KR101105479B1 - Method and Apparatus for manufacturing high quality silicon single crystal - Google Patents

Abstract

The present invention discloses a method and apparatus for producing high quality silicon single crystals. The method for producing a silicon single crystal according to the present invention is a method for producing a silicon single crystal by using a Czochralski method in which a seed crystal is immersed in a silicon melt, and then the seed crystal is gradually raised to grow a single crystal. Melt gap change amount is calculated by weight change or diameter change to correct melt gap change, thereby maintaining V / G of the solid-liquid interface within a flawless margin.

According to the present invention, it is possible to control the V / G of the solid-liquid interface within a flawless margin by correcting the melt gap change according to the change in the consumption amount of the silicon melt in the shoulder process or the body process. Therefore, even when manufacturing a large diameter silicon single crystal of 300 mm or more having a large change in silicon melt consumption, it is possible to produce a high quality silicon single crystal without V defects or I defects.

V defect, I defect, defect margin, melt gap, melt gap compensation

Description

Method and Apparatus for manufacturing high quality silicon single crystal

The present invention relates to a method for manufacturing a silicon single crystal, and more particularly, to a silicon single crystal grown by Czochralski (abbreviated as CZ) method of the high quality silicon single crystal that can minimize the occurrence of defects in the entire length of the single crystal It relates to a manufacturing method and apparatus.

In general, silicon single crystals used as materials for producing electronic components such as semiconductors are manufactured by the CZ method. In the CZ method, polycrystalline silicon is poured into a quartz crucible and melted, followed by immersing seed crystals in the molten silicon melt and slowly pulling the crystals. For a detailed description, see S.wolf and R.N. Tauber's paper, Silicon Processing for the VLSI Era, volume 1, Lattice Press (1986), Sunset Beach, CA.

When the silicon single crystal is grown by the CZ method, vacancy and interstitial silicon are introduced into the single crystal through the solid-liquid interface. When the concentration of bacon and lattice silicon introduced into a single crystal reaches a supersaturated state, the bacon and lattice silicon diffuse and aggregate to form baconic defects (hereinafter referred to as V defects) and interstitial defects (I defects). ).

However, since V defects and I defects adversely affect the characteristics of the wafer, it is necessary to suppress formation of V defects and I defects as much as possible during single crystal growth. Conventionally, in order to suppress the occurrence of V defects and I defects, a method of controlling V / G, which is the ratio of the pulling rate V of the single crystal and the temperature gradient G at the solid-liquid interface, is mainly used.

For example, US6,045,610 and JP8-330316 minimize V / G deviation in the radial direction when growing silicon single crystal by CZ method and adjust pulling speed V to make V / G perfect. Margin 'V / G _lower ~ It is _disclosed that a defect-free single crystal can be produced by controlling to be within V / G _upper '.

The range of V / G where V defects and I defects do not occur is called a defect margin. If V / G is greater than the upper limit of the flawless margin, excessive Bacon will be introduced into single crystal, causing V defect. If V / G becomes less than the lower limit of the defect-free margin, the lattice silicon tends to be excessively introduced into the single crystal, causing an I defect.

Among the parameters included in the V / G, G is controlled through the hot-zone design of the single crystal growth apparatus, mainly by adjusting the melt gap, which is a gap between the silicon melt and the heat shield means. Here, the heat shield means refers to a heat shield member that prevents external emission of radiant heat generated from the surface of the single crystal in order to reduce the temperature variation between the surface and the center of the single crystal when the silicon single crystal is pulled up. The melt gap is determined such that the temperature gradients of the single crystal center and the single crystal surface satisfy the same conditions. That is, the temperature gradient change profile of the center and the surface of the single crystal is obtained according to the change of the melt gap, and the melt gap at the point where the two temperature gradient profiles intersect is set as the process condition.

However, as silicon single crystals have been largely cured to more than 300 mm, it is increasingly difficult to control V / G within a flawless margin. This is because the larger the diameter of the silicon single crystal is, the more the silicon gap is consumed during the single crystal pulling process, thereby increasing the variation of the melt gap. When the melt gap is changed, the G value is accompanied. In this case, even if the single crystal pulling speed V is controlled at the zero pulling speed, V / G is out of the defect free margin, which causes V defects or I defects in the single crystal. . Therefore, when growing a large-diameter silicon single crystal, it is necessary to control the melt gap along with the pulling speed of the single crystal by the length of the single crystal to control the V / G within the defect free margin throughout the single crystal growth process.

