JP2020001977A - Determination method for oxygen stripe flattening manufacture condition of silicon single crystal, manufacturing method for silicon single crystal using the same, and silicon single crystal wafer - Google Patents

Abstract

To provide a determination method for a manufacture condition that can flatten oxygen stripes of a silicon single crystal more than before in manufacture of the silicon single crystal by an MCZ method, and a manufacturing method for the silicon single crystal using the same.SOLUTION: The present invention relates to a determination method for an oxygen stripe flattening manufacturing condition of a silicon single crystal in manufacture of the silicon single crystal by a magnetic field application Czochralski process. The determination method for the oxygen stripe flattening manufacture condition comprises: calculating a ratio Δt/M, where Δt is a temperature difference between a maximum temperature and a minimum temperature of silicon molten liquid and M is magnetic field intensity; finding correlation between the ratio Δt/M and oxygen stripes in the silicon single crystal raised at the ratio Δt/M; and setting a condition of the ratio Δt/M at which a desired silicon single crystal is obtained based upon the correlation; and determining the oxygen stripe flattening manufacture condition of the silicon single crystal.SELECTED DRAWING: Figure 4

Description

  The present invention relates to a silicon single crystal grown by a magnetic field application Czochralski method (MCZ method) and containing oxygen in the crystal, and a method for determining manufacturing conditions under which the crystal can be grown and using the method. The present invention relates to a method for manufacturing a silicon single crystal.

  2. Description of the Related Art A silicon single crystal rod for cutting a silicon single crystal wafer used as a substrate of a semiconductor device such as a memory, a CPU, and an image sensor is mainly manufactured by a Czochralski (CZ) method. In the CZ method, since a quartz crucible is generally used, oxygen atoms elute from the crucible and are taken into a silicon single crystal. Therefore, oxygen atoms are contained in the CZ crystal (silicon single crystal manufactured by the CZ method). ing.

  A silicon single crystal wafer cut from a silicon single crystal rod passes through many stages of device processes, and through the heat treatment process at that time, silicon atoms and oxygen atoms combine to form oxygen precipitates in the wafer. The oxygen precipitate formed in the bulk is called BMD (Bulk Micro Defect) or the like. It is known that this BMD has the ability to improve the device characteristics by capturing contaminant atoms such as heavy metals inside the wafer. For some devices such as an imaging device that is greatly affected by heavy metal contamination, a device with higher performance and higher reliability can be obtained as the BMD density increases. On the other hand, if there are too many BMDs, electrical characteristics such as leakage current may be degraded due to them.

  Therefore, for example, in an image sensor, when the BMD is small, deterioration caused by heavy metal is caused, and when the BMD is large, image unevenness is caused by deterioration caused by the BMD itself. It is known that the occurrence of the BMD in the wafer surface is caused by the unevenness of the oxygen concentration taken into the original crystal. This unevenness of the oxygen concentration is called an oxygen fringe. I have.

  Therefore, in order to obtain a good image sensor, a technique for precisely suppressing the periodic fluctuation of the BMD density in the wafer surface on the order of millimeters, that is, oxygen stripes in the crystal is desired. Such technology contributes to the improvement of the characteristics of not only the image sensor but also the materials for memories, CPUs, and solar cells. Therefore, the technology has an extremely wide range of applications, and improves the electrical characteristics and the occurrence of warpage or slip dislocation during the process. There are also effects such as preventing.

  By the way, in the production of a silicon single crystal by the CZ method, a silicon melt in a quartz crucible flows by convection, oxygen in the quartz crucible elutes, and is taken into the silicon single crystal at a high concentration. Therefore, an MCZ method for suppressing convection by applying a magnetic field to a silicon melt in a quartz crucible has been performed. As a result, it is known that the oxygen concentration in the crystal is suppressed, and a crystal having a large diameter is easily grown.

  It is disclosed that oxygen stripes are reduced in the MCZ method. For example, Patent Literature 1 discloses a technique in which the convection speed of a silicon melt is set to 7 mm / sec or less to eliminate growth fringes and control the oxygen concentration. If convection could be completely suppressed, oxygen streaks might disappear. However, it is currently impossible to completely suppress convection at present, and for example, microscopically, the micro oxygen concentration fluctuates drastically as disclosed in Patent Document 2. In Patent Document 2, by controlling the coil center position, it is possible to suppress microscopic unevenness in the oxygen concentration distribution. However, in some cases, it may not be sufficient when viewed with a crystal evaluation technique suitable for an image sensor in recent years as disclosed in Patent Document 3, for example.

