A crystal growth detection method and system for crystal growth equipment

Abstract

The present invention relates to the field of crystal growth technology, and in particular to a crystal growth detection method and system for crystal growth equipment. The present invention first determines the melt flow vector at each detection moment, then determines the abnormal deposition direction during the crystal growth process based on the crystal contour in the contour image, and obtains the abnormal deposition vector, and further combines the crystal contour in the contour image at the adjacent previous detection moment and the crystal diffraction peak in the XRD spectrum to regulate the magnetic field in the crystal growth furnace. The present invention analyzes the melt concentration distribution and the morphological deviation of the crystal cross-sectional contour in the crystal growth furnace to perform crystal growth detection, evaluate the influence of abnormal convection during the crystal growth process, and determine how to apply an intervening magnetic field to suppress the influence in combination with the crystal contour and the crystal diffraction peak, thereby improving the quality of crystal growth.

Description

Technical Field

[0001] The present invention relates to the technical field of crystal growth, and in particular to a crystal growth detection method and system for crystal growth equipment. Background Art

[0002] Crystals play a vital role in the semiconductor industry and are widely used in the manufacture of infrared optical devices, detectors, solar cells, and other products. Currently, crystals are typically prepared using the Czochralski method. A crystal growth furnace contains a crystal melt, and high-quality single crystals are grown by controlling the continuous growth of a seed crystal (single crystal seed) within the growth furnace. However, during the crystal growth process, the melt within the growth furnace can generate abnormal convection due to fluid dynamics, leading to uneven solute distribution near the crystal growth interface, which in turn affects crystal growth quality. Therefore, real-time monitoring and control of crystal growth is crucial.

[0003] At present, an external magnetic field is generally used to assist in stabilizing melt convection. However, due to the complex changes in the fluid in the crystal growth furnace, the direction and size of abnormal convection may change at any time. The preset fixed magnetic field direction may not be able to cover different convections in time, resulting in uneven melt flow in the crystal growth furnace, causing defects such as vacancies during the crystal growth process, affecting the crystal growth quality. Summary of the Invention

[0004] In order to solve the technical problem of poor crystal growth quality, the purpose of the present invention is to provide a crystal growth detection method and system for crystal growth equipment. The technical solution adopted is as follows:

[0005] A crystal growth detection method for a crystal growth device, the method comprising:

[0006] At each detection moment, the melt concentration at each measuring point in the crystal growth furnace, as well as the contour image and XRD pattern of the crystal interface are obtained, where the main viewing angle of the contour image is parallel to the crystal lifting direction;

[0007] Determining the melt flow vector at each detection moment based on the difference in melt concentration at different measuring points at each detection moment and the change in melt concentration at each measuring point between adjacent detection moments; determining the abnormal deposition direction during crystal growth based on the crystal contour in the contour image at each detection moment, and obtaining the abnormal deposition vector based on the difference in melt concentration at the crystal contour position along the abnormal deposition direction;

[0008] At each detection moment, the magnetic field in the crystal growth furnace is regulated based on the abnormal deposition vector and the melt flow vector, combined with the crystal contour in the contour image and the crystal diffraction peak in the XRD spectrum at the previous detection moment.

[0009] Furthermore, the method for obtaining the melt flow vector includes:

[0010] Obtain the maximum concentration gradient vector at each detection moment; take any detection moment as the target moment, and the previous detection moment before the target moment as the reference moment; at the reference moment, take the melt concentration at the target moment corresponding to the maximum melt concentration as the maximum reference concentration, and take the melt concentration at the target moment corresponding to the minimum melt concentration as the minimum reference concentration;

[0011] Based on the difference between the maximum melt concentration at the reference moment and the maximum reference concentration, and the difference between the minimum melt concentration at the reference moment and the minimum reference concentration, it is determined whether the maximum concentration gradient vector at the reference moment has changed; if it has changed, the maximum concentration gradient vector at the target moment is used as the melt flow vector; if it has not changed, the sum of the maximum concentration gradient vector at the target moment and the maximum concentration gradient vector at the reference moment is used as the melt flow vector at the target moment.

[0012] Furthermore, the method for obtaining the maximum concentration gradient vector includes:

[0013] At each detection moment, the measurement points are clustered based on the clustering algorithm and all melt concentrations; the area corresponding to all measurement points in each cluster is regarded as a concentration-similar area, and the melt concentration of the measurement point corresponding to the cluster center is regarded as the representative concentration of the area;

[0014] The range of the regional representative concentration is taken as the gradient modulus, and the direction from the cluster center corresponding to the maximum regional representative concentration to the cluster center corresponding to the minimum regional representative concentration is taken as the gradient direction. The maximum concentration gradient vector is determined based on the gradient modulus and gradient direction.

[0015] Furthermore, the method for determining whether the maximum concentration gradient vector at the reference moment has changed includes:

[0016] The difference between the maximum melt concentration and the maximum reference concentration is used as the first change parameter; the difference between the minimum reference concentration and the minimum melt concentration is used as the second change parameter; if both the first change parameter and the second change parameter are greater than 0, it is determined that the maximum concentration gradient vector has changed, otherwise there is no change.