The present invention was devised to solve the above-mentioned problems of the prior art, and by controlling the melt gap during the growth of a large diameter silicon single crystal of 300 mm or more by the CZ method, the single liquid crystal is controlled by controlling the V / G of the solid-liquid interface within a flawless margin. To provide a high quality silicon single crystal manufacturing method and apparatus capable of realizing defects in length.

In order to achieve the above technical problem, a high quality silicon single crystal manufacturing method is a silicon single crystal manufacturing method using a Czochralski method in which a seed crystal is immersed in a silicon melt and then the seed crystal is gradually raised to grow a single crystal. In this case, the melt gap change amount is calculated by weight change or diameter change of the silicon single crystal to correct the melt gap change, thereby maintaining V / G of the solid-liquid interface within a defect free margin.

According to an aspect of the present invention, the melt gap correction step may include: growing a single crystal shoulder by increasing the diameter of the silicon single crystal to a target diameter through controlling the pulling speed of the single crystal; Calculating a weight change amount of the grown shoulder based on the target weight of the shoulder; And calculating a melt gap change amount using the calculated shoulder weight change amount.

According to another aspect of the invention, the melt gap correction step, the step of measuring the diameter of the single crystal body while pulling the body of the single crystal; Calculating a body diameter change amount by comparing the measured body diameter with a target diameter; And calculating a melt gap change amount using the calculated body diameter change amount.

High quality silicon single crystal production apparatus for achieving the above technical problem, the quartz crucible containing silicon melt; Quartz crucible rotating means for moving the quartz crucible up and down while rotating the quartz crucible; Single crystal pulling means for pulling a single crystal from the silicon melt using a seed crystal wire; Heat shield means for shielding heat emitted from an outer circumferential surface of the single crystal being pulled up and forming a melt gap with a surface of the silicon melt; A diameter sensor measuring the diameter of the single crystal to be pulled and outputting diameter data; And receiving diameter data from the diameter sensor during growth of the single crystal body, calculating a melt gap change according to the consumption change of the silicon melt according to the change in the body diameter in comparison with the received diameter and the target diameter, and calculating the melt gap change amount. Control means for correcting the melt gap accordingly.

Preferably, the apparatus of the present invention further comprises a weight sensor for measuring the weight of the single crystal applied through the wire and outputting weight data, wherein the control means has a shoulder process for increasing the diameter of the single crystal to a target diameter. The shoulder weight change amount is calculated by comparing the weight of the grown shoulder and the target weight, the melt gap change amount is calculated according to the calculated shoulder weight change amount, and the melt gap is corrected according to the calculated melt gap change amount.

Preferably, the correction of the melt gap is made in a direction to cancel the calculated melt gap change amount.

Alternatively, the correction of the melt gap is made when the amount of melt gap change is greater than the threshold. At this time, the correction of the melt gap is made in a direction to cancel the melt gap change amount.

In accordance with the present invention, melt gap correction means converging the melt gap to a defect-free melt gap with a specific value or changing the melt gap to a value within the margin of the defect-free melt gap with a specific range.

According to the present invention, the V / G of the solid-liquid interface can be controlled within a flawless margin by correcting a melt gap change caused by a change in the consumption amount of silicon melt in a shoulder process or a body process. Therefore, even when manufacturing a large diameter silicon single crystal of 300 mm or more having a large change in silicon melt consumption, it is possible to produce a high quality silicon single crystal without V defects or I defects.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Prior to this, terms or words used in the specification and claims should not be construed as having a conventional or dictionary meaning, and the inventors should properly explain the concept of terms in order to best explain their own invention. Based on the principle that can be defined, it should be interpreted as meaning and concept corresponding to the technical idea of the present invention. Therefore, the embodiments described in the specification and the drawings shown in the drawings are only the most preferred embodiment of the present invention and do not represent all of the technical idea of the present invention, various modifications that can be replaced at the time of the present application It should be understood that there may be equivalents and variations.

1 is a cross-sectional view of an apparatus showing a schematic configuration of a single crystal production apparatus used in the silicon single crystal production method according to the present invention.