  Therefore, a technique for improving the oxygen concentration in-plane distribution and oxygen stripes has been disclosed since then. Patent Document 4 discloses a manufacturing method in which either a high-temperature portion or a low-temperature portion on the surface of a silicon melt is always positioned at a solid-liquid interface. However, this method requires a temperature measurement technique during operation and its control, but does not mention a specific control method. Patent Literature 5 discloses a method for producing a temperature distribution that is not rotationally symmetric in a solidification front region in a melt. As a specific method, a device such as a movable magnetic field or a shielding unit that breaks the rotational symmetry of the magnetic field is required, and is not a technology that can be immediately applied to existing devices. Further, Patent Document 6 discloses a manufacturing method in which a silicon melt flows from one side to the other under a solid-liquid interface. This patent document describes specific parameters, and may be applicable to existing machines. However, it is not easy to grow a crystal while satisfying the condition while the liquid level height falls under the condition and the amount of the melt decreases along with the crystal growth as in the CZ method. And the degree of freedom of operation is small. Furthermore, the crystal rotation that greatly affects the macroscopic oxygen plane distribution is included in the constraints, and even if the microscopic oxygen distribution is improved, it may adversely affect the macroscopic distribution. is there.

JP-A-6-92774 JP-A-9-188590 JP 2014-148448 A JP 2000-264784 A JP 2004-196655 A WO 2017/077701

  As described above, in the production of a silicon single crystal by the MCZ method, it is desired to further reduce oxygen fringes in the silicon single crystal. However, in the conventional method, a crystal evaluation technique suitable for a recent image sensor is desired. There was a problem that it could not be said that it was enough when viewed from the above.

SUMMARY OF THE INVENTION The present invention has been made in view of the above problems, and in the production of a silicon single crystal by the MCZ method, a method for determining production conditions that can flatten oxygen fringes of a silicon single crystal as compared with the conventional method, and a silicon single crystal using the same. An object of the present invention is to provide a method for producing a crystal.
Another object of the present invention is to provide a silicon single crystal wafer suitable for a device, which has few defects such as image unevenness even when an imaging element is manufactured.

In order to achieve the above object, the present invention provides
A method for determining oxygen fringe flattening production conditions of a silicon single crystal in the production of a silicon single crystal by a magnetic field application Czochralski method,
Δt is the temperature difference between the highest temperature and the lowest temperature of the silicon melt.
Assuming that the magnetic field strength is M,
Calculating a ratio Δt / M between the Δt and the M,
The correlation between the ratio Δt / M and oxygen fringes in the silicon single crystal grown at the ratio Δt / M is determined, and based on the correlation, the condition of the ratio Δt / M at which a desired silicon single crystal is obtained is determined. The present invention provides a method for determining oxygen fringe flattening manufacturing conditions, wherein the setting is made and oxygen fringe flattening manufacturing conditions of silicon single crystal are determined.

  Such a method for determining oxygen fringe flattening production conditions according to the present invention is directed to a method for manufacturing a silicon single crystal by the MCZ method, by using the above-mentioned oxygen fringe flattening production conditions to reduce a silicon single crystal with reduced oxygen fringes. Can be determined.

  At this time, the calculation of the ratio Δt / M is performed by calculating the temperature distribution of the quartz crucible in the initial melt amount by simulation in a pulling machine configuration used for manufacturing a silicon single crystal, and calculating the calculated temperature distribution of the quartz crucible. The highest temperature of the silicon melt, the temperature at the solid-liquid interface as the lowest temperature of the silicon melt, the difference is Δt, and the central magnetic field strength of the silicon melt is M. Is preferred.

  In the structure of a pulling machine used for manufacturing a silicon single crystal (combination of a pulling machine (silicon single crystal growing apparatus) and a so-called HZ (Hot Zone) part such as a graphite part and a heat insulating material provided therein), By calculating Δt / M in this manner, the production conditions for flattening oxygen stripes of a silicon single crystal can be determined with high accuracy.

  Further, a silicon single crystal wafer sample obtained by cutting out oxygen fringes in the silicon single crystal grown at the above-mentioned ratio Δt / M from the silicon single crystal grown at the above-mentioned ratio Δt / M by 2 mm pitch using the FT-IR method. The oxygen concentration gradient is obtained by measuring the oxygen concentration at 6 mm intervals of the measured oxygen concentration, and the maximum in-plane value of the determined oxygen concentration gradient is 0.04 (ppma-JEIDA / mm). It is preferable to set the condition of the ratio Δt / M so as to be as follows.

