KR20170081562A - Ingot growth control device and control method of it - Google Patents

Abstract

The present invention relates to an ingot growth control apparatus capable of correcting oxygen concentration control variables and crystal defect control variables in a current ingot growth process by correlating them with accumulated data stored in a previous ingot growing process and a control method thereof .
In the ingot growth control apparatus and the control method thereof according to the present invention, the oxygen concentration control variable is corrected in the current process in consideration of the accumulated oxygen concentration performance data in the previous process, and the correction amount of the oxygen concentration control variable or the hot zone change It is possible to correct the crystal defect control data accumulated in the previous process by correcting the correction coefficient of the hot zone according to the correction coefficient of the hot zone and to correct the crystal defect control parameter in the present process in consideration of the corrected crystal defect performance data.

Description

[0001] Ingot growth control device and control method [

The present invention relates to an ingot growth control apparatus capable of correcting oxygen concentration control variables and crystal defect control variables in a current ingot growth process by correlating them with accumulated data stored in a previous ingot growing process and a control method thereof .

Generally, in the single crystal ingot growing apparatus according to the Czochralski method, polycrystalline silicon is loaded in the crucible, and the polycrystalline silicon contained in the crucible is melted by the heat radiated from the heater to form a silicon melt , The seed is gradually rotated while being immersed in the silicon melt, and simultaneously raised.

Generally, a quartz crucible is essentially used for preparing a single crystal ingot using the Czochralski method in order to contain a molten silicon melt by a heater.

However, the quartz crucible is transferred into the SiOx form as it is dissolved in the melt by reflecting the high-temperature silicon melt, and eventually is mixed into the single crystal through the solid-liquid interface.

At this time, the SiOx mixed in the single crystal forms a strengthening effect of the wafer and a micro internal defect (BMD) to act as a gettering site for metal impurities during the semiconductor process, or induce various crystal defects and segregation in the wafer Which is a factor that adversely affects the yield of semiconductor devices.

Therefore, when growing a silicon single crystal using the Czochralski method, it is necessary to appropriately control the oxygen concentration introduced into the crystal through the solid-liquid interface.

According to the conventional technique, the dissolution rate of the quartz crucible, the flow pattern of the silicon melt, and the oxygen concentration of the single crystal ingot are controlled through the Ox volatilization control from the melt surface, but the crystal growth condition is changed And is mainly limited to controlling the axial oxygen concentration of a single crystal ingot.

Japanese Patent Laid-Open Publication No. 2015-089854 discloses a method of manufacturing a silicon single crystal which controls the oxygen concentration of crystals by controlling the volatilization of Ox from the melt surface by controlling the inert gas flow rate introduced between the outer peripheral surface of the single crystal ingot and the bottom opening of the heat shield. .

However, according to the prior art, it is possible to control the axial oxygen concentration of the ingot by controlling the flow rate of the inert gas. However, since the flow rate of the inert gas affects the oxygen concentration at the melt surface, especially at the periphery of the crystal, There is a problem that the uniformity of the directional oxygen concentration is deteriorated.

Therefore, it is required to control the uniformity of the axial oxygen concentration of the ingot, the oxygen concentration level of the ingot, and the uniformity of the radial oxygen concentration of the ingot in accordance with the main variables.

On the other hand, since crystal defects exist in a single crystal and crystal defects depend sensitively on crystal growth and cooling conditions, the kind and distribution of growth defects are controlled by controlling the thermal environment near the growth interface.

Generally, crystal defects are largely divided into vacancy-type and interstitial-type, and the formation of these defects depends on the growth rate V and the radial temperature gradient G Of the total population.

In detail, a vacancy type is formed when the value of V / G exceeds a certain threshold, while an interstitial type defect is formed under the condition of V / G.

Therefore, when crystals are grown in the hot zone, the type, size, and density of the defects present in the crystal are affected by the pulling speed (P / S).

Japanese Patent Registration No. 4428038 discloses one or more of the pulling rate of the current process, the temperature pattern, the heater temperature, the ingot diameter, the crucible rotation number, the seed crystal rotation number, A silicon single crystal manufacturing system capable of realizing uniform quality of ingot length by automatically correcting manufacturing conditions in real time is disclosed.

However, according to the related art, since the actual data is reflected in the current process, it is difficult to exclude the influence of the noise that may occur during the process, and since the performance data interrelated with each other such as the pulling rate and the ingot diameter are individually reflected It is difficult to precisely control the quality of the ingot because it affects both oxygen concentration and crystal defects.

SUMMARY OF THE INVENTION The present invention has been conceived to solve the problems of the prior art described above, and it is an object of the present invention to provide a method and apparatus for adjusting the oxygen concentration control parameter and the crystal defect control parameter in a current ingot growth process, And to provide a control method for the same.

The present invention relates to an ingot growth control apparatus for immersing a seed crystal in a silicon melt contained in a crucible and growing the single crystal ingot as the seed crystal is slowly rotated and pulled up, An oxygen concentration control unit for controlling an oxygen concentration control variable depending on the oxygen concentration; An oxygen concentration calculation unit for correcting the oxygen concentration control parameter in the current process according to the oxygen concentration performance data accumulated in the previous process; A crystal defect control unit for controlling a crystal defect control parameter influencing crystal defects in the radial direction and the axial direction of the ingot in the current process; And a crystal defect calculator for correcting the crystal defect control parameter in the current process in accordance with the crystal defect accumulation data accumulated in the previous process and the correction amount of the oxygen concentration control parameter in the current process.

The present invention also provides an ingot growth control method for immersing a seed crystal in a silicon melt contained in a crucible and growing the single crystal ingot as the seed crystal is gradually rotated and pulled up. And controlling an oxygen concentration control parameter for controlling an oxygen concentration in an ingot growing process in a current process; A second step of storing crystal defect defect data during an ingot growth process in a previous process and controlling a crystal defect control parameter influencing a crystal defect in the ingot growing process in the current process; A third step of correcting the oxygen concentration control parameter in the current process according to the oxygen concentration performance data in a previous process; And a fourth step of correcting the crystal defect control parameter in the current process according to the crystal defect achievement data and the correction amount of the oxygen concentration control parameter in the current process in the previous process.