Referring to FIG. 1, the single crystal manufacturing apparatus includes a quartz crucible 10 in which a silicon melt (SM) in which polycrystalline silicon is melted is accommodated; A crucible housing 20 surrounding the outer circumferential surface of the quartz crucible 10 and supporting the outer circumferential surface of the quartz crucible 10 in a predetermined form; A crucible rotating means (30) installed at the bottom of the crucible housing (20) to rotate the quartz crucible (10) together with the housing (20) and to move the quartz crucible (10) up and down to adjust the melt gap (MG); Heating means 40 for heating the quartz crucible 10 spaced a predetermined distance from a side wall of the crucible housing 20; Heat insulation means (50) installed on the outside of the heating means (40) to prevent heat generated from the heating means (40) from flowing out; Single crystal pulling means (70) for pulling the single crystal (C) from the silicon melt (SM) accommodated in the quartz crucible (10) by using a seed crystal (60) attached to a wire; And heat shield means 80 for reflecting heat emitted from the single crystal C at a predetermined distance from the outer circumferential surface of the single crystal C pulled by the single crystal pulling means 70. Since these components are typical components of the single crystal manufacturing apparatus using the CZ method, which is well known in the art, detailed description of each component will be omitted.

The single crystal manufacturing apparatus according to the present invention includes a weight sensor 100 for detecting the weight of the single crystal pulled by the single crystal pulling means (70); Diameter sensor 110 for measuring the diameter of the single crystal (C) during the pulling up of the single crystal (C); And control means 140 for controlling the size of the melt gap MG by controlling the driving means 130 to drive the quartz crucible rotating means 30 up and down.

The weight sensor 100 detects the weight of the single crystal applied to the single crystal pulling wire when the single crystal C is pulled, and outputs the weight data of the single crystal to the control means 140. To this end, the weight sensor 100 is electrically connected to the control means 140. As the weight sensor 100, a load cell may be used. A load cell is a sensor element that converts the weight of a measurement object into an electrical signal. The load cell may use a pancake & miniature load cell. For example, an MNT model manufactured by CAS Korea may be used. However, the present invention is not limited by the specific type of the weight sensor 100. The weight sensor 100 is installed in the single crystal pulling means (not shown) located above the single crystal growth apparatus.

The diameter sensor 110 measures the diameter of the single crystal (C) when the single crystal (C) is pulled out and outputs diameter data to the control means (140). To this end, the diameter sensor 110 is electrically connected to the control means 140. The diameter sensor 110 may be an Automatic Optic Pyrometer Series 1100 manufactured by Ircon, Inc. of the United States, but the present invention is not limited thereto.

The control unit 140 detects a shoulder weight change amount using the weight sensor 100 in the shoulder process. Sensing the shoulder weight change can indirectly indicate the change in the consumption of the silicone melt, which is directly related to the change in the melt gap. Here, the shoulder step is a step of gradually increasing the diameter of the single crystal (C) to the target diameter before growing the single crystal (C) as an initial step of the single crystal pulling step. In FIG. 1, the single crystal part pulled up by the shoulder process is represented by S. In FIG.

The shoulder weight change amount is a change amount based on a target weight. The target weight refers to the target weight of the shoulder for the melt gap MG to become a defect-free melt gap after the shoulder process is completed. The defect free melt gap refers to a melt gap such that the V / G of the solid-liquid interface is within the defect free margin. The defect free melt gap is not limited to a single value but may have a constant margin. That is, when the melt gap MG is in a certain range, the V / G of the solid-liquid interface is in a defect free margin.

Immediately after the shoulder process is completed, if the amount of change in the shoulder weight is +, the melt gap MG increases than the defect-free melt gap set as the process condition. On the contrary, if the shoulder weight change amount is-immediately after the shoulder process is completed, the melt gap MG is reduced than the defect-free melt gap set as the process condition.

The control means 140 calculates the melt gap change amount according to the shoulder weight change amount by using Equation 1 immediately after the shoulder process is completed.

[Equation 1]

△ Melt_gap = F (△ W _shoulder )

In Equation 1, ΔW _shoulder is a shoulder weight change amount based on a target weight, F is a function of converting a shoulder weight change amount into a melt gap change amount, and ΔMelt_gap is a melt gap change amount corresponding to a shoulder weight change amount. .