  By setting the condition of the ratio Δt / M in this way, it is possible to determine the oxygen fringe flattening production conditions that can more reliably produce a silicon single crystal with reduced oxygen fringes.

  Further, the ratio Δt / M is preferably set to 0.05 (K / gauss) or less.

  With such a ratio Δt / M, it is possible to determine the oxygen fringe flattening manufacturing conditions that can more reliably manufacture a silicon single crystal with reduced oxygen fringes.

  Further, the present invention determines the condition of the ratio Δt / M in advance by using the above-described method for determining the oxygen fringe flattening operation condition, and based on the determined condition of the ratio Δt / M, applies a magnetic field application Czochralski. A method for producing a silicon single crystal, which comprises growing a silicon single crystal by a method.

  According to such a method of manufacturing a silicon single crystal, a silicon single crystal with reduced oxygen fringes can be manufactured by using a condition of a predetermined ratio Δt / M.

  Further, the present invention relates to a silicon single crystal wafer containing oxygen, wherein the in-plane maximum value of the oxygen concentration gradient value at 6 mm intervals of the silicon single crystal wafer is 0.04 (ppma-JEIDA / mm) or less. The present invention provides a silicon single crystal wafer characterized by the following.

  With such a silicon single crystal wafer, the in-plane maximum value of the oxygen concentration gradient value is as small as 0.04 (ppma-JEIDA / mm) or less, and the unevenness of the oxygen concentration is sufficiently suppressed. Even if it is manufactured, there are few defects such as image unevenness, and it is suitable for a device.

As described above, the method for determining the oxygen fringe flattening manufacturing condition of the present invention determines the correlation between the ratio Δt / M and the oxygen fringe in the silicon single crystal grown at the ratio Δt / M, and determines the correlation. The conditions for the ratio Δt / M at which a desired silicon single crystal is obtained are set based on this, and the oxygen fringe flattening production conditions for the silicon single crystal are determined. Thereby, when manufacturing the silicon single crystal by the MCZ method, the above-described oxygen fringe flattening manufacturing condition is used to determine the oxygen fringe flattening manufacturing condition that enables the manufacturing of the silicon single crystal with reduced oxygen fringe. be able to.
In addition, in the case of the silicon single crystal wafer of the present invention, since the in-plane maximum value of the oxygen concentration gradient value is 0.04 (ppma-JEIDA / mm) or less, even if an imaging device is manufactured, defects such as image unevenness will occur. Less and suitable for the device.

It is a figure which shows the oxygen concentration in-plane distribution measured by 2 mm pitch by FT-IR method in Experimental example 1 and Experimental example 2. FIG. 9 is a diagram showing the in-plane distribution of oxygen concentration gradient values at 6 mm intervals obtained from the in-plane distribution of oxygen concentration in Experimental Examples 1 and 2. It is the schematic of the general silicon single crystal growing apparatus. FIG. 9 is a diagram in which the oxygen fringe index (the maximum value of the oxygen concentration gradient) obtained in the experimental example is plotted with respect to the ratio Δt / M.

  Hereinafter, the present invention will be described in detail with reference to the drawings as an example of an embodiment, but the present invention is not limited to this.

  As described above, in the production of a silicon single crystal by the MCZ method, it is desired to further reduce oxygen fringes in the silicon single crystal. However, in the conventional method, a recent imaging device, memory, CPU, It has been difficult to produce silicon single crystals that can be said to be sufficient for improving the properties of materials for batteries.

  The present inventor has conducted intensive studies on such a problem, and previously determined the oxygen fringe flattening manufacturing conditions from the temperature difference between the highest temperature and the lowest temperature of the silicon melt and the magnetic field intensity, and determined the oxygen fringe flattening condition. It has been conceived that if a silicon single crystal is manufactured using the chemical manufacturing conditions, a silicon single crystal with reduced oxygen fringes can be manufactured, and the present invention has been completed.