In the ingot growth control apparatus and the control method thereof according to the present invention, the oxygen concentration control variable is corrected in the current process in consideration of the accumulated oxygen concentration performance data in the previous process, and the correction amount of the oxygen concentration control variable or the hot zone change It is possible to correct the crystal defect control data accumulated in the previous process by correcting the correction coefficient of the hot zone according to the correction coefficient of the hot zone and to correct the crystal defect control parameter in the present process in consideration of the corrected crystal defect performance data.

Therefore, it is possible to minimize the influence of noise occurring in the current process by setting the growth condition of the current process by reflecting the cumulative stored quality result of the previous process, and to accurately reflect the change of the control variable interlocked with the oxygen concentration and the crystal defect There is an advantage that the quality of the ingot included can be precisely controlled.

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a view showing an example of an ingot growing apparatus of the present invention. Fig.
Fig. 2 is a block diagram showing a control device applied to Fig. 1; Fig.
3 is a block diagram showing an oxygen concentration calculation unit applied to FIG.
Fig. 4 is a block diagram showing a crystal defect arithmetic unit applied to Fig. 2; Fig.
5 is a flowchart showing an example of the ingot growth control method of the present invention.
6 is a flowchart showing details of step S1 applied to FIG.
7 is a flowchart showing details of step S5 applied to FIG.

Hereinafter, the present embodiment will be described in detail with reference to the accompanying drawings. It should be understood, however, that the scope of the inventive concept of the present embodiment can be determined from the matters disclosed in the present embodiment, and the spirit of the present invention possessed by the present embodiment is not limited to the embodiments in which addition, Variations.

1 is a view showing an example of an ingot growing apparatus of the present invention.

1, the single crystal ingot growing apparatus of the present invention includes a chamber 110, a crucible 120, a heater 130, a heat insulating material 140, a heat shielding member 150, a cooling tube 160, a diameter measuring sensor 170, and a control device 200. [

The chamber 110 is a closed space for forming a hot zone for growing an ingot from a silicon melt. The crucible 120, the heater 130, and the heat insulating material 140 are embedded in the chamber 110.

The seed crystal driving unit 213 (shown in FIG. 2) and the pulling rate control unit 231 (also shown in FIG. 2), which will be described later, can be installed on the upper side of the chamber 110, (S / R) and pulling rate (P / S) of the seed crystals can be controlled by means of the seed crystal (see Fig. 2).

The crucible 120 is a container containing a silicon melt, and the polycrystalline silicon is melted to form a silicon melt. In the embodiment, the crucible 120 is composed of an inner peripheral portion 121 made of quartz and an outer peripheral portion 122 made of graphite.

The crucible 120 may be rotated and elevated while the single crystal ingot growing process is being performed by connecting a drive shaft and a pedestal to a lower side of the crucible 120. The crucible driving unit 211, 2) and the crucible elevating portion 232 (shown in Fig. 2) can control the rotational speed and the elevating speed of the crucible 120. [

In the embodiment, the crucible elevation portion 232 (shown in FIG. 2) may be formed in the crucible 120 (10) so as to keep the melt gap (M / G) constant as the silicon melt decreases as the single crystal ingot growing process proceeds But it is not limited to this.

The heater 130 is a heat source for heating the crucible 120 and is provided around the crucible 120.

The heat insulating material 140 is provided on the inner circumferential surface of the chamber 110 so as to surround the heater 130 to prevent the heat of the heater 130 from escaping through the side surface of the chamber 110.

The upper end of the heating end member 150 is fixed on the upper side of the heat insulating member 140 and the lower end of the heat insulating member 140 is made of silicone contained in the crucible 120. [ And is maintained to maintain the melt interface and melt gap (M / G).

The cooling tube 160 is formed in a cylindrical case with a tube through which the cooling water flows. The cooling tip 160 has an inner upper end of the chamber 110 to allow the single crystal ingot passing through the heat end member 150 to pass therethrough. Respectively.

Accordingly, the single crystal ingot grown from the silicon melt is cooled while sequentially passing through the heat block member 150 and the cooling pipe 160.

The diameter measuring sensor 170 is installed to face a meniscus formed between the silicon melt interface and the ingot through a transparent window provided on one side of the chamber 110, So that the brightness can be measured.

Also, the diameter of the ingot can be controlled by the automatic diameter control unit 233 (shown in FIG. 2), which will be described below, in consideration of the value measured by the diameter measuring sensor 170.

In addition, a plurality of magnets (not shown) capable of forming a magnetic field are mounted on the outside of the chamber 110, and a convection phenomenon of the silicon melt contained in the crucible 120 is controlled by a magnetic field formed by the magnet And can control the intensity of the magnetic field acting on the silicon melt contained in the crucible 120 by the magnetic field controller 212 (shown in FIG. 3) described below.

The controller 200 controls the oxygen concentration control variable and the crystal defect control parameter to be linked with each other in the current process, reflecting the oxygen concentration data, the crystal defect data, and the hot zone change accumulated in the previous process.

In detail, the controller 200 corrects the oxygen concentration control parameter according to the oxygen concentration data accumulated in the previous process, reflects the correction amount of the oxygen concentration control parameter and the hot zone correction coefficient according to the hot zone change, After correcting cumulative stored crystal defect data and correcting the crystal defect control variable according to the corrected crystal defect data, a detailed configuration will be described below.

Fig. 2 is a block diagram showing the control device applied to Fig. 1. Fig.

2, the controller includes an oxygen concentration control unit 210 for controlling oxygen concentration control parameters for controlling the oxygen concentration in the radial direction and the axial direction of the ingot, A crystal defect controller 230 for controlling a crystal defect control parameter for determining a crystal defect in a radial direction and an axial direction of the ingot, And a crystal defect calculator (240) for correcting the crystal defect control parameter according to the defect accumulation data and the correction amount of the oxygen concentration control parameter.

The oxygen concentration control unit 210 includes a crucible rotation unit 211, a magnetic field control unit 212 and a seed crystal driving unit 213. The oxygen concentration control unit 210 controls the crucible rotation speed C / R, the magnetic field intensity MI, The oxygen concentration control variable including the number (S / R) can be varied according to the target values.