The function F determines a defect margin for the V / G, a defect melt gap, and a target weight of the shoulder in consideration of the diameter of the quartz crucible, the target diameter of the single crystal body, the design conditions of the hot zone, and the like. The melt gap change amount corresponding to the shoulder weight change amount of the condition may be calculated, and the functional correlation between the shoulder weight change amount and the melt gap change amount may be obtained by numerical analysis.

Equation 1 may be embodied as Equation 2 below, but the present invention is not limited thereto.

[Equation 2]

△ Melt_gap = A △ W _shoulder

In Equation 2, A is a proportional constant that converts the shoulder weight change amount into the melt gap change amount, and is dependent on the diameter of the quartz crucible. A may have a value of 0.6 ~ 1.2mm / kg.

The control means 140 corrects the melt gap based on the melt gap change amount calculated by Equation (1). That is, the control means 140 reduces the melt gap by the absolute value of the melt gap change amount when the melt gap change amount has a positive value. In addition, the control means 140 increases the melt gap by the absolute value of the melt gap change amount if the melt gap change amount has a-value.

Alternatively, the control means 140 corrects the melt gap MG when a condition in which the melt gap change exceeds a threshold is established. Since the defect-free melt gap may have a constant margin, the melt gap MG may be included in the margin of the defect-free melt gap unless the amount of change in the melt gap is excessive. Therefore, correcting the melt gap MG in the present invention means converging the melt gap MG to a certain value of the defect-free melt gap or changing it to a value within a margin of the defect-free melt gap having a specific range.

The correction of the melt gap MG is made by adjusting the gap between the quartz crucible 10 and the heat shield means 80. The adjustment of the interval is possible by adjusting the height of the quartz crucible 10 by driving the quartz crucible rotating means 30 up and down. Alternatively, the melt gap MG may be corrected by adjusting the position of the heat shield means 80 up and down.

Preferably, the control means 140 has a memory (not shown) that stores a function F for converting the melt gap change amount according to the shoulder weight change amount. The memory may be provided inside the control means or may be provided outside the control means. The memory is an inactive memory that is commonly used.

As described above, the control means 140 calculates the melt gap change amount according to the change amount of the shoulder weight by using Equation 1 after the shoulder process is completed, and corrects the melt gap based on the calculated melt gap change amount In the initial stage of the body process, the V / G at the solid-liquid interface (CM) can be prevented from escaping the defect margin, thereby preventing the occurrence of V defects or I defects in the single crystal formed at the beginning of the body process.

The control means 140 proceeds to the body process to grow a single crystal to a desired length when the shoulder process is completed. In Fig. 1, the single crystal body is marked B. The control means 140 receives the diameter data from the diameter sensor 110 during the body process to monitor the body diameter change amount. Receiving the diameter data takes place at a pre-scheduled time point. The body diameter change amount is a change amount based on a target diameter. The target diameter is the diameter of the single crystal that is the target of control in the course of the body process. For example, the target diameter for a 300 mm wafer is set to 304 to 310 mm. The reason why the target diameter is larger than the diameter of the wafer is that the outer circumferential surface of the single crystal is ground during single wafer processing.

As the body process proceeds, the silicon melt SM is consumed to increase the melt gap MG. Accordingly, the control unit 140 gradually raises the quartz crucible 10 in accordance with the consumption of the silicon melt SM to maintain the melt gap MG at the defect-free melt gap set in the process conditions as the body process proceeds. However, when the diameter of the single crystal body B deviates from the target diameter due to instability of the single crystal growth process or the like, the consumption amount of the silicon melt SM is changed to change the melt gap MG by the amount of change in the body diameter. As a result, if the melt gap MG becomes larger or smaller than the defect-free melt gap and leaves the margin, the V / G at the solid-state interface CM may escape the defect margin even if the pulling speed is maintained at the pulling rate. .

Therefore, the control means 140 monitors the diameter data of the single crystal received from the diameter sensor 110 during the body process. This is because monitoring the diameter data of single crystals can indirectly identify changes in the consumption of silicon melt (SM). If the change in the body diameter becomes positive, the consumption of the silicon melt SM increases, and the melt gap MG increases more than the defect free melt gap set as the process condition. When the change in the body diameter becomes-, the consumption of the silicon melt SM decreases more than expected, so that the melt gap MG decreases from the defect-free melt gap set as the process condition. This increase and decrease of the melt gap affects the temperature gradient G at the solid-liquid interface (CM), so that the single crystal pulling rate is intact, but the V / G can be out of the defect margin even if the pulling rate is maintained.