[0017] Furthermore, the method for determining the abnormal deposition direction includes:

[0018] Based on the difference between the crystal contour in the contour image and the preset standard growth contour, the maximum convex vector and the maximum concave vector of the crystal in the contour image are determined; the direction of the corresponding vector of the sum of the maximum convex vector and the maximum concave vector is used as the abnormal deposition direction; wherein the preset standard growth contour is different at different detection moments.

[0019] Furthermore, the method for obtaining the maximum convex vector and the maximum concave vector includes:

[0020] At each detection moment, the preset standard growth contour is projected onto the contour image to determine the maximum protrusion direction and maximum protrusion width, as well as the maximum concave direction and maximum concave width of the crystal contour; the maximum protrusion vector is determined based on the maximum protrusion direction and maximum protrusion width, and the maximum concave vector is determined based on the maximum concave direction and maximum concave width.

[0021] Furthermore, the method for obtaining the abnormal deposition vector includes:

[0022] At each detection moment, two contour intersection points on the crystal contour are obtained along the abnormal deposition direction, and the melt concentration at the measurement point closest to each contour intersection point is used as the reference concentration of the corresponding contour intersection point. The difference between the two reference concentrations is used as the modulus of the abnormal deposition vector, and the abnormal deposition vector is determined in combination with the abnormal deposition direction.

[0023] Furthermore, the method for regulating the magnetic field in the crystal growth furnace includes:

[0024] At each detection moment, the difference between the abnormal deposition vector and the melt flow vector is used as the abnormal convection vector, the opposite direction of the vector direction of the abnormal convection vector is used as the magnetic field control direction, and a control weight is determined according to the modulus of the abnormal convection vector. The preset magnetic field intensity is weighted using the control weight to obtain the magnetic field control intensity;

[0025] At each detection moment, a crystal growth quality parameter is obtained based on the degree of overlap between the area enclosed by the crystal outline in the outline image and the area enclosed by the preset standard growth outline, as well as the deviation of the width of the crystal diffraction peak in the XRD pattern relative to the preset standard width; any detection moment is taken as a target moment, and the previous detection moment immediately preceding the target moment is taken as a reference moment; and the rate of change between the crystal growth quality parameters corresponding to the target moment and the reference moment is used as the magnetic field control effect parameter at the reference moment;

[0026] If the magnetic field control effect parameter is greater than or equal to 0, the magnetic field in the crystal growth furnace is controlled based on the magnetic field control direction and magnetic field control intensity at the reference moment; if the magnetic field control effect parameter is less than 0, the magnetic field in the crystal growth furnace is controlled based on the magnetic field control direction and magnetic field control intensity at the target moment.

[0027] Furthermore, the method for obtaining the crystal growth quality parameters includes:

[0028] The overlapping area between the area enclosed by the crystal contour in the contour image and the area enclosed by the preset standard growth contour is used as the first quality parameter, and the negative correlation mapping result of the absolute value of the difference between the width of the crystal diffraction peak in the XRD pattern and the preset standard width is used as the second quality parameter; the first quality parameter and the second quality parameter are combined to obtain the crystal growth quality parameter.

[0029] A crystal growth detection system for crystal growth equipment, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of a crystal growth detection method for crystal growth equipment when executing the computer program.

[0030] The present invention has the following beneficial effects:

[0031] The present invention obtains the melt concentration at each measuring point in the crystal growth furnace, as well as the contour image and XRD spectrum of the crystal interface at each detection moment, with the main viewing angle of the contour image being parallel to the crystal lifting direction, so as to prepare for subsequent crystal growth detection and analysis; then, based on the difference in melt concentration at different measuring points at each detection moment and the change in melt concentration at each measuring point between adjacent detection moments, the melt flow vector reflecting the overall macroscopic flow condition of the melt at each detection moment is determined, so as to prepare for the subsequent comprehensive evaluation of abnormal convection in combination with the crystal deposition condition; at each detection moment, based on the crystal contour in the contour image, the abnormal deposition direction in the crystal growth process is determined, and combined with the difference in melt concentration at the crystal contour position along the abnormal deposition direction, the abnormal deposition vector reflecting the influence of abnormal convection in the crystal growth process is obtained, and then combined with the crystal contour in the contour image at the adjacent previous detection moment and the crystal diffraction peak in the XRD spectrum, the magnetic field in the crystal growth furnace is regulated. The present invention analyzes the melt concentration distribution in the crystal growth furnace and the morphological deviation of the crystal cross-sectional profile to perform crystal growth detection, evaluate the influence of abnormal convection during the crystal growth process, and determine how to apply an intervening magnetic field to suppress the influence in combination with the crystal profile and crystal diffraction peaks, thereby improving the crystal growth quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0033] Figure 1A method flow chart of a crystal growth detection method for a crystal growth device provided by one embodiment of the present invention;

[0034] Figure 2 A flow chart of a method for obtaining a melt flow vector provided by one embodiment of the present invention;

[0035] Figure 3 A schematic diagram of a crystal outline provided by one embodiment of the present invention;

[0036] Figure 4 A flow chart of a method for controlling the magnetic field in a crystal growth furnace provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0037] To further illustrate the technical means and effectiveness of the present invention in achieving its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail a crystal growth detection method and system for a crystal growth device according to the present invention, including its specific implementation, structure, features, and effectiveness. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0038] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0039] The following describes in detail a specific solution of a crystal growth detection method and system for crystal growth equipment provided by the present invention in conjunction with the accompanying drawings.