Specifically, the crucible rotation section 211 controls the rotation speed of the crucible 120 (shown in FIG. 1). The crucible rotation section 211 controls the crucible rotation speed C / R, which determines the axial oxygen concentration uniformity of the ingot, (Target crucible rotation: T_C / R).

In the embodiment, the oxygen concentration calculator 220 varies the target crucible rotation speed T_C / R in order to adjust the designed oxygen concentration according to the axial length of the ingot in advance as the ingot growing process proceeds, The target crucible rotation speed T_C / R is corrected in consideration of the number C / R.

The magnetic field control unit 212 controls the maximum magnet field position of the magnet based on the silicon melt interface contained in the crucible 120 (shown in FIG. 1), or the current applied to the magnet The magnitude of the magnetic field (MI) that controls the oxygen concentration level of the ingot is controlled by the target magnetic induction (T_MI).

In the embodiment, the oxygen concentration calculator 220 maintains the target magnetic field intensity T_MI constant and corrects the target magnetic field intensity T_MI in consideration of the magnetic field strength MI of the previous process even if the ingot growing process proceeds .

The seed crystal driving unit 213 rotates a wire connected to the seed crystal. The seed crystal rotation speed S / R, which determines the uniformity of oxygen concentration in the radial direction of the ingot, T_S / R).

In the embodiment, the oxygen concentration calculation unit 220 sets the target seed crystal rotation speed T_S / R relatively high, while the shouldering process of growing the seed crystal from the seed crystal to the diameter of the ingot is performed after the ingot body growth is completed In the tailing process in which the diameter of the ingot is reduced, the target seed crystal rotation speed (T_S / R) is set relatively small and the target seed crystal rotation speed (T_S / R) is calculated in consideration of the seed crystal rotation speed .

The oxygen concentration calculation unit 220 stores the crucible rotation number C / R, the magnetic field strength MI and the seed crystal rotation number S / R according to the axial length of the ingot as oxygen concentration performance data , The target crucible rotation speed (T_C / R) and the target magnetic field intensity (T_MI) are calculated in order to satisfy the axial oxygen concentration uniformity, oxygen concentration level and radial oxygen concentration uniformity of the ingot in the present process, ) And the target seed crystal rotation speed T_S / R, the detailed structure will be described below.

Of course, the oxygen concentration calculation unit 220 accumulates and utilizes the performance data at the previous process at least five times in order to secure the reliability of the oxygen concentration performance data.

The crystal defect controller 230 includes a pulling rate control unit 231, a crucible lifting unit 232 and an automatic diameter control unit 233. The pulling rate control unit 231 controls the pulling rate P / S, the melt gap M / The crystal defect control variable including the set value SP can be varied according to the target values.

In detail, the pulling rate control unit 231 controls the operation of a wire driving unit for pulling up a wire (W: see FIG. 1) to which an ingot is suspended. The pulling rate control unit 231 controls the pulling speed P / S to the target pulling-up speed T_P / S.

In the embodiment, the crystal defect calculator 240 corrects the target pulling rate T_P / S in consideration of the scoring score of the ingot by the copper haze evaluation method in the previous process.

The crucible lifting portion 232 may be formed in the crucible lifting portion 232 so as to maintain the target melt gap T_M / G even if the silicon melt interface in the crucible 120 (FIG. 1) The melt gap (M / G) that controls the balance of the inside / outside ratio as a result of the copper haze scoring is controlled to the target melt gap T_M / G to raise and lower the crucible 120 (shown in FIG.

In the embodiment, the crystal defect calculator 240 corrects the target malt gap (T_M / G) in consideration of the inside / outside ratio of the ingot by the copper haze evaluation method in the previous process.

In addition, the automatic diameter control unit 233 controls the automatic diameter control unit 233 so that the value measured by the brightness of the meniscus by the diameter measuring sensor 170 (shown in Fig. 1) keeps the target set value T_S.P P / S), the set value SP indicating the brightness of the meniscus that controls the scattering of the diameter of the ingot is controlled to the target set value T_S.P.

In the embodiment, when the measured value of the diameter measuring sensor 170 (shown in FIG. 1) is 2000 or more, the automatic diameter control unit 233 determines that the diameter of the ingot is large and increases the pulling rate P / S , And if the measured value of the diameter measuring sensor 170 (shown in FIG. 1) is lower than 2000, it is determined that the diameter of the ingot is reduced and the pulling speed P / S is lowered to control the diameter of the ingot to be constant .

However, as the ingot growth process progresses, the diameter of the ingot increases in the longitudinal direction due to the change of the surrounding environment such as decrease in the melt of the silicon without changing the pulling rate (P / S), thereby affecting the crystal quality and the margin .

Accordingly, the crystal defect calculator 240 decreases the target set value T_S.P to keep the measured value of the diameter measuring sensor 170 (shown in FIG. 1) constant even if the process proceeds, The target set value T_S.P is corrected in consideration of the diameter of the ingot.

The crystal defect calculator 240 stores the scoring score of the ingot by the copper haze evaluation method, the melt gap and the diameter of the ingot as the determination defect performance data for each axial length of the ingot in the previous process, (T_P / S), the target melt gap (T_M / G) and the target set value (T_S) in order to satisfy the balance of the O / .

The copper haze (CUH) evaluation method described above uses copper contamination solution, which is a mixed solution of BOE (Buffered Oxide Etchant) solution, to contaminate Cu at a high concentration on single crystal silicon, to perform short diffusion heat treatment, And the opposite surface of the contaminated surface is visually observed under light condensation or the like to distinguish crystal defect regions.

At this time, the scoring score of the ingot is obtained by scoring the defect-free region according to the copper haze (CUH) evaluation method, and the inside / outside ratio (C / E) of the ingot is determined by the copper haze (CUH) A VDP (Vacancy Dominant Point Defect Zone) region appearing at the center and corners of the monocrystalline ingot is calculated based on the radial length of the defect-free region appearing as heat treatment.