That is, the present invention is a method for determining the production condition of oxygen fringe flattening of silicon single crystal in the production of silicon single crystal by magnetic field application Czochralski method,
Δt is the temperature difference between the highest temperature and the lowest temperature of the silicon melt.
When the magnetic field strength is M,
Calculating a ratio Δt / M between the Δt and the M,
The correlation between the ratio Δt / M and oxygen fringes in the silicon single crystal grown at the ratio Δt / M is determined, and based on the correlation, the condition of the ratio Δt / M at which a desired silicon single crystal is obtained is determined. This is a method for determining oxygen fringe flattening manufacturing conditions, wherein the conditions are set and oxygen fringe flattening manufacturing conditions for silicon single crystal are determined.

  Such a method for determining the oxygen fringe flattening manufacturing condition of the present invention obtains a correlation between the ratio Δt / M and oxygen fringes in the silicon single crystal grown at the ratio Δt / M, and based on the correlation. The conditions for the ratio Δt / M to obtain a desired silicon single crystal are set, and the production conditions for flattening oxygen stripes of the silicon single crystal are determined. Thereby, when manufacturing the silicon single crystal by the MCZ method, the above-described oxygen fringe flattening manufacturing condition is used to determine the oxygen fringe flattening manufacturing condition that enables the manufacturing of the silicon single crystal with reduced oxygen fringe. be able to.

  As described in Patent Document 1, in order to suppress oxygen stripes, it is preferable to suppress the convection velocity of a silicon melt (melt). Fluids flow due to non-uniformity mainly due to temperature. Therefore, the strength of this convection depends on the temperature difference. That is, it is considered that the smaller the difference Δt between the maximum temperature and the minimum temperature of the melt is, the more the convection is suppressed. On the other hand, in the MCZ method, convection is suppressed by applying a magnetic field. However, it is considered that convection can be suppressed as the magnetic field intensity M increases.

  Therefore, it is considered that the strength of convection can be estimated by taking the ratio Δt / M between these two parameters. That is, if a condition with a small ratio Δt / M is selected, an operation with low convection is possible, and thus, it is possible to determine a production condition in which oxygen stripes are suppressed.

  Here, there are only two operational constraints: the difference Δt between the maximum and minimum temperatures of the melt and the magnetic field strength M. Other than minute oxygen concentration distribution, such as crystal growth speed, crystal rotation speed, crucible rotation speed, furnace pressure in the pulling machine and the amount of inert gas flowing during crystal growth, for example, macroscopic oxygen distribution and crystal defects There are no restrictions on operating parameters that have a significant effect on the crystal quality of. Of course, it is not denied that these operation parameters also affect the minute oxygen concentration distribution, but since the temperature difference Δt and the magnetic field strength M that have a greater effect are defined, the minute difference can be obtained by using these operation parameters. It is basically possible to freely control the quality other than the optimum oxygen concentration distribution.

  At this time, the calculation of the ratio Δt / M is performed by calculating the temperature distribution of the quartz crucible in the initial melt amount by simulation in a pulling machine configuration used for manufacturing a silicon single crystal, and calculating the calculated temperature distribution of the quartz crucible. The highest temperature of the silicon melt, the temperature at the solid-liquid interface as the lowest temperature of the silicon melt, the difference is Δt, and the central magnetic field strength of the silicon melt is M. Is preferred.

  By calculating the ratio Δt / M in this way, it is possible to accurately determine the oxygen fringe flattening production conditions of the silicon single crystal.

  The strength of convection is expected to some extent at Δt / M, as indicated above. However, the strength of convection is not determined simply by this value, but also depends on the amount of melt. That is, if the amount of the melt is large, the convection becomes strong, and if the amount of the melt is small, the convection becomes weak. Therefore, it is preferable to calculate Δt / M in the state of the initial melt amount that is considered to have the strongest convection.

  Further, if it is possible to actually measure the temperature in order to obtain Δt, there is nothing beyond that, but in general, it is almost impossible to find the position of the highest temperature in the melt and measure the temperature. Therefore, it is realistic to obtain the value using a simulation.

  In the CZ method, a melt is generally placed in a quartz crucible placed in a graphite crucible. This is heated from outside the graphite crucible with a heater. Therefore, generally, the point indicating the highest temperature in the melt is the point in contact with the quartz crucible.

  Therefore, the maximum temperature of the quartz crucible obtained by simulation can be used as the maximum melt temperature. This is a quartz crucible with fewer elements for calculation than calculating the maximum temperature at the melt, which has many elements for calculation (that is, a volume for a three-dimensional simulation and a large cross-sectional area for a two-dimensional simulation). It is easier to find the maximum temperature at Of course, it is apparent that the maximum melt temperature may be properly determined and that temperature may be used.