[0040] See also Figure 1 , which shows a method flow chart of a crystal growth detection method for a crystal growth device according to an embodiment of the present invention, specifically comprising:

[0041] Step S1, at each detection moment, obtaining the melt concentration at each measuring point in the crystal growth furnace, as well as the contour image and XRD spectrum of the crystal interface, wherein the main viewing angle of the contour image is parallel to the crystal lifting direction.

[0042] It should be noted that the growth detection and control methods for different crystals are the same, and an embodiment of the present invention is analyzed and described using germanium crystal as an example.

[0043] In one embodiment of the present invention, an existing detection system in a crystal growth furnace is used or measurement points are evenly distributed on the inner wall of the crystal growth furnace and conductivity sensors are installed to collect the melt concentration at each measurement point in real time, in preparation for the subsequent analysis of the melt flow in the crystal growth furnace and the evaluation of abnormal convection; at the same time, X-ray imaging technology is used to collect a contour image of the crystal interface, wherein the main viewing angle of the contour image is parallel to the crystal lifting direction, that is, the circular cross-sectional profile of the crystal rod is characterized, in preparation for the subsequent evaluation of abnormal deposition caused by abnormal convection during crystal growth; and X-ray diffraction technology is used to detect the crystal growth interface and obtain an XRD spectrum, in preparation for the subsequent evaluation of the crystal interface growth quality to regulate the magnetic field to interfere with the growth.

[0044] The acquisition frequency of melt concentration, contour image and XRD spectrum is set to 1 time per second, and synchronous acquisition starts from the moment the germanium seed crystal (single crystal seed) first penetrates into the melt, so as to obtain the melt concentration at each measuring point at each detection moment, as well as the contour image and XRD spectrum of the crystal interface.

[0045] It should be noted that the above-mentioned collection processes are all existing technologies and will not be described in detail. The implementer can also set the collection frequency by himself.

[0046] Step S2, determining the melt flow vector at each detection moment based on the difference in melt concentration at different measuring points at each detection moment and the change in melt concentration at each measuring point between adjacent detection moments; at each detection moment, determining the abnormal deposition direction during the crystal growth process based on the crystal contour in the contour image, and obtaining the abnormal deposition vector in combination with the difference in melt concentration at the crystal contour position along the abnormal deposition direction.

[0047] During the germanium crystal growth process, an external magnetic field can reduce the impact of internal abnormal convection on the crystal growth quality, thereby allowing the crystal to grow according to the ideal lattice arrangement. However, since abnormal convection cannot be directly detected, the overall macroscopic flow of the melt and abnormal convection will affect crystal growth to a certain extent. Therefore, the embodiment of the present invention first evaluates the overall macroscopic flow of the melt to prepare for the subsequent evaluation of abnormal convection in combination with the crystal growth condition.

[0048] Considering that the distribution of melt concentration is closely related to the flow direction of the melt, and due to the buoyancy effect, the flow direction of the melt will tend from the high-concentration area to the low-concentration area, the melt flow situation can be preliminarily evaluated based on the difference in melt concentration at different measuring points at each detection moment. However, since melt flow is dynamic, the above melt flow situation is preliminarily evaluated based on static data at a single detection moment, which may have errors. Therefore, it needs to be further determined in combination with the change in melt concentration.

[0049] Therefore, the embodiment of the present invention will determine the melt flow vector at each detection moment based on the difference in melt concentration at different measuring points at each detection moment and the change in melt concentration at each measuring point between adjacent detection moments; wherein the melt flow vector reflects the overall macroscopic flow direction of the melt in the crystal growth furnace and the concentration gradient difference in the flow direction.

[0050] Preferably, in one embodiment of the present invention, the method for obtaining the melt flow vector includes:

[0051] See also Figure 2 , which shows a flow chart of a method for obtaining a melt flow vector provided by one embodiment of the present invention, specifically comprising:

[0052] Step S201, obtaining the maximum concentration gradient vector at each detection moment; taking any detection moment as the target moment, and the previous detection moment adjacent to the target moment as the reference moment; at the reference moment, taking the melt concentration of the measuring point corresponding to the maximum melt concentration at the target moment as the maximum reference concentration, and taking the melt concentration of the measuring point corresponding to the minimum melt concentration at the target moment as the minimum reference concentration.

