target setting 2: <3 mΩ·cm. Data were selected and compared within ±5 hours of the time difference for equal diameter infeed. N=10. Except for parameters showing differences, the following experimental parameters were identical: SR: 12 rpm; CR: 2 rpm.

[0049] Target setting parameter table (Table 4): Comparative Example 12 110 7.0 8.5 9.0 4.0 Example 15 90 7.3 8.8 8.3 4.5

[0050] Based on the above experimental parameters and the parameters in Table 4, crystal pulling was performed on the comparative examples and embodiments to obtain the corresponding heavily doped monocrystalline silicon. The head resistivity of the heavily doped monocrystalline silicon pulled each time was recorded, and the results were plotted as shown in the figure. Figure 4 The diagram shows a comparison of the resistivity of the head. Among them, Figure 4 The curve corresponding to the original method is the head resistivity curve of heavily doped monocrystalline silicon pulled using the control method of the embodiment, while the curve corresponding to the original method is the head resistivity curve of heavily doped monocrystalline silicon pulled using the comparative control method. Figure 4 It can be seen that the head resistivity of the comparative example is controlled between 2.64-3.06 mΩ·cm (U1-L1), while the head resistivity of the embodiment is controlled between 2.65-2.95 mΩ·cm (U2-L2). Therefore, the head resistivity controlled by the embodiment meets the required control range, indicating that the head resistivity of the embodiment is effectively controlled. Furthermore, from... Figure 4 The standard deviations in the data show that the standard deviation of the comparative example is greater than that of the example. The smaller the standard deviation, the more concentrated the data is, the smaller the fluctuation, and the fluctuation range is reduced by 48%.

[0051] The above-disclosed embodiments are merely preferred embodiments of the present invention and should not be construed as limiting the scope of the invention. Those skilled in the art will understand that implementing all or part of the above-described embodiments and making equivalent changes in accordance with the claims of the present invention are still within the scope of the invention.

Claims

1. A method for systematically controlling the resistivity of heavily doped head, characterized in that, Includes the following steps: S1. Based on the classification of heavily doped head resistivity, systematically set the parameter values ​​for pressure, argon, initial doping multiple, total doping multiple, supplementary doping coefficient, and test temperature time. The heavily doped head resistivity classification includes three levels: Level 1 (2.35-2.65 mΩ·cm), Level 2 (2.65-2.95 mΩ·cm), and Level 3 (3.65-3.95 mΩ·cm). The systematic setting includes setting a uniform pressure value for each category and adjusting the parameters for argon, initial doping multiple, total doping multiple, supplementary doping coefficient, and test temperature time accordingly to compensate for volatilization losses caused by the uniform pressure setting. When setting the argon, a low argon design is used for Level 2 and Level 3 heavily doped head resistivity processes, while a high argon design is used for Level 1 heavily doped head resistivity processes. The initial doping multiple, total doping multiple, and supplementary doping coefficient are also set... At that time, the initial doping multiple, total doping multiple, and supplementary doping coefficient of the second-stage heavy doping head resistivity process are set in series. The initial doping multiple of the third-stage heavy doping head resistivity process is set to be reduced from the initial doping multiple of the second-stage heavy doping head resistivity process, and the initial doping multiple of the first-stage heavy doping head resistivity process is set to be increased from the initial doping multiple of the second-stage heavy doping head resistivity process. The total doping multiple of the third-stage heavy doping head resistivity process is set to be reduced from the total doping multiple of the second-stage heavy doping head resistivity process, and the total doping multiple of the first-stage heavy doping head resistivity process is set to be increased from the total doping multiple of the second-stage heavy doping head resistivity process. The supplementary doping coefficient of the third-stage heavy doping head resistivity process is set to be reduced from the supplementary doping coefficient of the second-stage heavy doping head resistivity process, and the supplementary doping coefficient of the first-stage heavy doping head resistivity process is set to be increased from the supplementary doping coefficient of the second-stage heavy doping head resistivity process. S2. Based on the parameter values ​​systematically set in step S1, crystal pulling is performed to obtain heavily doped monocrystalline silicon with the corresponding head resistivity.

2. The method for systematically regulating the resistivity of heavily doped head according to claim 1, characterized in that, Based on the pressure control level of the release stage controlled by the large-diameter butterfly valve, the control range is determined as (Min1, Max1). The pressure values ​​of various processes are uniformly set as Min1 + (20%-40%) × (Max1-Min1). Among them, Min1 refers to the minimum pressure of the release stage controlled by the large-diameter butterfly valve, and Max1 refers to the maximum pressure of the release stage controlled by the large-diameter butterfly valve.

3. The method for systematically controlling the resistivity of heavily doped head according to claim 1, characterized in that, Based on the argon control level of the equipment during the ignition and release phase, the control range is determined to be (Min2, Max2). The argon for the two processes of the second-level heavy-doped head resistivity and the third-level heavy-doped head resistivity is uniformly set to Min2 + (20%-40%) × (Max2-Min2), and the argon for the first-level heavy-doped head resistivity process is set to Max2. Here, Min2 refers to the minimum argon value of the equipment during the ignition and release phase, and Max2 refers to the maximum argon value of the equipment during the ignition and release phase.

4. The method for systematically controlling the resistivity of heavily doped head according to claim 3, characterized in that, The initial doping ratio for the second-stage heavy doping head resistivity process is set to 6.5-8.0, the total doping ratio is set to 8.0-9.5, and the supplementary doping coefficient is set to 7.0-9.

0.

5. The method for systematically regulating the resistivity of a heavily doped head according to claim 4, characterized in that, The initial doping factor for the third-stage heavy doping head resistivity process is set to be reduced by (20-35)% from the initial doping factor for the second-stage heavy doping head resistivity process, and the initial doping factor for the first-stage heavy doping head resistivity process is set to be increased by (25-40)% from the initial doping factor for the second-stage heavy doping head resistivity process.

6. The method for systematically regulating the resistivity of a heavily doped head according to claim 5, characterized in that, The total doping factor of the third-stage heavy doping head resistivity process is set to be reduced by (15-25)% from the total doping factor of the second-stage heavy doping head resistivity process, and the total doping factor of the first-stage heavy doping head resistivity process is set to be increased by (15-20)% from the total doping factor of the second-stage heavy doping head resistivity process.

7. The method for systematically controlling the resistivity of a heavily doped head according to claim 6, characterized in that, The doping coefficient for the third-stage heavy doping head resistivity process is set to be reduced by (30-40%) from the base of the doping coefficient for the second-stage heavy doping head resistivity process, and the doping coefficient for the first-stage heavy doping head resistivity process is set to be increased by (25-30%) from the base of the doping coefficient for the second-stage heavy doping head resistivity process.

8. The method for systematically controlling the resistivity of a heavily doped head according to claim 7, characterized in that, The temperature testing time for all processes is uniformly set to 4±0.5h.

9. The method for systematically controlling the resistivity of a heavily doped head according to any one of claims 1-8, characterized in that, When the quality of the crystallizer is not up to standard, it is melted back and the crystallizer is re-induced. No additional doping is added this time. And when the total amount of additional doping is greater than or equal to the total doping multiple × the amount of material fed, the next doping will stop.

10. The method for systematically controlling the resistivity of a heavily doped head according to claim 1, characterized in that, The liquid outlet distance for all processes is uniformly set to 25-30mm.

Images