  On the other hand, as the minimum temperature of the melt, the temperature at the solid-liquid interface can be used. Generally, a solid-liquid interface is a portion where a liquid and a solid are in contact, and has a melting point as a temperature, and is considered to be the lowest temperature in a melt. Therefore, the temperature at the solid-liquid interface, that is, the melting point can be used as the minimum melt temperature, and the difference from the maximum melt temperature (here, the maximum temperature of the quartz crucible obtained by simulation) can be used as Δt. Of course, it is clear that the lowest melt temperature may be obtained by a proper simulation and that temperature may be used.

  Further, the central magnetic field strength of the melt can be used as the magnetic field strength M. The magnetic field strength applied to the melt generally has a distribution. Therefore, in order to perform the calculation more accurately, a value obtained by integrating the magnetic field component applied to the melt and considering the direction of the convection with respect to the entire melt as an effective magnetic field intensity for suppressing the convection is used. It can be used. However, it is not easy to calculate them one by one. Therefore, here, the ratio of Δt / M is calculated assuming that the central magnetic field strength of the melt is the ability to suppress convection, and a condition in which this value is small can be obtained as a manufacturing condition for flattening oxygen stripes.

  Since the temperature difference Δt is basically determined by the structure of the lifting machine, it does not greatly change during operation. Of course, to be precise, Δt changes due to a decrease in the melt accompanying the crystal growth, but here, it is determined by simulation under the condition of the initial melt amount where the convection is strongest. It is unlikely that oxygen stripes will deteriorate. Thus, for example, it is also possible to reduce the magnetic field strength during operation.

  Further, a silicon single crystal wafer sample obtained by cutting out oxygen fringes in the silicon single crystal grown at the above-mentioned ratio Δt / M from the silicon single crystal grown at the above-mentioned ratio Δt / M by 2 mm pitch using the FT-IR method. The oxygen concentration gradient is obtained by measuring the oxygen concentration at 6 mm intervals of the measured oxygen concentration, and the maximum in-plane value of the determined oxygen concentration gradient is 0.04 (ppma-JEIDA / mm). It is preferable to set the condition of the ratio Δt / M so as to be as follows.

By setting the condition of the ratio Δt / M in this way, it is possible to determine the oxygen fringe flattening production conditions that can more reliably produce a silicon single crystal with reduced oxygen fringes.
Hereinafter, this will be described using experimental examples.

(Experimental example)
Using a silicon single crystal growing apparatus of a pulling machine configuration schematically shown in FIG. 3, using a quartz crucible having an inner diameter of about 780 mm, various silicon single crystals having a final product diameter of 300 mm were grown by the MCZ method. .

  The external appearance of the silicon single crystal growing apparatus 14 shown in FIG. 3 includes a main chamber 1, a top chamber 11, and a pulling chamber 2 communicating with the main chamber. A graphite crucible 6 and a quartz crucible 5 are provided inside the main chamber 1. A heater 7 is provided so as to surround the graphite crucible 6 and the quartz crucible 5, and the material silicon 4 contained in the quartz crucible 5 is melted by the heater 7 to form a raw material melt 4. Further, a heat insulating member 8 is provided to prevent the radiant heat from the heater 7 from affecting the main chamber 1 and the like.

  In growing a crystal by the MCZ method, a magnetic field is applied by a magnetic field application device (not shown).

  On the melt surface of the raw material melt 4, a shielding member 13 is arranged opposite to the melt surface at a predetermined interval, and shields radiant heat from the melt surface of the raw material melt 4. After the seed crystal is immersed in this crucible, the rod-shaped single crystal rod 3 is pulled up from the raw material melt (melt) 4 under the conditions determined by the method for determining the production conditions for oxygen fringe flattening of the present invention. The crucible can be moved up and down in the direction of the crystal growth axis, and the crucible is raised during growth so as to compensate for the decrease in the liquid level of the raw material melt 4 which has been reduced by the growth of the single crystal. The height of the melt surface is kept approximately constant.

  Further, an inert gas such as an argon gas is introduced as a purge gas from the gas inlet 10 during the growth of the single crystal, passes between the single crystal rod 3 being pulled up and the gas purge cylinder 12, and then the shielding member 13 and the raw material The liquid 4 passes between the surface of the melt and the gas 4 and is discharged from the gas outlet 9. By controlling the flow rate of the gas to be introduced and the amount of gas discharged by a pump or a valve (not shown), the pressure in the chamber during pulling is controlled.