[0053] Considering that the maximum concentration gradient in the crystal growth furnace generally reflects the overall flow condition of the melt, one embodiment of the present invention first obtains the maximum concentration gradient vector at each detection moment.

[0054] In a preferred embodiment of the present invention, considering that the maximum concentration gradient is usually directed from a region with a larger melt concentration to a region with a smaller melt concentration, clustering can cluster measurement point regions corresponding to similar melt concentrations, thereby helping to evaluate the overall melt flow and determine the maximum concentration gradient; therefore, the method for obtaining the maximum concentration gradient vector includes:

[0055] At each detection moment, the measuring points are clustered based on the clustering algorithm and the melt concentrations at all measuring points. The corresponding areas of all measuring points in each cluster are regarded as a concentration-similar area, and the melt concentration of the measuring point corresponding to the cluster center is regarded as the regional representative concentration. The range of the regional representative concentration is taken as the gradient modulus, and the direction from the cluster center (i.e., the measuring point with the maximum regional representative concentration) to the cluster center (i.e., the measuring point with the minimum regional representative concentration) is taken as the gradient direction. Based on the gradient modulus and gradient direction, the maximum concentration gradient vector is determined.

[0056] As an example, the ISODATA algorithm is specifically used to cluster the melt concentration difference as a metric distance. This and the determination of vectors based on modulus and direction are both existing well-known technologies. Implementers can also use other clustering methods, which will not be repeated here.

[0057] In another embodiment of the present invention, the extreme difference in melt concentration at different measuring points can be directly used as the gradient modulus, and the direction from the measuring point corresponding to the maximum melt concentration to the measuring point corresponding to the minimum melt concentration can be used as the gradient direction. Based on the gradient modulus and the gradient direction, the maximum concentration gradient vector can be determined.

[0058] Furthermore, considering that the concentration difference in the direction corresponding to the maximum concentration gradient will change over time, that is, as the melt flows, at each detection moment, the concentration in the area or measuring point corresponding to the maximum melt concentration at the previous adjacent detection moment will decrease, while the concentration in the area or measuring point corresponding to the minimum melt concentration will increase;

[0059] However, if the above situation does not occur, it means that the maximum concentration gradient at the previous detection moment has not changed, and other extreme concentration gradients may appear inside the melt. This extreme concentration gradient and the maximum concentration gradient at the previous detection moment jointly affect the overall flow of the melt; therefore, it is also necessary to determine the concentrations of the measuring points corresponding to the maximum melt concentration and the minimum melt concentration at the detection moment as a reference, in order to evaluate whether the maximum concentration gradient at the previous detection moment has changed and to prepare for judging whether other extreme concentration gradients appear at the detection moment.

[0060] In one embodiment of the present invention, any detection moment is taken as the target moment, and the previous detection moment adjacent to the target moment is taken as the reference moment; at the reference moment, the melt concentration of the measuring point corresponding to the maximum melt concentration at the target moment is taken as the maximum reference concentration, and the melt concentration of the measuring point corresponding to the minimum melt concentration at the target moment is taken as the minimum reference concentration; by changing the target moment, the concentration reference of the measuring points corresponding to the maximum melt concentration and the minimum melt concentration at the previous detection moment at each detection moment can be obtained.

[0061] Step S202, based on the difference between the maximum melt concentration at the reference moment and the maximum reference concentration, and the difference between the minimum melt concentration at the reference moment and the minimum reference concentration, determine whether the maximum concentration gradient vector at the reference moment has changed; if it has changed, use the maximum concentration gradient vector at the target moment as the melt flow vector; if it has not changed, use the sum of the maximum concentration gradient vector at the target moment and the maximum concentration gradient vector at the reference moment as the melt flow vector at the target moment.

[0062] If the concentration difference in the direction corresponding to the maximum concentration gradient changes at the target time, the concentration at the region or measuring point corresponding to the maximum melt concentration at the reference time will relatively decrease at the target time, and the concentration at the region or measuring point corresponding to the minimum melt concentration will relatively increase at the target time. Based on this, in a preferred embodiment of the present invention, the method for determining whether the maximum concentration gradient vector at the reference time has changed includes:

[0063] The difference between the maximum melt concentration and the maximum reference concentration is used as the first change parameter; the difference between the minimum reference concentration and the minimum melt concentration is used as the second change parameter; if both the first change parameter and the second change parameter are greater than 0, it is determined that the maximum concentration gradient vector has changed, otherwise there is no change.

[0064] Among them, the first change parameter reflects the decrease in concentration at the target time in the area or measuring point corresponding to the maximum melt concentration at the reference time; the second change parameter reflects the increase in concentration at the target time in the area or measuring point corresponding to the minimum melt concentration at the reference time; when the maximum concentration gradient vector at the reference time does not change, it means that there are other extreme concentration gradients, and the maximum concentration gradient at the target time is the extreme concentration gradient.