Therefore, the control means 140 corrects the melt gap MG using Equation 3 below when a change in the body diameter occurs during the process of the body process.

&Quot; (3) "

ΔMelt_gap = G (ΔD)

In Equation 2, ΔD is the body diameter change amount based on the target diameter of the body, G is a function of converting the body diameter change amount to the melt gap change amount, △ Melt_gap is the melt gap change amount corresponding to the body diameter change amount to be.

The function G determines a defect margin for the V / G, a defect melt gap, and a target diameter of the single crystal body in consideration of the diameter of the quartz crucible, the design conditions of the hot zone, the amount of polycrystalline silicon used, and the like. Melt gap variation corresponding to body diameter variation under various conditions can be calculated, and the functional correlation between body diameter variation and melt gap variation can be obtained by numerical analysis.

For example, Equation 3 may be embodied as Equation 4 below, but the present invention is not limited thereto.

&Quot; (4) "

△ Melt_gap = B △ D

In Equation 4, B is a proportional constant that converts the change in the body diameter into the melt gap change, and depends on the diameter of the quartz crucible and the length of the single crystal body. B may have a value of 1 to 2.

The control means 140 corrects the melt gap MG when the melt gap change amount corresponding to the body diameter change amount is calculated. If the melt gap change amount has a positive value, it means that the melt gap MG is increased more than the defect-free melt gap set as the process condition, thereby reducing the melt gap MG by the absolute value of the melt gap change amount. If the melt gap change amount has a negative value, it means that the melt gap MG is reduced from the defect-free melt gap set as the process condition, thereby increasing the melt gap MG by the absolute value of the melt gap change amount.

Alternatively, the control means 140 corrects the melt gap MG when a condition in which the melt gap change exceeds a threshold is established. Since the defect-free melt gap may have a constant margin, the melt gap MG may be included in the margin of the defect-free melt gap if the amount of change in the melt gap is not excessive.

The correction of the melt gap MG is made by adjusting the gap between the quartz crucible 10 and the heat shield means 80. The adjustment of the interval is possible by adjusting the height of the quartz crucible 10 by driving the quartz crucible rotating means 30 up and down. Alternatively, the melt gap MG may be corrected by adjusting the position of the heat shield means 80 up and down.

As described above, when the control means 140 corrects the melt gap MG according to the change in the body diameter during the process of the body process, the overall body process is prevented by preventing the V / G from leaving the defect margin at the solid-liquid interface CM. It is possible to prevent the occurrence of V defects or I defects in the single crystal. As a result, a large diameter single crystal of 300 mm or more can be grown into a high quality defect free single crystal.

2 is a flowchart illustrating a method of manufacturing a silicon single crystal according to the present invention.

1 and 2, in step S10, the control means 140 loads the target weight of the single crystal shoulder and the target diameter of the single crystal body set in the memory.

In step S20, the control means 140 controls the single crystal raising means 70 to soak the seed crystal 60 in the silicon melt SM and then proceed with the shoulder process. In the shoulder process, the pulling speed of the seed crystal is controlled to increase the diameter of the single crystal to the target diameter. Although not shown in the drawing, the control means 140 may proceed with the necking process prior to the shoulder process to remove the potential present in the seed crystal 60.

In step S30, the control means 140 obtains the weight data of the single crystal shoulder S by using the weight sensor 100 when the shoulder process is completed.

In step S40, the control means 140 calculates the shoulder weight change amount by comparing the shoulder weight with the target weight.

In step S50, the control means 140 calculates a melt gap change amount corresponding to the shoulder weight change amount using Equation (1).

In step S60, the control means 140 adjusts the distance between the quartz crucible 10 and the heat shield means 80 in response to the melt gap change amount to turn the melt gap MG into a defect-free melt gap of a certain value. Converge or adjust to a value within a certain range of flawless melt gap margins.

Alternatively, the control means 140 may correct the melt gap under conditions in which the melt gap change amount is larger than the threshold value.

In step S70, the control means 140 starts the body process and initializes the body diameter measurement time k to 1 when the correction of the melt gap MG according to the shoulder weight change amount is completed. When the body process is started, the control means 140 raises the single crystal by the defect free pulling speed set in the memory, and in order to keep the melt gap MG at the beginning of the body process constant as the single crystal body B is grown. The quartz crucible rotating means 30 is controlled to gradually raise the quartz crucible 10.