Likewise, the crystal defect calculator 240 accumulates and utilizes the historical data at least five times in the previous process in order to secure the reliability of the crystal defect defect data.

The crystal defect calculator 240 receives the correction amount of the oxygen concentration control variable from the oxygen concentration calculator 220 and calculates the correction amount of the oxygen concentration control variable from the crucible outer periphery 122 The hot zone correction coefficient corresponding to the hot zone change is calculated.

In the embodiment, the hot zone correction coefficient may be a physical property including resistance and thermal conductivity of the material constituting the hot zone, and the physical property value may be changed according to the number of times of use of the hot zone It is calculated as empirically reflected values.

Therefore, the crystal defect calculator 240 corrects the crystal defect performance data by reflecting the correction amount of the oxygen concentration control variable and the hot zone correction coefficient, and then corrects the crystal defect control variable by reflecting the corrected crystal defect performance data , The detailed configuration will be described below.

FIG. 3 is a block diagram showing the oxygen concentration calculation unit applied to FIG.

The oxygen concentration calculator 220 of the present invention calculates the oxygen concentration of the present process by using the oxygen concentration data of the previous process, the crucible rotation rate C / R, the magnetic field strength MI and the seed crystal rotation speed S / 3, a crucible revolution rate calculator 221, a magnetic field calculator 222, a seed crystal revolution rate calculator 223, a crucible revolution rate correction calculator 225, And a seed crystal rotation speed correction computing unit 226.

First, if the axial oxygen concentration uniformity of the ingot is not satisfied in consideration of the cumulative crucible rotation rate (C / R) of the previous process, the crucible rotation rate ratio calculator 221 calculates the crucible rotation rate C / R ) To calculate the target crucible rotation speed T_C / R (see (1)).

If the oxygen concentration level of the ingot is not satisfied in consideration of the accumulated magnetic field intensity MI of the previous process, the magnetic field calculator 222 processes the magnetic field strength MI of the previous process to obtain the target magnetic field intensity T_MI (See ②)

If the radial oxygen concentration uniformity of the ingot is not satisfied in consideration of the cumulative seed crystal rotation speed (S / R) of the previous process, the seed crystal revolution rate calculator 223 calculates the seed crystal revolution number S / R) is processed to calculate the target seed crystal rotation speed T_S / R (refer to (3)).

When the target magnetic field strength T_MI which varies the oxygen concentration level of the ingot is changed, the target crucible rotation speed T_C / R and the target seed crystal rotation speed T_S / R, which determine the axial / radial oxygen concentration uniformity of the ingot, ) Is also changed.

Therefore, the crucible rotation speed correction computing unit 225 corrects the target crucible rotation speed T_C / R in consideration of the axial direction variation value AOi of the target magnetic field intensity MI, and the seed crystal rotation speed corrector 226 ) Corrects the target seed crystal rotation speed T_S / R in consideration of the radial variation value ROi of the target magnetic field strength T_MI (see ④⑤)

Further, if the target seed crystal rotation speed T_S / R that affects the radial oxygen uniformity of the ingot is changed, the target crucible rotation rate T_C / R that determines the axial oxygen concentration uniformity of the ingot also changes.

Therefore, the crucible rotation speed correction computing unit 225 may further correct the target crucible rotation speed T_C / R in consideration of the variation value of the target seed crystal rotation speed T_S / R, but it may be omitted. (See ⑥⑦)

When the oxygen concentration calculator 220 corrects the target crucible rotation speed T_C / R, the target magnetic field strength T_MI and the target seed crystal rotation speed T_S / R, the crucible rotation speed C / R ), The magnetic field intensity (MI), and the seed crystal rotation speed (S / R) are varied, and the respective oxygen concentration control parameter correction amounts can be calculated.

The target crucible rotation speed T_C / R, which determines the axial oxygen concentration uniformity of the ingot by the oxygen concentration calculation unit 220, the target magnetic field strength T_MI which determines the oxygen concentration level of the ingot, By controlling the target seed crystal rotation speed T_S / R that determines the radial direction oxygen concentration uniformity of the ingot, it is possible to relatively easily control the uniformity of the axial oxygen concentration of the ingot, the oxygen concentration level of the ingot, It is possible to precisely control the oxygen concentration by dividing the oxygen concentration into uniformity.

In addition, the oxygen concentration calculation unit 220 reflects the oxygen concentration performance data that influences the oxygen concentration of the ingot accumulated in the previous step, and sets the growth condition of the current process. Thus, The influence of noise can be minimized.

Further, the reliability of the crystal defect defect data can be further improved by reflecting the calculated correction value of the oxygen concentration control parameter into the defect defect performance data, which will be described in detail below.

4 is a block diagram showing a crystal defect arithmetic unit applied to FIG.

The crystal defect calculator 240 of the present invention is configured to reflect the scoring score of the ingot by the copper haze evaluation method, the inner / outer circumference ratio, and the diameter of the ingot in the current process, And includes a scoring operator 241, an inner / outer ratio calculator 242, a set value calculator 243, a pulling rate correction calculator 245, and a diameter correction calculator 246 as shown in FIG.

First, when the hot zone correction coefficient reflecting the change in the physical property of the hot zone according to the number of times of use of the hot zone used in the present process including the oxygen concentration control variable correction amount calculated in the oxygen concentration calculation unit 220 (shown in FIG. 3) , The crystal defect calculating unit 240 corrects the crystal defect performance data of the previous process by reflecting the oxygen concentration control parameter correction amount and the hot zone correction coefficient (refer to a)

Therefore, the scoring score, ingress / egress ratio, and ingot diameter of the ingot by the copper haze evaluation method as the crystal defect performance data of the previous step are corrected in advance and then applied to the crystal defect control of the current process as follows.

Specifically, if the O-band free range is not satisfied on the basis of the scoring score of the ingot by the copper haze evaluation method previously accumulated, the scoring operator 241 processes the pulling rate (P / S) And corrects the pulling speed T_P / S (see (9)).

If the inner / outer circumference balance of the ingot is not satisfied on the basis of the inner / outer circumferential ratio of the ingot by the previously accumulated copper haze evaluation method, the inner / outer circumference ratio calculator 242 calculates the melt gap (M / G ) To calculate the target melt gap T_M / G (see (10)).