[0065] After determining that an extreme concentration gradient exists at the target moment, the maximum concentration gradient vector at the target moment can be further summed with the maximum concentration gradient vector at the reference moment to determine the melt flow vector at the target moment; wherein, vector operation is a well-known technical means and will not be repeated here.

[0066] By changing the target time, the melt flow vector at each detection time can be obtained.

[0067] Considering that during the growth process of germanium crystal, if the melt flow direction is stable and uniform, the formed crystal will appear as a regular cylinder with uniform diameter; if abnormal convection occurs, the deposition of single crystal germanium will shift with the melt flow, that is, the diameter of the growth interface changes, forming abnormal deposition and resulting in uneven cylindrical shape; at the same time, if abnormal convection promotes the melt flow at a certain position of the crystal growth interface, the deposition at this position will appear as a convex, and vice versa, the deposition will appear as a concave.

[0068] Among them, the uneven growth interface is the result of the combined action of melt flow and abnormal convection. Therefore, after obtaining the melt flow vector at each detection moment, the embodiment of the present invention further analyzes the morphological uniformity of the crystal growth interface based on the crystal contour in the contour image, thereby analyzing the abnormal deposition situation during the crystal growth process and determining the abnormal deposition direction; further combined with the difference between the melt concentrations at the crystal contour position in the abnormal deposition direction, the degree of deposition unevenness of the crystal growth interface, that is, the abnormal deposition modulus length, is evaluated, thereby obtaining the abnormal deposition vector.

[0069] See also Figure 3 , which shows a schematic diagram of a crystal outline provided by an embodiment of the present invention; Figure 3 The medium-thick irregular shape represents the crystal outline, while the thin regular circle represents the preset standard growth outline.

[0070] Preferably, in one embodiment of the present invention, considering that the abnormal deposition direction is generally from depression to deposition, the method for determining the abnormal deposition direction includes:

[0071] Based on the difference between the crystal contour in the contour image and the preset standard growth contour, the maximum convex vector and the maximum concave vector of the crystal in the contour image are determined; the direction of the corresponding vector of the sum of the maximum convex vector and the maximum concave vector is used as the abnormal deposition direction; wherein the preset standard growth contour is different at different detection moments.

[0072] It should be noted that when the crystal contour in the contour image completely coincides with the preset standard growth contour, it is determined that the crystal is growing normally, and no additional intervention or subsequent analysis is required.

[0073] In a preferred embodiment of the present invention, the method for obtaining the maximum convex vector and the maximum concave vector includes:

[0074] At each detection moment, the preset standard growth contour is projected onto the contour image to determine the maximum protrusion direction and maximum protrusion width, as well as the maximum concave direction and maximum concave width of the crystal contour; the maximum protrusion vector is determined based on the maximum protrusion direction and maximum protrusion width, and the maximum concave vector is determined based on the maximum concave direction and maximum concave width.

[0075] As an example, firstly, the closed edge in the contour image, i.e., the crystal contour, is obtained based on the edge detection algorithm. Then, the preset standard growth contour at the detection moment is projected onto the contour image for comparison reference to determine the maximum protruding area and the maximum concave area, i.e. Figure 3 The maximum protruding area A and the maximum concave area B in the image are obtained, and the maximum protruding width in the maximum protruding area A and the maximum concave width in the maximum concave area B are determined;

[0076] The maximum bulge direction is the direction from the preset standard growth contour to the crystal contour at the maximum bulge width, and the maximum concave direction is the direction from the preset standard growth contour to the crystal contour at the maximum concave width; the maximum bulge direction is used as the vector direction, and the maximum bulge width is used as the vector modulus to determine the maximum bulge vector, and similarly, the maximum concave vector is determined; the vectors are then summed, and the direction of the summed vector is the abnormal deposition direction.

[0077] It should be noted that obtaining the crystal contour, determining and mapping the preset standard growth contour at each detection moment, determining the maximum protruding area and the maximum concave area, obtaining the maximum protruding direction and the maximum protruding width, obtaining the maximum concave direction and the maximum concave width, vector construction and vector operations are all existing technologies and will not be repeated here.

[0078] After determining the abnormal deposition direction at each detection moment, the abnormal deposition degree can be further determined, thereby constructing the abnormal deposition vector.

[0079] Preferably, in one embodiment of the present invention, considering that abnormal deposition is caused by melt concentration difference, the melt concentration difference can be reflected by evaluating the melt concentration at the measuring point closest to the crystal contour in the abnormal deposition direction, thereby evaluating the degree of deposition non-uniformity at the crystal growth interface at the detection moment; therefore, the method for obtaining the abnormal deposition vector includes:

[0080] At each detection moment, two contour intersection points on the crystal contour are obtained along the abnormal deposition direction. The melt concentration at the measurement point closest to each contour intersection point is used as the reference concentration of the corresponding contour intersection point. The difference between the two reference concentrations is used as the modulus of the abnormal deposition vector, and the abnormal deposition vector is determined in combination with the abnormal deposition direction.