  FIG. 3 is a schematic diagram of the silicon single crystal growing apparatus, but there are actually various types of HZ set in the pulling machine. Therefore, the temperature distribution in the initial melt amount of each puller configuration used for the production of the silicon single crystal was calculated by FEMAG (F. Dupret, P. Nicodeme, Y. Ryckmans, P. Waters, and M). J. Crochet, Int. J. Heat Mass Transfer, 33, 1849 (1990)).

  The maximum temperature of the quartz crucible obtained by the simulation was taken as the maximum temperature of the silicon melt, and the difference from the melting point of silicon, 1685 K (1412 ° C.), was found as Δt. As a result, Δt in the initial melt amount in the structure of the pulling machine used in this experimental example fluctuated in the range of 125 to 185 (K). The magnetic field strength M when growing a silicon single crystal was the central magnetic field strength of the silicon melt, and the central magnetic field strength was varied in the range of 2500 to 4000 gauss. Here, a silicon single crystal was manufactured without changing the central magnetic field strength during crystal growth. Using these results, the ratio Δt / M under the production conditions under which each crystal was grown was calculated.

These silicon single crystals were evaluated using the FT-IR method. The evaluation method is as follows. First, a silicon single crystal wafer sample having a thickness of about 1 mm cut out from a grown silicon single crystal was mirror-finished and then divided into quarters. Next, the silicon single crystal wafer sample was scanned at a pitch of 2 mm in the radial direction by the FT-IR method, and the interstitial oxygen concentration was measured using a peak at 1107 cm ⁻¹ caused by interstitial oxygen. At that time, the spatial resolution was set to 100 μm × 100 μm by a method called a so-called micro FT-IR method or μFT-IR method instead of a general FT-IR method having a spatial resolution of the order of mm, and the measurement variation of the oxygen concentration was measured. Was controlled to 0.01 ppma or less. This is because a method called a so-called micro FT-IR method or μFT-IR method has an order of magnitude smaller than a measurement method in which the spatial resolution, in which one measurement point is an average value in the mm order region, is mm order. This is because the oxygen concentration gradient can be obtained more accurately.

  At this time, the value of the sample edge portion of the silicon single crystal wafer sample may be incorrect due to the influence of the edge and the like. . An oxygen concentration gradient at 6 mm intervals was obtained from the oxygen concentration value obtained by this measurement method, and the maximum value in the plane was obtained as an oxygen fringe index.

  As described in Patent Document 3, it is important to suppress the oxygen concentration gradient of 4 to 10 mm in order to suppress adverse effects on devices such as an image sensor due to oxygen fringes. FIG. 1 shows the in-plane oxygen concentration distribution at a pitch of 2 mm measured by the FT-IR method using the evaluation method described in the experimental example. FIG. 2 shows the in-plane distribution of the oxygen concentration gradient at 6 mm intervals obtained from the measured values.

  In each of the puller configurations of the above experimental examples, among the silicon single crystals pulled up by shaking Δt and M, an example showing a small in-plane variation of oxygen concentration by the FT-IR method is shown in Experimental Examples 1 and 2. Another example showing a larger in-plane variation in oxygen concentration than in Example 1 is shown in FIGS.

  The silicon single crystal shown in Experimental Example 1 was a crystal in which the in-plane variation of the oxygen concentration was small. When an image sensor was manufactured using a silicon single crystal wafer cut from this silicon single crystal, a defect considered to be caused by oxygen fringes was obtained. Did not occur. On the other hand, the silicon single crystal shown in Experimental Example 2 is a crystal in which the in-plane variation of the oxygen concentration is large. When an image sensor was manufactured using a silicon single crystal wafer cut out from this crystal, a defect considered to be caused by oxygen fringes was obtained. There has occurred. However, even in the silicon single crystal of Experimental Example 2, the total amount was not defective, but was a partial defect. From the comparison between the defect distribution and the distribution shown in FIG. 2, a defect occurred at a large portion where the oxygen concentration gradient at intervals of 6 mm was approximately 0.05 (ppma-JEIDA / mm) or more, and 0.03 (ppma-JEIDA). / Mm) was analyzed so that no defect occurred in a small portion.

  Therefore, in a wafer in which the maximum value of the oxygen concentration gradient value (oxygen fringe index) of the grown crystal is 0.04 (ppma-JEIDA / mm) or less, the oxygen concentration gradient value is 0.03 (ppma-JEIDA). / Mm) or more, and the average value is sufficiently small, so it is considered that the wafer is suitable for manufacturing an imaging device.

