JP2015161675A - Scribe surface quality measuring and controlling device and scribe surface quality measuring and controlling method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a scribe surface quality measuring and controlling device that realizes acquisition of a high-quality scribe surface by appropriately applying automatic control to a controlled object such as a scribing wheel, a laser beam or the like in scribing processing.SOLUTION: A scribe surface quality measuring and controlling device that applies scribing to a hard but brittle material by controlling prescribed scribing means comprises: polarization high-speed image sensors 10 and 20 that measure polarization-property spatial distributions of the hard but brittle material at high speed and in real time during scribing of the hard but brittle material; a storage unit that stores a database where a scribing condition for the hard but brittle material is associated with the polarization-property spatial distributions of the hard but brittle material; and a scribing control unit 40 that obtains an optimal scribing condition for the scribing means by referring to the database on the basis of the polarization-property spatial distributions measured by the polarization high-speed image sensors, and uses the obtained scribing condition to control the scribing means in real time.

Description

  The present invention relates to a cleaving quality measurement control device and a cleaving quality measurement control method, and more particularly to a cleaving quality measurement control device and a cleaving quality measurement control method for measuring and controlling cleaving quality of hard and brittle materials such as glass in real time.

  Cleaving is a technology that splits cracks formed in the surface or inside of hard and brittle materials such as glass by mechanical and thermal stress. In the industry, mechanical cleaving and laser cleaving are used. The law is widely used. In the mechanical cleaving method, an initial crack is formed on glass using a tool called a scribing wheel, and the initial crack is propagated by an external force. On the other hand, the laser cleaving method can be classified into two types: a method of cleaving by locally heating and applying thermal stress using laser light absorbed by the glass, and condensing the laser light inside the glass. There is a method of cleaving by altering the inside of the glass or generating cracks. In recent years, a hybrid cleaving method that simultaneously performs crack generation by a scribing wheel and division by laser irradiation has been proposed.

  In the glass cleaving field using these cleaving methods, processing conditions are often determined empirically, and no technology for measuring cleaving quality in real time during processing is found.

  On the other hand, there is an example in which alignment characteristics of a liquid crystal alignment film formed on a glass substrate are diagnosed by polarization analysis as a method for determining whether the processing quality of a transparent material is good or bad (Patent Document 1).

  A photoelastic method is known as a method for measuring a main stress difference and a main stress direction when a load stress is applied to a substance exhibiting birefringence due to stress.

Japanese Patent Application Laid-Open No. 09-125891

  The strength of hard and brittle materials such as glass is greatly affected by the finish of the end face. For example, if there is a crack or the like in the fractured surface of the glass after cleaving, the glass is cracked from the crack just by being bent slightly, and the strength is significantly reduced. For this reason, the finish of the fractured surface is very important, and it is required to improve the quality of the fractured surface, that is, to obtain a smooth fractured surface free from cracks.

  As described above, the cleaving of glass or the like is performed using a scribing wheel or laser light. In this case, it is considered that a high-quality fractured surface can be obtained by appropriately controlling the load and moving speed of the scribing wheel on the glass or appropriately controlling the intensity of the laser beam. However, as described above, there is no prior art for measuring data related to cleaving quality in real time during cleaving, and there is no prior art for controlling the scribing wheel and laser light by feeding back the measured data. .

  The present invention has been made in view of the above circumstances, and a cleaving quality measurement control device and a cleaving device that can appropriately and automatically control an object to be cleaved, such as a scribing wheel or a laser beam, to obtain a high quality cleaved surface. An object is to provide a quality measurement control method.

  In order to solve the above-mentioned problem, the cleaving quality measurement control device according to the present invention controls a predetermined cleaving means to cleave a hard and brittle material, while cleaving the hard and brittle material, A polarization high-speed image sensor that measures the spatial distribution of polarization characteristics of hard and brittle materials at high speed and in real time, and a database that correlates the cleaving conditions for the hard and brittle materials and the spatial distribution of polarization characteristics of the hard and brittle materials. By storing the storage unit and referring to the database based on the spatial distribution of the polarization characteristics measured by the polarization high-speed image sensor, an optimum cleaving condition for the cleaving means is obtained, and the obtained cleaving condition is used. And a cleaving control unit for controlling the cleaving means in real time.