[0081] As an example, the two contour points where the extended straight line of the abnormal deposition direction intersects the crystal contour are respectively used as contour intersection points, and the corresponding reference concentrations are obtained; then the difference is measured by the absolute value of the difference to determine the modulus of the abnormal deposition vector, and then the abnormal deposition vector is constructed.

[0082] Step S3, at each detection moment, according to the abnormal deposition vector and the melt flow vector, combined with the crystal contour in the contour image of the previous detection moment and the crystal diffraction peak in the XRD spectrum, the magnetic field in the crystal growth furnace is regulated.

[0083] Since abnormal deposition is the result of the combined effects of melt flow and abnormal convection, the embodiments of the present invention can preliminarily assess the abnormal convection situation at each detection moment based on the abnormal deposition vector and melt flow vector, thereby regulating the magnetic field for intervention to reduce the impact of abnormal convection on crystal growth.

[0084] In addition, since the production cost of frequently adjusting the magnetic field is high, when the growth quality of the crystal is improved after the magnetic field is adjusted at the previous detection moment, it means that the magnetic field intervention at the previous detection moment can appropriately reduce the impact of abnormal convection on crystal growth. The magnetic field can be kept unchanged at this detection moment, thereby reducing the adjustment frequency of the magnetic field while ensuring that abnormal convection can be suppressed, and optimizing the cost of magnetic field adjustment. Therefore, the embodiment of the present invention will further combine the crystal contour in the contour image of the adjacent previous detection moment and the crystal diffraction peak in the XRD spectrum to regulate the magnetic field in the crystal growth furnace.

[0085] Preferably, in one embodiment of the present invention, the method for regulating the magnetic field in the crystal growth furnace includes:

[0086] See also Figure 4, which shows a flow chart of a method for controlling the magnetic field in a crystal growth furnace provided by one embodiment of the present invention, specifically comprising:

[0087] Step S301: At each detection moment, the difference between the abnormal deposition vector and the melt flow vector is used as the abnormal convection vector, the opposite direction of the vector direction of the abnormal convection vector is used as the magnetic field control direction, and the control weight is determined according to the module length of the abnormal convection vector. The preset magnetic field strength is weighted using the control weight to obtain the magnetic field control intensity.

[0088] Considering that the abnormal convection vector can be obtained by subtracting the abnormal deposition vector from the melt flow vector, and the influence of abnormal convection on crystal growth can be suppressed by applying a magnetic field in the opposite direction to the abnormal convection vector, the direction of magnetic field control can be preliminarily determined, and the intensity of magnetic field control needs to be further determined; considering that the larger the abnormal convection vector, the greater the corresponding magnetic field intervention intensity, in order to suppress the influence of abnormal convection.

[0089] As an example, the modulus of the abnormal convection vector is used as the numerator, the modulus of the melt flow vector is used as the denominator, and the fractional ratio is used as the control weight. Then, the control weight is multiplied by the preset magnetic field strength to obtain the magnetic field control intensity. The larger the adjustment weight, the more severe the abnormal convection, and the greater the magnetic field strength is required to suppress the influence of abnormal convection.

[0090] It should be noted that during the growth of germanium crystals, the magnetic field strength is generally set within the range of 50-500 Gauss. The embodiment of the present invention sets the preset magnetic field strength to 200, and the implementer can also adjust it according to the actual situation in the crystal growth furnace.

[0091] Step S302: At each detection moment, the crystal growth quality parameter is obtained based on the degree of overlap between the area enclosed by the crystal contour in the contour image and the area enclosed by the preset standard growth contour, and the deviation of the width of the crystal diffraction peak in the XRD spectrum from the preset standard width; any detection moment is used as the target moment, and the previous detection moment adjacent to the target moment is used as the reference moment; and the rate of change between the crystal growth quality parameters corresponding to the target moment and the reference moment is used as the magnetic field control effect parameter at the reference moment.

[0092] Considering that if the crystal growth quality improves after the magnetic field is adjusted at the previous detection moment, it means that the magnetic field intervention at the previous detection moment can appropriately reduce the impact of abnormal convection on crystal growth, and the magnetic field can be kept unchanged at this detection moment; therefore, the embodiment of the present invention first evaluates the crystal growth quality parameters at each detection moment, and then evaluates the magnetic field control effect parameters at the previous detection moment;

[0093] It is also considered that if the degree of overlap between the area enclosed by the crystal contour in the contour image and the area enclosed by the preset standard growth contour is higher, the crystal growth quality is higher; at the same time, the diffraction peak width in the XRD pattern of the crystal interface can reflect the vacancy defect situation in the crystal. Vacancy defects usually cause local irregularities in the crystal structure, which in turn causes the diffraction peak to become wider. Therefore, when the deviation of the width of the crystal diffraction peak in the XRD pattern relative to the preset standard width is smaller, the crystal growth quality is higher.