  Conventionally, it was difficult to lower the oxygen concentration gradient value to 0.02 (ppma-JEIDA / mm) or less. However, according to the method of the present invention, as shown in Experimental Example 1 in FIG. An extremely good silicon single crystal wafer having an oxygen fringe index of 0.02 (ppma-JEIDA / mm) or less can be produced.

  Further, the ratio Δt / M is preferably set to 0.05 (K / gauss) or less.

  With such a ratio Δt / M, it is possible to determine the oxygen fringe flattening manufacturing conditions that can more reliably manufacture a silicon single crystal with reduced oxygen fringes.

  The oxygen fringe indices in the respective puller configurations obtained by varying Δt and M obtained by the method of the above experimental example were plotted against the ratio Δt / M. As shown in FIG. 4, the oxygen fringe index is calculated based on the ratio Δt / M between Δt calculated using the integrated heat transfer analysis software FEMAG used for the simulation and the central magnetic field intensity M as the magnetic field intensity M. When plotted, there was a good correlation between the two. From the correlation shown by the dotted line in FIG. 4, if the ratio Δt / M is about 0.05 (K / gauss) or less, a silicon single crystal having an oxygen concentration index of 0.04 (ppma-JEIDA / mm) or less can be obtained. It turned out to be obtained. Therefore, it is preferable to set the production conditions such that Δt / M is 0.05 (K / gauss) or less.

  The central magnetic field strength is preferably at least 500 gauss or more in order to sufficiently obtain the effect of suppressing convection. On the other hand, the central magnetic field strength is preferably 6000gauss or less. With such a magnetic field strength, when convection is suppressed too much and oxygen concentration non-uniformity is accumulated in the melt and cannot be suppressed by the applied magnetic field strength, it is opened at once and conversely The occurrence of a phenomenon in which the non-uniformity is enhanced can be prevented. It is desirable that Δt is at least 50K or more. With such Δt, no solidification occurs during crystal growth. Further, it is preferable that the maximum melt temperature, that is, the maximum crucible temperature be equal to or lower than the crucible softening point. Thereby, softening of the crucible can also be suppressed.

  However, the temperatures shown here are calculated values under the conditions used for calculation by the present inventors. Since these values greatly change depending on the calculation conditions, the values shown here are not absolute values, but are only reference values, and all the operating condition determination methods based on such a concept fall into the category of this technology. .

  Further, the present invention determines the condition of the ratio Δt / M in advance by using the above-described method for determining the oxygen fringe flattening operation condition, and based on the determined condition of the ratio Δt / M, applies a magnetic field application Czochralski. A method for producing a silicon single crystal, which comprises growing a silicon single crystal by a method.

  According to such a method of manufacturing a silicon single crystal, a silicon single crystal with reduced oxygen fringes can be manufactured by using a condition of a predetermined ratio Δt / M.

  As a manufacturing apparatus used in the method for manufacturing a silicon single crystal, for example, the above-described silicon single crystal growing apparatus 14 in FIG. 3 can be used.

  Further, the present invention relates to a silicon single crystal wafer containing oxygen, wherein the in-plane maximum value of the oxygen concentration gradient value at 6 mm intervals of the silicon single crystal wafer is 0.04 (ppma-JEIDA / mm) or less. The present invention provides a silicon single crystal wafer characterized by the following.

  If the oxygen fringe index is 0.04 (ppma-JEIDA / mm) or less, it can be said that the wafer is suitable for a device with few defects such as image unevenness even when an imaging device is manufactured. Wafers exceeding 0.04 (ppma-JEIDA / mm) cause defects such as image unevenness when an imaging device is manufactured, and can be said to be not suitable for devices.

  Hereinafter, the present invention will be described specifically with reference to Examples and Comparative Examples, but the present invention is not limited thereto.

(Example 1)
From the relationship shown in FIG. 4, a good crystal with flattened oxygen stripes was obtained. Using the HZ (hot zone) configuration in which the temperature difference Δt = 137.4 (K) in the initial melt amount calculated using FEMAG from the puller configuration used in the experimental example, the central magnetic field intensity was used. The crystal was grown with M being constant at 4000 (gauss). At this time, the ratio Δt / M is 0.034 (K / gauss), which is lower than 0.05 (K / gauss). Therefore, good oxygen stripe quality is expected.