In step S80, the control means 140 determines whether the measuring period of the body diameter has arrived. If the measuring cycle of the body diameter arrives, the control means performs step S90.

In step S90, the control means 140 measures the diameter of the single crystal body B using the diameter sensor 110. Then, in step S100, the control means 140 calculates the body diameter change amount compared to the measured body diameter and the target diameter.

In step S110, the control means 140 calculates a melt gap change amount corresponding to the body diameter change amount by using Equation 3.

In step S120, the control means 140 converges the melt gap to a certain value of the defect-free melt gap or adjusts the melt gap MG to a value within a specific range of defect-free melt gap margins according to the melt gap variation. Correct.

Alternatively, the control means 140 may correct the melt gap under conditions in which the melt gap change amount is larger than the threshold value.

In step S130, the control means 140 determines whether the next measurement period of the body diameter set in the memory has arrived. The control means 140 performs step 140 if the next measurement period of the body diameter has arrived.

In step S140, the control means 140 increases the body diameter measurement turn k by one. In step S150, the control means 140 determines whether the body process is finished. If the body process is not finished, the control means 140 returns the process to step S90 to perform subsequent steps including step S90. On the other hand, if the body process is finished, the control means 140 ends the melt gap correction process.

Steps S90 to S150 are periodically repeated during the body process, so that even if the body diameter is out of the target diameter and the melt gap is changed, the melt gap is corrected so that the V / It is possible to prevent G from falling out of a flawless margin. Therefore, even when the large diameter silicon single crystal of 300 mm or more, which consumes a large amount of silicon melt, is grown, it is possible to produce high quality silicon single crystal without V defect or I defect.

[Experimental Example]

1. Melt Gap Change and Defect Type Identification Experiment

In this experiment, when the shoulder weight is out of the target weight, the V gap or I defect may occur in the single crystal due to the change of the melt gap, and in order to prevent such defects, it is desirable to correct the melt gap according to the change in the shoulder weight. Shows that Table 1 below summarizes the results of Experiment 1.

[Table 1]

Shoulder weight (kg) Melt Gap Variation (mm) Occurrence Defect Type % Defective Example 1 6.3 - - - Comparative Example 1 8.4 + 1.7 V defect 18.3 Comparative Example 2 4.9 1.1 I fault 13.1

Example 1

In Example 1, the shoulder was grown to a weight of 6.3 kg with the diameter of the single crystal set to 308 mm. As a result, the melt gap increased by 5 mm based on the melt gap before proceeding with the shoulder process. The 5 mm increased melt gap corresponds to a defect free melt gap for growing defect free single crystals over the entire area of the single crystal. Hereinafter, the melt gap with an increase of 5 mm is referred to as a defect-free melt gap. After the shoulder process was completed, the single crystal body was grown to a length of 1100 mm while calculating the defect free pulling speed margin based on the defect free melt gap and controlling the single crystal pulling speed within the calculated margin. When growing the single crystal body, the quartz crucible was slowly raised as the single crystal body was grown in order to maintain the melt gap as a defectless melt gap. As a result of processing the grown single crystal into a wafer, a high quality wafer without V defects or I defects was obtained.

[Comparative Example 1]

Compared to Example 1, the shoulder process was carried out longer, and the weight of the shoulder was increased by 2.1 kg to grow the shoulder to a weight of 8.4 kg. As a result, the melt gap increased by 6.7 mm based on the melt gap before proceeding with the shoulder process. This is a 1.7 mm increase over the flawless melt gap. In the body process, the single crystal body was grown to a length of 1100 mm while adjusting the pulling speed within the pulling speed margin calculated in Example 1. In addition, the growth rate of the quartz crucible was controlled during growth of the single crystal body to maintain the melt gap 1.7 mm larger than the defect free melt gap. As a result of processing the wafer with the grown single crystal, the percentage of wafers with V defects was found to be 18.3%.

Comparative Example 2

Compared to Example 1, the shoulder process was rapidly performed to reduce the weight of the shoulder by 1.4 kg to grow the shoulder to a weight of 4.9 kg. As a result, the melt gap increased by 3.9 mm based on the melt gap before the shoulder process. This value is 1.1 mm smaller than the defect free melt gap of Example 1. In the body process, the single crystal body was grown to a length of 1100 mm while controlling the single crystal pulling speed within the pulling speed margin calculated in Example 1. In addition, the growth rate of the quartz crucible was controlled during growth of the single crystal body to keep the melt gap 1.1 mm smaller than the defect free melt gap. As a result of processing the grown single crystal into a wafer, it was found that the ratio of the wafer having I defect was 13.1%.