[0094] Based on this, in a preferred embodiment of the present invention, the method for obtaining crystal growth quality parameters includes:

[0095] The overlapping area between the area enclosed by the crystal contour in the contour image and the area enclosed by the preset standard growth contour is used as the first quality parameter, and the negative correlation mapping result of the absolute value of the difference between the width of the crystal diffraction peak in the XRD spectrum and the preset standard width is used as the second quality parameter; the first quality parameter and the second quality parameter are combined to obtain the crystal growth quality parameter.

[0096] As an example, the absolute value of the difference is added to a preset minimum positive parameter of 0.001 and then a reciprocal operation is performed to perform a negative correlation mapping to obtain the second quality parameter; then the first quality parameter and the second quality parameter are multiplied and fused to obtain the crystal growth quality parameter at the corresponding detection moment;

[0097] Then, any detection moment is taken as the target moment, and the previous detection moment of the target moment is taken as the reference moment; at the reference moment, the crystal growth quality parameter of the target moment is taken as the numerator, and the crystal growth quality parameter of the reference moment is taken as the denominator, and then the ratio of the fraction is subtracted by 1 to obtain the rate of change, that is, the magnetic field control effect parameter of the reference moment; by changing the target moment, the magnetic field control effect parameter of the previous adjacent detection moment of each detection moment can be obtained, thereby evaluating the magnetic field control frequency.

[0098] It should be noted that obtaining the preset standard width is already an existing technology, and implementers can obtain it by referring to relevant information on germanium crystals, and the acquisition process will not be described in detail.

[0099] Step S303: If the magnetic field control effect parameter is greater than or equal to 0, the magnetic field in the crystal growth furnace is controlled based on the magnetic field control direction and magnetic field control intensity at the reference moment; if the magnetic field control effect parameter is less than 0, the magnetic field in the crystal growth furnace is controlled based on the magnetic field control direction and magnetic field control intensity at the target moment.

[0100] When the magnetic field control effect parameter is greater than or equal to 0, it indicates that the crystal growth quality is stable or improved, further indicating that the magnetic field controlled at the reference time can suppress the influence of abnormal convection. Therefore, magnetic field control is not required at the target time, and the magnetic field control direction and intensity at the reference time are maintained.

[0101] When the magnetic field control effect parameter is less than 0, it indicates that the quality of crystal growth has declined, further indicating that the magnetic field controlled at the reference time is insufficient to suppress the influence of abnormal convection, causing the crystal growth to shift or abnormal deposition. In this case, the magnetic field needs to be re-controlled at the target time, based on the magnetic field control direction and magnetic field control intensity at the target time obtained in step S301.

[0102] It should be noted that adjusting the magnetic field strength and direction in the crystal growth furnace is already an existing technology and will not be described in detail.

[0103] The present invention also proposes a crystal growth detection system for crystal growth equipment. The system includes a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the computer program, the steps of a crystal growth detection method for crystal growth equipment are implemented.

[0104] In summary, the present invention determines the melt flow vector at each detection moment based on the difference in melt concentration at different measuring points and the change in melt concentration at each measuring point between adjacent detection moments, then determines the abnormal deposition direction during crystal growth based on the crystal contour in the contour image, and obtains the abnormal deposition vector in combination with the difference between the melt concentrations at the crystal contour position in the abnormal deposition direction, and further regulates the magnetic field in the crystal growth furnace based on the abnormal deposition vector and the melt flow vector, combined with the crystal contour in the contour image at the adjacent previous detection moment and the crystal diffraction peak in the XRD spectrum. The present invention analyzes the melt concentration distribution in the crystal growth furnace and the morphological deviation of the crystal cross-sectional contour to perform crystal growth detection, evaluate the influence of abnormal convection during crystal growth, and determine how to apply an intervening magnetic field to suppress the influence in combination with the crystal contour and the crystal diffraction peak, thereby improving the quality of crystal growth.

[0105] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0106] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A crystal growth detection method for a crystal growth device, characterized in that: The method comprises: At each detection moment, the melt concentration at each measuring point in the crystal growth furnace, as well as the contour image and XRD pattern of the crystal interface are obtained, where the main viewing angle of the contour image is parallel to the crystal lifting direction; Determining the melt flow vector at each detection moment based on the difference in melt concentration at different measuring points at each detection moment and the change in melt concentration at each measuring point between adjacent detection moments; determining the abnormal deposition direction during crystal growth based on the crystal contour in the contour image at each detection moment, and obtaining the abnormal deposition vector based on the difference in melt concentration at the crystal contour position along the abnormal deposition direction; At each detection moment, according to the abnormal deposition vector and the melt flow vector, combined with the crystal contour in the contour image and the crystal diffraction peak in the XRD spectrum at the previous detection moment, the magnetic field in the crystal growth furnace is regulated; The method for determining the abnormal deposition direction includes: Determining a maximum convex vector and a maximum concave vector of the crystal in the contour image based on a difference between the crystal contour in the contour image and a preset standard growth contour; using the direction of a corresponding vector of the sum of the maximum convex vector and the maximum concave vector as the abnormal deposition direction; wherein the preset standard growth contour is different at different detection moments; Methods for regulating the magnetic field in a crystal growth furnace include: At each detection moment, the difference between the abnormal deposition vector and the melt flow vector is used as the abnormal convection vector, the opposite direction of the vector direction of the abnormal convection vector is used as the magnetic field control direction, and a control weight is determined according to the modulus of the abnormal convection vector. The preset magnetic field intensity is weighted using the control weight to obtain the magnetic field control intensity; At each detection moment, a crystal growth quality parameter is obtained based on the degree of overlap between the area enclosed by the crystal outline in the outline image and the area enclosed by the preset standard growth outline, as well as the deviation of the width of the crystal diffraction peak in the XRD pattern relative to the preset standard width; any detection moment is taken as a target moment, and the previous detection moment immediately preceding the target moment is taken as a reference moment; and the rate of change between the crystal growth quality parameters corresponding to the target moment and the reference moment is used as the magnetic field control effect parameter at the reference moment; If the magnetic field control effect parameter is greater than or equal to 0, the magnetic field in the crystal growth furnace is controlled based on the magnetic field control direction and magnetic field control intensity at the reference moment; if the magnetic field control effect parameter is less than 0, the magnetic field in the crystal growth furnace is controlled based on the magnetic field control direction and magnetic field control intensity at the target moment.