If the diameter dispersion of the ingot is not satisfied on the basis of the diameter of the ingot accumulated in advance, the set value calculator 243 processes the set value SP of the previous process and corrects the set value SP to the target set value T_S.P (See ⑪)

However, if the melt gap (M / G) that determines the ratio of ingot inner / outer circumference is changed, the pulling speed P / S of the ingot influencing the O-band free range and the set value SP ).

Therefore, the pulling rate correction calculator 245 calculates the target pulling speed T_P (S / P) by reflecting the change (P / S) of the pulling speed P / S of the ingot according to the melt gap / S), and the diameter auxiliary operator 246 likewise corrects the diameter of the ingot according to the melt gap (M / G) (Dc: Diameter conversion) to obtain the target set value T_S.P ) (See ⑫ ⑬)

Further, if the set value S.P that affects the scattering of the diameter of the ingot is changed, the pulling speed P / S of the ingot which influences the O-band free range also fluctuates.

Therefore, the pulling rate correction calculator 245 further corrects the target pulling speed T_P / S by reflecting the change of the pulling speed of the ingot according to the set value S. (refer to ⑭ ⑮)

As described above, the crystal defect calculator 240 calculates the target pulling speed T_P / S that determines the O-band free range of the ingot, the target malt gap T_M / G that determines the inward / , By controlling the target set value T_S.P that affects the scattering of the diameter of the ingot, the O-band free range of the ingot, the inner / outer circumference balance of the ingot and the diameter scattering of the ingot can be determined relatively easily during the ingot growing step The defect can be precisely controlled.

In addition, the crystal defect calculating unit 240 reflects the crystal defect accumulation data that influences the crystal defects accumulated in the previous steps, and sets the growth conditions of the current process. The influence of noise can be minimized.

Further, the crystal defect calculator 240 corrects the crystal defect performance data of the previous process in consideration of the influence of the oxygen concentration control parameter correction amount and the hot zone, and then reflects the crystal defect defect control in the current process, Can be controlled.

5 is a flowchart showing an example of the ingot growth control method of the present invention.

In the ingot growth control method of the present invention, the oxygen concentration control parameter is controlled according to a predetermined target value during the current ingot growing process. As shown in FIG. 5, The oxygen concentration control parameter is corrected and will be described in detail below (see S1).

In the embodiment, the oxygen concentration data may be the crucible rotation speed (C / R), the magnetic field strength (MI) and the seed crystal rotation speed (S / R) according to the axial length of the ingot. Values stored at least five times in the process can be applied.

The oxygen concentration control parameter includes a target crucible rotation rate (T_C / R) that determines the axial oxygen concentration uniformity of the ingot, a target magnetic field strength (T_MI) that determines the oxygen concentration level of the ingot, The target seed crystal rotation speed T_S / R that determines the target seed crystal rotation speed.

Therefore, the uniformity of the oxygen concentration in the axial direction of the ingot, the oxygen concentration level of the ingot, and the uniformity of the oxygen concentration in the radial direction of the ingot can be separately controlled by using the oxygen concentration data of the previous process.

Next, the oxygen concentration control variable correction amount is calculated in real time during the present process (S2 calculation)

In an embodiment, the oxygen concentration control parameter correction amount may be a target crucible rotation speed (T_C / R) variation value, a target magnetic field strength (T_MI) variation value, and a target seed crystal rotation speed (T_S / R) variation value.

Next, the hot zone correction coefficient according to the hot zone change is calculated (refer to S3)

In the embodiment, the hot zone may be a heater and a cooling member as well as an outer circumferential portion of the crucible, and the hot zone correction coefficient may be calculated by empirically reflecting a change in physical property including a resistance and a thermal conductivity for each component of the hot zone .

For example, in a case where a heater is replaced among a plurality of batches used for about three years, the data is synthesized to calculate the oxygen concentration and crystal defect level for each ingot length, and then divided by the number of times of use, Can be unitized by length.

Of course, if the number of times of use before and after the heater change is different, the correction coefficient of the heater can be calculated by multiplying the difference in the use frequency before and after the heater replacement by the heater coefficient, and thereby the oxygen concentration and the crystal defect variation in the present process can be predicted .

Next, the crystal defect performance data of the previous process is corrected by reflecting the oxygen concentration control parameter correction amount and the hot zone correction coefficient (see S4)

In the embodiment, the crystal defect achievement data may be a scoring score of the ingot by the copper haze evaluation method, an inner / outer circumference ratio, and the ingot diameter according to the axial length of the ingot. In order to increase the reliability, The stored value can be applied.

However, as the current process progresses, the oxygen concentration control variable is changed, and since the replacement and use frequency of the hot zone may be changed, the defect defect performance data of the previous process should also be corrected to reflect the defect defect control of the present process .

Next, the crystal defect control variable according to the corrected crystal defect performance data is corrected, and will be described in detail below (see S5).

In the embodiment, the crystal defect control parameter includes a target pulling speed (T_P / S) that determines the O-band free range of the ingot, a target melt gap (T_M / G) that determines the in / And may be a target set value T_S.P that affects the scattering.

Therefore, the O-band free range of the ingot, the inner / outer balance of the ingot, and the diameter dispersion of the ingot can be separately controlled by using one variable, using the crystal defect defect data of the previous process.

Furthermore, the correction amount of the oxygen concentration control parameter in the present process and the hot zone correction coefficient according to the hot zone replacement and change are reflected in real time to correct the crystal defect performance data of the previous process, It is possible to precisely and precisely control crystal defects.

FIG. 6 is a flowchart showing details of step S1 applied to FIG.