From the experimental results shown in Table 1 it can be confirmed the following facts. If the shoulder weight is not controlled to the target weight, the melt gap at the completion of the shoulder process does not meet the melt gap condition. Therefore, if the body process is performed without correcting the melt gap change caused by the deviation of the target value of the shoulder weight, V defects or I defects are generated in the single crystal. V defects occur when the melt gap at the completion of the shoulder process is greater than the faultless melt gap. The reason for this is that even if the single crystal is pulled up at the rate of zero defect increase due to the decrease of G due to the increase of the melt gap, the bacon is excessively introduced into the single crystal. I defects also occur when the melt gap at the completion of the shoulder process is less than the faultless melt gap. The reason for this is that the lattice silicon is excessively introduced into the single crystal even though the single crystal is pulled up at a defect free pulling speed by increasing G due to the decrease of the melt gap. Therefore, if the melt gap is changed when the shoulder process is completed, if the melt gap is corrected according to the present invention and the body process is performed, V defects or I defects may be prevented from occurring in the single crystal.

2. Melt Gap Change and Defect Type Identification Experiments

The diameter change of the single crystal when growing the single crystal body also changes the amount of silicon melt that solidifies as well as the change in shoulder weight. Therefore, the change in diameter of the single crystal also causes a change in the melt gap. When the melt gap is changed, the temperature gradient G at the solid-liquid interface is changed, which may cause V defects or I defects in the single crystal.

Table 2 below shows the experimental results for the melt gap variation and the defect generation type under various body diameter conditions. In Table 2 below, the melt gap change amount is a relative change amount of the melt gap measured based on the melt gap of Example 2 when the body process is completed. The defect incidence rate represents the percentage of wafers where defects are found when wafer processing is performed with the single crystal body obtained in each experiment.

TABLE 2

Average diameter
(mm) Melt-Gap Variation
(mm) Occurrence Defect Type Defect rate
(%) Example 2 308 - - - Comparative Example 3 310.3 + 3.9 V defect 23.5 Comparative Example 4 309.8 + 3.1 V defect 20.1 Comparative Example 5 309.1 + 1.9 V defect 12.6 Comparative Example 6 308.4 + 0.7 - 0 Comparative Example 7 307.1 -1.5 I fault 3.4 Comparative Example 8 306.7 2.2 I fault 13.7 Comparative Example 9 306.5 2.6 I fault 15.8

Example 2

In Example 2, the body was grown to a length of 1100 mm by setting the target diameter of the body to 308 mm. During the body process, the rate of rise of the quartz crucible was controlled to keep the melt gap constant as a defectless melt gap. The pulling speed was also controlled within the flawless pulling speed range. As a result, a high-quality wafer without V defects or I defects in a single crystal was obtained.

[Comparative Examples 3 to 9]

In Comparative Examples 3 to 6, the target diameter of the body was set larger than 308 mm. The single crystal pulling speed and the melt gap were controlled under the same conditions as in Example 2. As a result, in Comparative Examples 3 to 6, the amount of change in the melt gap was greater than that of Example 2, showing a positive value. The size of the melt gap change increased as the diameter of the body was larger than 308 mm. As a result, wafers with V defects were found in the wafers produced by Comparative Examples 3 to 5, and the proportion of defective wafers increased as the target diameter of the body was larger than 308 mm. This is because the melt gap is intact during the process of the body process, and the G decreases because the melt gap is exceeded. As a result, the bake is excessively introduced into the single crystal even though the pulling speed is controlled in the margin.

On the other hand, in the comparative example 5, although the target diameter was set larger than 308 mm, the wafer with a V defect was not found. This is because the decrease in G is considerably smaller because the melt gap increase was not as large as in Example 2, and as a result, even if G was reduced, the V / G of the solid-liquid interface did not escape the margin. This means that even if G is reduced, the vacancy is not excessively introduced to cause V defects. In other words, even if the melt gap is changed due to the diameter change of the single crystal body, if the change width is within a certain value, it means that the defects do not occur in the single crystal by not leaving the margin without defects of the V / G of the solid-liquid interface.