  When the oxygen fringe index was determined from the silicon single crystal grown under these conditions by the same evaluation method as in the experimental example, values less than 0.022 (ppma-JEIDA / mm) and 0.04 (ppma-JEIDA / mm) were obtained. Was done. When the wafer cut out of the crystal was put into an image sensor process, good results with few defective products could be obtained.

(Comparative Example 1)
Using a silicon single crystal growing apparatus having a pulling machine configuration schematically shown in FIG. 3, a crystal having a final product diameter of 300 mm was grown using a quartz crucible having an inner diameter of about 780 mm. At this time, the crystal was grown without concern for the index of the ratio Δt / M. When confirmed later, the temperature difference Δt at the initial hot water amount calculated using FEMAG was 183.6 (K), and the central magnetic field intensity M was constant at 2500 (gauss). The ratio Δt / M calculated from these was 0.073 (K / gauss), which was greater than 0.05 (K / gauss).

  When the oxygen fringe index was determined from the crystal grown under these conditions by the same evaluation method as that of the experimental example, the values were 0.078 (ppma-JEIDA / mm) and more than 0.04 (ppma-JEIDA / mm). It was gone. When a silicon single crystal wafer cut out of this crystal was put into the process of an image sensor, good results could not be obtained because of many defective products.

  Note that the present invention is not limited to the above embodiment. The above embodiment is an exemplification, and the present invention has substantially the same configuration as the technical idea described in the claims of the present invention, and has the same effect. Within the technical scope of

1. Main chamber, 2. Pulling chamber, 3. Single crystal rod,
4: raw material melt (melt), 5: quartz crucible, 6: graphite crucible,
7 ... heater, 8 ... heat insulating member, 9 ... gas outlet, 10 ... gas inlet,
11: top chamber, 12: gas purge cylinder, 13: shielding member,
14 ... Silicon single crystal growing device (pulling machine).

Claims (6)

A method for determining oxygen fringe flattening production conditions of a silicon single crystal in the production of a silicon single crystal by a magnetic field application Czochralski method,
Δt is the temperature difference between the highest temperature and the lowest temperature of the silicon melt.
When the magnetic field strength is M,
Calculating a ratio Δt / M between the Δt and the M,
The correlation between the ratio Δt / M and oxygen fringes in the silicon single crystal grown at the ratio Δt / M is determined, and based on the correlation, the condition of the ratio Δt / M at which a desired silicon single crystal is obtained is determined. A method for determining oxygen fringe flattening manufacturing conditions, comprising setting and setting oxygen fringe flattening manufacturing conditions of a silicon single crystal.   The calculation of the ratio Δt / M is performed in a pulling machine configuration used for the production of a silicon single crystal, and the temperature distribution of the quartz crucible in the initial melt amount is obtained by simulation, and the highest temperature of the obtained temperature distribution of the quartz crucible is obtained by the above-mentioned method. The maximum temperature of the silicon melt, the temperature at the solid-liquid interface is the minimum temperature of the silicon melt, the difference is calculated as Δt, and the central magnetic field strength of the silicon melt is calculated as M. Item 3. The method for determining oxygen stripe flattening production conditions according to Item 1.   Oxygen fringes in the silicon single crystal grown at the ratio Δt / M were cut out from a silicon single crystal wafer sample cut from the silicon single crystal grown at the ratio Δt / M at a pitch of 2 mm using the FT-IR method. Obtained by measuring the concentration, the oxygen concentration gradient value of the measured oxygen concentration at 6 mm intervals is obtained, and the maximum value of the obtained oxygen concentration gradient value in the plane is 0.04 (ppma-JEIDA / mm) or less. 3. The method according to claim 1, wherein the condition of the ratio .DELTA.t / M is set so as to satisfy the condition.   The method for determining oxygen stripe flattening production conditions according to any one of claims 1 to 3, wherein the ratio Δt / M is set to 0.05 (K / gauss) or less.   The condition of the ratio Δt / M is determined in advance by using the method for determining the operation condition of the oxygen fringe flattening according to claim 1, and a magnetic field application choke is determined based on the determined condition of the ratio Δt / M. A method for producing a silicon single crystal, comprising growing a silicon single crystal by a Ralsky method.   What is claimed is: 1. A silicon single crystal wafer containing oxygen, wherein an in-plane maximum value of an oxygen concentration gradient value at 6 mm intervals of the silicon single crystal wafer is 0.04 (ppma-JEIDA / mm) or less. Crystal wafer.

Images