As shown in FIG. 6, the crucible rotation rate (C / R) and the magnetic field strength (MI) of the previous process, which are oxygen concentration performance data, And seed crystal rotation speed (S / R) are cumulatively stored (see S111, S121 and S131)

First, when the axial oxygen uniformity of the ingot is satisfied based on the crucible rotation speed (C / R) of the previous process, the target crucible rotation speed (T_C / R) is calculated based on the crucible rotation speed The target crucible rotation rate T_C / R is primarily corrected through the crucible rotation rate calculation based on the crucible rotation rate (C / R) of the previous process (S112, S113, S114). Reference)

If the oxygen concentration level of the ingot is satisfied on the basis of the magnetic field intensity MI of the previous process, the target magnetic field intensity T_MI is firstly corrected based on the magnetic field intensity MI of the previous process, The target magnetic field strength T_MI of the current process is primarily corrected through the magnetic field intensity calculation based on the magnetic field intensity MI of the magnetic field strength of the current process (see S122, S123, and S124).

At this time, if the hot zone is changed, the target field strength T_MI of the current process is corrected by the magnetic field strength calculation based on the magnetic field strength MI of the previous process, The final target magnetic field strength T_MI of the current process is calculated on the basis of the calculated target magnetic field strength T_MI (see S124 ', S125 and S126)

Also, if the radial oxygen uniformity of the ingot is satisfied on the basis of the seed crystal rotation speed (S / R) of the previous process, the target seed crystal rotation speed T_S / R), but if not, the target seed determination revolution speed (T_S / R) is firstly corrected through the seed determination revolution speed ratio calculation based on the seed determination revolution speed (S / R) of the previous process. (See S132, S133, and S134)

However, if the final target magnetic field strength T_MI, which determines the oxygen concentration level of the ingot, fluctuates, it affects the axial / radial concentration of the ingot, and the target crucible rotation speed T_C / R, The number of revolutions T_S / R must be changed.

Therefore, the target crucible rotation speed T_C / R is secondarily corrected through the crucible rotation speed correction calculation in consideration of the axial variation value AOi of the target magnetic field strength T_MI (see S127, S128, and S115).

The target seed crystal rotation speed T_S / R is secondarily corrected through the seed crystal rotation speed correction calculation in consideration of the radial variation value ROi of the target magnetic field intensity T_MI, and the final target seed determination (S_S / R) (see S127, S129, S135, and S136)

However, if the target seed crystal rotation speed T_S / R that determines the uniformity of the oxygen concentration in the ingot is changed, the target crucible rotation speed T_C / R, which depends on the axial oxygen concentration uniformity of the ingot, do.

Therefore, it is possible to further correct the target crucible rotation speed T_C / R through the crucible rotation speed correction operation in consideration of the variation value of the target seed determination rotation speed T_S / R, It is acceptable.

As described above, the target crucible rotation speed T_C / R is corrected based on the crucible rotation speed C / R of the previous process, the target magnetic field strength T_MI of the current process, and the target seed crystal rotation speed T_S / Next, when the target crucible rotation speed T_C / R satisfies the axial oxygen uniformity of the ingot, the final target crucible rotation speed T_C / R of the present process is calculated (see S116 and S117)

Therefore, it is possible to control the uniformity of the axial oxygen concentration of the ingot, the oxygen concentration level of the ingot, and the uniformity of the radial oxygen concentration of the ingot separately by using one parameter, using the oxygen concentration data of the previous process, It is possible to precisely control the oxygen concentration by complementing the influence of the change and to minimize the occurrence of noise during the oxygen concentration control of the present process by using reliable oxygen concentration performance data.

FIG. 7 is a flowchart illustrating in detail the step S5 of FIG.

The process of controlling the crystal defect control variables according to the corrected crystal defect performance data according to the present invention will be described with reference to FIG. 7. As shown in FIG. 7, the scoring score, the inner / outer circumference ratio, (See S111, S121 and S131)

Of course, the crystal defect performance data as described above is reflected by reflecting the oxygen concentration control parameter correction amount and the hot zone correction coefficient in the present process, and then the corrected crystal defect performance data is stored. (See S4)

First, if the current pulling rate P / S satisfies the scoring score, the current pulling rate P / S is first corrected to the target pulling rate T_P / S, and if the current pulling rate P / S is unsatisfied, The target pulling-up speed T_P / S is primarily corrected to the pulling-up speed P / S (refer to S512, S512, S513, and S114)

If the current melt gap M / G satisfies the inner / outer circumference ratio, the current melt gap M / G is calculated as the target melt gap T_M / G, and if the current melt gap M / G is unsatisfied, The target melt gap T_M / G is calculated by the melt gap (M / G) calculated through the above-described calculation (S522, S523, S524)

If the current ingot diameter satisfies the set value SP, the current set value SP is calculated as the target set value T_S.P, and if it is unsatisfied, the set value SP calculated through the set value calculation ) To correct the target set value T_S.P (see S532, S533, and S534)

However, if the melt gap (M / G) that influences the ingress / egress ratio of the ingot fluctuates, the O-band free range of the ingot and the diameter scattering of the ingot are affected, and the target pulling rate T_P / S ) And the target set value T_S.P must be changed.

Therefore, the target pulling-up speed T_P / S is secondarily corrected through the pulling rate correction calculation in consideration of the target pulling-up speed P / S variation value Tc according to the change in the target melt gap T_M / G. S525, S526, and S515)

Further, the target set value T_S.P is secondarily corrected through the diameter correction calculation in consideration of the diameter variation value Dc of the ingot according to the change of the target melt gap T_M / G, and the final target set value (T_S.P) (see S525, S527, S535, and S536)

Of course, since the pulling speed P / S of the ingot changes according to the diameter of the ingot, the target pulling rate T_P / S is changed to the second pulling rate T_P / S by the pulling rate correction calculation, taking into account the calculated target set value T_S.P (Refer to S536, S526, and S515)

As described above, the step of correcting the target pulling-up speed T_P / S in accordance with the crystal defect defect data may be performed by calculating the target pulling rate T_P / S by reflecting the scoring score, the inner / And further corrects the target pulling rate T_P / S reflecting the pulling speed P / S of the ingot interlocked with the melt gap M / G and the diameter of the ingot.

Therefore, the target pulling rate T_P / S is corrected as the current process progresses, and the current pulling rate P / S is controlled along the target pulling rate T_P / S.

At this time, if the current pulling rate P / S satisfies the scoring score, the current pulling rate P / S is calculated as the target pulling rate T_P / S, and the current pulling rate P / (S516 and S517)

However, if the current pulling rate P / S is unsatisfactory for the scoring score, the above process is repeated to correct the target pulling rate T_P / S.