In Comparative Examples 7 to 9, the target diameter of the body was set smaller than 308 mm. The single crystal pulling speed and the melt gap were controlled under the same conditions as in Example 2. As a result, in Comparative Examples 7 to 9, the consumption amount of the silicon melt was smaller than that of Example 2, and the change amount of the melt gap was-. The magnitude of the melt gap change increased as the target diameter of the body was smaller than 308 mm. As a result, the wafer with I defect was found in the wafer manufactured by Comparative Examples 7-9, and the ratio of the defective wafer increased as the target diameter of the body was smaller than 308 mm. This is because the melt gap decreases below the melt gap during the process of the body process, and the G increases. As a result, the lattice silicon is excessively introduced into the single crystal even though the pulling speed is controlled within the margin.

From the above experimental results, the following facts can be confirmed. When carrying out the body process, it is preferable to precisely control the diameter of the single crystal to the target diameter. If the diameter of the single crystal body is out of the target value during the process of the body process, the G is changed due to the change of the melt gap, and the V / G of the solid-liquid interface can be released. It is preferable. This melt gap correction enables the production of high quality defect-free silicon single crystals without V defects or I defects even when growing large diameter silicon single crystals of 300 mm or more, which consume a large amount of silicon melt.

As described above, although the present invention has been described by way of limited embodiments and drawings, the present invention is not limited thereto and is intended by those skilled in the art to which the present invention pertains. Of course, various modifications and variations are possible within the scope of the claims to be described.

The following drawings, which are attached to this specification, illustrate exemplary embodiments of the present invention, and together with the detailed description of the present invention serve to further understand the technical spirit of the present invention, the present invention includes matters described in such drawings. It should not be construed as limited to.

1 is a block diagram of a silicon single crystal manufacturing apparatus according to an embodiment of the present invention.

2 is a flowchart of a method of manufacturing silicon single crystal according to an embodiment of the present invention.

<Main reference number in drawing>

10: quartz crucible 20: crucible housing

30: quartz crucible rotating means 40: heating means

50: heat insulation means 60: seed crystal

70: single crystal pulling means 80: heat shield means

100: weight sensor 110: diameter sensor

130: drive means 140: control means

Claims (15)

In the silicon single crystal manufacturing method using the Czochralski method in which seed crystals are immersed in silicon melt and the seed crystals are gradually raised to grow single crystals. Measuring the diameter of the single crystal body while pulling up the body of silicon single crystal; Calculating a body diameter change amount by comparing the measured body diameter with a target diameter; Calculating a melt gap variation using the calculated body diameter variation; and Compensating the melt gap in the direction to cancel the calculated melt gap change; Silicon single crystal manufacturing method comprising the V / G of the solid-liquid interface to maintain within the defect margin. delete delete delete delete delete The method of claim 1, wherein the melt gap correction step, And if the melt gap change amount is greater than a threshold, correcting the melt gap in a direction to cancel the melt gap change amount. The method of claim 1, wherein the correction of the melt gap,  A method for producing a silicon single crystal, characterized in that the melt gap is converged into a defect-free melt gap having a specific value or the melt gap is changed to a value within a margin of the defect-free melt gap having a specific range. Quartz crucibles containing silicon melt; Quartz crucible rotating means for moving the quartz crucible up and down while rotating the quartz crucible; Single crystal pulling means for pulling a single crystal from the silicon melt using a seed crystal wire; Heat shield means for shielding heat emitted from an outer circumferential surface of the single crystal being pulled up and forming a melt gap with a surface of the silicon melt; A diameter sensor measuring the diameter of the single crystal to be pulled and outputting diameter data; And Receives diameter data from the diameter sensor during growth of the single crystal body, calculates the body diameter change compared to the received diameter and the target diameter, calculates the melt gap change according to the consumption change of the silicon melt according to the diameter change, and calculates And control means for correcting the melt gap change amount in a direction to cancel the melt gap change amount according to the melt gap change amount. 10. The method of claim 9, And the control means corrects the melt gap by adjusting the relative positions of the quartz crucible and the heat shield means. delete 10. The method of claim 9, And the control means corrects the melt gap by converging the melt gap to a defect-free melt gap of a certain value or by changing the melt gap to a value within a defect-free melt gap margin of a specific range. delete delete delete

Images