2. The crystal growth detection method for crystal growth equipment according to claim 1, characterized in that: The method for obtaining the melt flow vector includes: Obtain the maximum concentration gradient vector at each detection moment; take any detection moment as the target moment, and the previous detection moment before the target moment as the reference moment; at the reference moment, take the melt concentration at the target moment corresponding to the maximum melt concentration as the maximum reference concentration, and take the melt concentration at the target moment corresponding to the minimum melt concentration as the minimum reference concentration; Based on the difference between the maximum melt concentration at the reference moment and the maximum reference concentration, and the difference between the minimum melt concentration at the reference moment and the minimum reference concentration, it is determined whether the maximum concentration gradient vector at the reference moment has changed; if it has changed, the maximum concentration gradient vector at the target moment is used as the melt flow vector; if it has not changed, the sum of the maximum concentration gradient vector at the target moment and the maximum concentration gradient vector at the reference moment is used as the melt flow vector at the target moment.

3. The crystal growth detection method for crystal growth equipment according to claim 2, characterized in that: The method for obtaining the maximum concentration gradient vector includes: At each detection moment, the measurement points are clustered based on the clustering algorithm and all melt concentrations; the area corresponding to all measurement points in each cluster is regarded as a concentration-similar area, and the melt concentration of the measurement point corresponding to the cluster center is regarded as the representative concentration of the area; The range of the regional representative concentration is taken as the gradient modulus, and the direction from the cluster center corresponding to the maximum regional representative concentration to the cluster center corresponding to the minimum regional representative concentration is taken as the gradient direction. The maximum concentration gradient vector is determined based on the gradient modulus and gradient direction.

4. The crystal growth detection method for crystal growth equipment according to claim 2, characterized in that: Methods for determining whether the maximum concentration gradient vector at a reference moment has changed include: The difference between the maximum melt concentration and the maximum reference concentration is used as the first change parameter; the difference between the minimum reference concentration and the minimum melt concentration is used as the second change parameter; if both the first change parameter and the second change parameter are greater than 0, it is determined that the maximum concentration gradient vector has changed, otherwise there is no change.

5. The crystal growth detection method for crystal growth equipment according to claim 1, characterized in that: The method for obtaining the maximum convex vector and the maximum concave vector includes: At each detection moment, the preset standard growth contour is projected onto the contour image to determine the maximum protrusion direction and maximum protrusion width, as well as the maximum concave direction and maximum concave width of the crystal contour; the maximum protrusion vector is determined based on the maximum protrusion direction and maximum protrusion width, and the maximum concave vector is determined based on the maximum concave direction and maximum concave width.

6. The crystal growth detection method for crystal growth equipment according to claim 1, characterized in that: The method for obtaining the abnormal deposition vector includes: At each detection moment, two contour intersection points on the crystal contour are obtained along the abnormal deposition direction, and the melt concentration at the measurement point closest to each contour intersection point is used as the reference concentration of the corresponding contour intersection point. The difference between the two reference concentrations is used as the modulus of the abnormal deposition vector, and the abnormal deposition vector is determined in combination with the abnormal deposition direction.

7. The crystal growth detection method for crystal growth equipment according to claim 1, characterized in that: The method for obtaining the crystal growth quality parameters includes: The overlapping area between the area enclosed by the crystal contour in the contour image and the area enclosed by the preset standard growth contour is used as the first quality parameter, and the negative correlation mapping result of the absolute value of the difference between the width of the crystal diffraction peak in the XRD pattern and the preset standard width is used as the second quality parameter; the first quality parameter and the second quality parameter are combined to obtain the crystal growth quality parameter.

8. A crystal growth detection system for a crystal growth device, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the crystal growth detection method for a crystal growth device as claimed in any one of claims 1 to 7 are implemented.

Images