Therefore, the O-band free range of the ingot, the inner / outer balance of the ingot, and the diameter dispersion of the ingot can be controlled separately by using one variable by using the crystal defect accumulation data of the previous process. It is possible to precisely control the crystal defects and to minimize the occurrence of noise during the crystal defect control of the current process by using the reliable crystal defect performance data.

Furthermore, crystal defect defects can be more precisely controlled by correcting the crystal defect defect data of the previous process in consideration of the influence of the oxygen concentration control parameter correction amount and the hot zone, and then reflecting the data in the crystal defect control in the present process.

110: chamber 120: crucible
130: heater 140: insulation
150: heat terminal member 160: cooling pipe
170: Diameter measuring sensor 200: Control device
210: oxygen concentration control unit 220: oxygen concentration calculation unit
230: crystal defect controller 240: crystal defect calculator

Claims (31)

1. An ingot growth control device for immersing a seed crystal in a silicon melt contained in a crucible and growing the single crystal ingot as the seed crystal is slowly rotated and pulled up,
An oxygen concentration control unit for controlling an oxygen concentration control parameter for controlling an oxygen concentration in a radial direction and an axial direction of the ingot in a current process;
An oxygen concentration calculation unit for correcting the oxygen concentration control parameter in the current process according to the oxygen concentration performance data accumulated in the previous process;
A crystal defect control unit for controlling a crystal defect control parameter influencing crystal defects in the radial direction and the axial direction of the ingot in the current process; And
And a crystal defect calculating unit for correcting the crystal defect control parameter in the current process in accordance with the crystal defect defect data accumulated in the previous process and the correction amount of the oxygen concentration control parameter in the current process. The method according to claim 1,
Wherein the oxygen concentration control unit comprises:
A crucible rotating part for controlling the rotation of the crucible to a target crucible rotation speed (T_C / R)
A magnetic field control unit for controlling the intensity MI of the magnetic field acting on the silicon melt to a target magnetic field intensity T_MI;
And a seed crystal driving section for controlling the rotation of the seed crystal to a target seed crystal rotation speed (T_S / R). 3. The method of claim 2,
Wherein the oxygen concentration calculation unit comprises:
The oxygen concentration data accumulated cumulatively for each axial length of the ingot at the time of five previous ingot growth,
Wherein the oxygen concentration performance data includes the crucible rotation speed (C / R), the magnetic field strength (MI), and the seed crystal rotation speed (S / R). The method of claim 3,
Wherein the oxygen concentration calculation unit comprises:
A crucible rotation rate ratio calculator for correcting the target crucible rotation rate T_C / R so as to satisfy the axial oxygen concentration uniformity of the ingot on the basis of the previously accumulated crucible rotation rate (C / R)
A magnetic field calculator for correcting the target magnetic field strength (T_MI) so as to satisfy the oxygen concentration level of the ingot on the basis of the previously stored magnetic field strength (MI)
And a seed crystal rotation speed ratio calculator for correcting the target seed crystal rotation speed (T_S / R) so as to satisfy the radial direction oxygen concentration uniformity of the ingot on the basis of the previously stored seed crystal rotation (S / R) Control device. 5. The method of claim 4,
The magnetic field calculator comprises:
And further corrects the target field strength (T_MI) by reflecting a correction coefficient according to the number of times of use of the hot zone. 6. The method of claim 5,
Wherein the oxygen concentration calculation unit comprises:
And a crucible rotation speed correction computing unit for further correcting the corrected target crucible rotation rate (T_C / R) in the crucible rotation rate ratio computing unit by reflecting the corrected target magnetic field strength (T_MI) in the magnetic field computing unit. Device. 6. The method of claim 5,
Wherein the oxygen concentration calculation unit comprises:
Further comprising a seed crystal rotation speed correction computing unit for further correcting the corrected target seed crystal rotation speed (T_S / R) by the seed crystal rotation rate rate calculator reflecting the corrected target magnetic field strength (T_MI) in the magnetic field computing unit Ingot growth control device. The method according to claim 1,
Wherein the crystal defect controller comprises:
A pulling rate control unit for controlling the pulling rate P / S of the ingot in accordance with the target pulling rate T_P / S,
A crucible for controlling the elevation of the crucible in order to control the melt gap (M / G), which is the interval between the heat end member cooling the ingot pulled in the silicon melt and the silicon melt interface, according to the target melt gap T_M / G A lifting portion,
And an automatic diameter control section for controlling the diameter of the ingot so as to control a set value (SP) indicating the brightness of a meniscus formed between the silicon melt interface and the ingot according to a target set value (T_S.P) Growth Control Device. 9. The method of claim 8,
Wherein the crystal defect calculating unit comprises:
The crystal defect performance data accumulated cumulatively for each axial length of the ingot at the time of five previous ingot growth,
Wherein the crystal defect achievement data includes an ingot scoring score by a copper haze evaluation method, an inside / outside ratio by a copper haze evaluation method, and a diameter of an ingot. 10. The method of claim 9,
Wherein the crystal defect calculating unit comprises:
A scoring operator for correcting the target pulling-up speed T_P / S so as to satisfy the O-band free range on the basis of the scoring score of the ingot by the copper haze evaluation method previously accumulated,
An inner / outer ratio calculator for correcting the target melt gap T_M / G so as to satisfy the inner / outer circumference balance of the ingot on the basis of the inner / outer circumferential ratio of the ingot by the previously accumulated cumulative copper haze evaluation method,
And a set value computing unit for correcting the target set value T_S.P so as to satisfy the diameter dispersion of the ingot based on the diameter of the ingot accumulated in advance. 11. The method of claim 10,
Wherein the crystal defect calculating unit comprises:
A pulling rate correction processor for further correcting the target pulling rate T_P / S by reflecting the change of the target pulling rate T_P / S in accordance with the target melt gap T_M / G calculated by the inside / Wherein the ingot growth control device further comprises: 12. The method of claim 11,
Wherein the crystal defect calculating unit comprises:
Further comprising a diameter correction computing unit for further correcting the target setting value (T_S.P) by reflecting a change in diameter according to a target melt gap (T_M / G) calculated by the inner / outer peripheral ratio computing unit. 13. The method of claim 12,
The pulling rate correction calculator comprises:
And further corrects the target pulling-up speed (T_P / S) by reflecting the target set value (T_S.P) corrected by the diameter correction computing unit. 14. The method according to any one of claims 1 to 13,
Wherein the crystal defect calculating unit comprises:
And corrects the crystal defect control variable by reflecting a correction coefficient according to the number of times of use of the hot zone. 15. The method of claim 14,
In the hot zone,
A crucible supporting section for surrounding the crucible; a heater for heating the crucible; and a heat block member for cooling the ingot pulled up in the silicon melt,
Wherein the crystal defect calculating unit comprises:
And the correction coefficient is calculated in consideration of the property value of the hot zone in accordance with the number of times of use of the hot zone. 16. The method of claim 15,
Wherein the crystal defect calculating unit comprises:
And reflects the resistance and the thermal conductivity of the hot zone to the physical properties of the hot zone. 1. An ingot growth control method for immersing a seed crystal in a silicon melt contained in a crucible and growing the single crystal ingot as the seed crystal is gradually rotated and pulled up,
A first step of storing oxygen concentration data in the ingot growing step in the previous step and controlling an oxygen concentration control variable that influences the oxygen concentration in the ingot growing step in the current step;
A second step of storing crystal defect defect data during an ingot growth process in a previous process and controlling a crystal defect control parameter influencing a crystal defect in the ingot growing process in the current process;
A third step of correcting the oxygen concentration control parameter in the current process according to the oxygen concentration performance data in a previous process; And
And a fourth step of correcting the crystal defect control parameter in the current process in accordance with the crystal defect achievement data and the correction amount of the oxygen concentration control parameter in the current process in the previous process. 18. The method of claim 17,
In the first step,
Accumulating the oxygen concentration performance data for each axial length of the ingot at the time of five previous ingot growth,
The oxygen concentration performance data includes,
(C / R) of the crucible, an intensity (MI) of a magnetic field acting on the silicon melt, and the seed crystal rotation speed (S / R). 19. The method of claim 18,
In the first step,
During the current ingot growing process,
Controlling the rotation of the crucible in accordance with the target crucible revolution number T_C / R,
Controlling the intensity of the magnetic field acting on the silicon melt according to the target field strength (T_MI)
And controlling the rotation of the seed crystal according to the target seed crystal rotation speed (T_S / R). 20. The method of claim 19,
In the third step,
A first step of correcting the target crucible rotation speed T_C / R so as to satisfy the axial oxygen concentration uniformity of the ingot on the basis of the previously accumulated crucible rotation speed (C / R)
A second step of correcting the target magnetic field intensity T_MI so as to satisfy the oxygen concentration level of the ingot on the basis of the previously stored magnetic field intensity MI,
And a third step of correcting the target seed crystal rotation speed (T_S / R) so as to satisfy the radial oxygen concentration uniformity of the ingot on the basis of the previously stored seed crystal rotation speed (S / R). 21. The method of claim 20,
In the third step,
And a fourth step of further correcting the target field strength (T_MI) by reflecting a correction coefficient according to the number of times of use of the hot zone. 22. The method of claim 21,
In the third step,
Further comprising a fifth step of further correcting the corrected target crucible rotation speed (T_C / R) in the first step by reflecting the corrected target magnetic field strength (T_MI) in the fourth step. 22. The method of claim 21,
In the third step,
Further comprising: a sixth step of further correcting the target seed determination revolution speed (T_S / R) corrected in the third step by reflecting the corrected target magnetic field strength (T_MI) in the fourth step. 18. The method of claim 17,
The second step comprises:
And cumulatively storing the crystal defect performance data for each axial length of the ingot at the time of five previous ingot growth,
The crystal defect performer data includes:
A scoring score of the ingot by the copper haze evaluation method, an inner / outer peripheral ratio by the copper haze evaluation method, and a diameter of the ingot. 25. The method of claim 24,
The second step comprises:
During the current ingot growing process,
Controlling the pulling of the ingot according to the target pulling speed T_P / S,
Controlling the lifting and lowering of the crucible in accordance with the target melt gap T_M / G,
And controlling the diameter of the ingot in accordance with a target set value T_S.P indicating the brightness of the meniscus. 26. The method of claim 25,
In the fourth step,
(T_P / S) so as to satisfy an O-band free range by reflecting a correction amount of the oxygen concentration control parameter based on a scoring score of an ingot by a copper haze evaluation method previously accumulated,
(T_M / G) that corrects the target melt gap (T_M / G) so as to satisfy the inner / outer circumferential balance of the ingot by reflecting the correction amount of the oxygen concentration control variable based on the inner / outer circumferential ratio of the ingot by the cumulative copper- Process,
And correcting the target set value (T_S.P) so as to satisfy the diameter dispersion of the ingot by reflecting the correction amount of the oxygen concentration control parameter based on the diameter of the ingot accumulated in advance. 27. The method of claim 26,
In the fourth step,
The target pulling speed T_P / S corrected in the first process is further corrected by reflecting the change in the target pulling rate T_P / S according to the target melt gap T_M / G calculated in the second process And further comprising a fourth step. 28. The method of claim 27,
In the fourth step,
Further comprising a fifth step of further correcting the corrected target setting value T_S.P in the third step by reflecting a change in diameter according to the target melt gap T_M / G calculated in the second step Ingot growth control method. 29. The method of claim 28,
In the fourth step,
Further comprising a sixth step of further correcting the target pulling up rate (T_P / S) corrected in the fourth step by reflecting the corrected target set value (T_S.P) in the fifth step. 30. The method according to any one of claims 17 to 29,
In the fourth step,
And correcting the crystal defect control parameter by reflecting a correction coefficient according to the number of times of use of the hot zone. 31. The method of claim 30,
The correction coefficient of the hot zone is calculated by:
Wherein the resistance and the thermal conductivity of the hot zone vary depending on the number of times of use.

Images