  According to the cleaving quality measurement control device and the cleaving quality measurement control method according to the present invention, it is possible to appropriately automatically control a cleaving wheel, laser beam, or the like to be cut and to obtain a high quality cleaved surface.

The figure which shows the structural example of the cleaving quality measurement control apparatus which concerns on this embodiment. The figure which shows typically the quality of cleaving quality. The figure which shows typically the characteristic structure of the polarization high-speed image sensor which produces | generates a phase difference image. The figure which illustrates typically a mode that a three-dimensional phase difference image is produced | generated from two phase difference images. The figure explaining the concept of the real-time control which a cutting control part performs.

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

  FIG. 1 is a diagram illustrating a configuration example of a cleaving quality measurement control device 1 according to an embodiment of the present invention. The cleaving quality measurement control device 1 includes a polarization high-speed image sensor (1), (2) (10, 20), a three-dimensional phase difference image generation unit 30, a cleaving control unit 40, and a storage unit that stores an analysis value / experiment database 60. 50 etc. are comprised.

  As shown in the upper left of FIG. 1, when a hard and brittle material such as glass 201 is cleaved, an initial crack is generated on the surface of the glass 201 by a cleaving means called a scribing wheel 200. After generating the initial crack, the glass 201 is divided by applying an external force to the glass 201. This is the cutting of the glass 201.

  Depending on how the initial cracks are generated, the quality of the cross section of the glass 201 after the cutting (hereinafter referred to as the cutting quality) greatly depends. FIG. 2 is a diagram schematically showing the quality of the cleaving quality. FIG. 2A shows a case of good cleaving quality, and the cleaved surface has no irregularities and is a smooth surface. On the other hand, FIG. 2B shows a case in which the cleaving quality is poor, and irregularities such as cracks are generated in the cleaved surface, which causes a decrease in strength.

  The cleaving quality measurement control device 1 of the embodiment measures the spatial distribution of the stress of the glass 201 generated during the cleaving of the glass 201 as the spatial distribution of the polarization characteristic of the glass 201, and uses this measurement value to determine the scribing wheel 200. The cleaving conditions for the cleaving means such as the above are controlled in real time, thereby obtaining a satisfactory cleaving quality.

  Polarization high-speed image sensors (1) 10 and (2) 20 shown in FIG. 1 are imaging cameras that measure the spatial distribution of the polarization characteristics of glass at high speed in real time while cleaving a hard and brittle material such as glass 201. The polarization high-speed image sensor (1) 10 images the cross section of the glass 201 from the traveling direction of the scribing wheel 200, that is, the direction parallel to the split cross section. A light source (not shown) is installed on the opposite side of the polarization high-speed image sensor (1) 10 across the glass 201, and light converted into linearly polarized light through a polarizing plate (not shown) is transmitted through the glass 201 and polarized. The high-speed image sensor (1) 10 is reached.

  On the other hand, the polarization high-speed image sensor (2) 20 images the cross section of the glass 201 from a direction orthogonal to the traveling direction of the scribing wheel 200. Similarly, a light source (not shown) is installed on the opposite side of the polarization high-speed image sensor (2) 20 with the glass 201 interposed therebetween, and light converted into linearly polarized light passes through the glass 201 and is polarized high-speed image sensor (2 ) 20 is reached.

  The number of polarization high-speed image sensors is not limited to two illustrated in FIG. 2, and imaging may be performed from three or more directions by three or more polarization high-speed image sensors. The reason why a plurality of high-speed polarization image sensors are provided is to generate a three-dimensional image from a plurality of two-dimensional images. When only a two-dimensional image is used, the number of high-speed polarization image sensors may be one.

  The polarization high-speed image sensors (1) 10 and (2) 20 measure the spatial distribution of the polarization characteristics of the glass. More specifically, the spatial distribution of the polarization phase difference δ is measured.

Now, let the traveling direction of light be Z, and let the two orthogonal directions in the plane perpendicular to the traveling direction be the X direction and the Y direction, respectively. At this time, assuming that the polarization components in the X direction and the Y direction are E _X and E _Y , respectively, E _X and E _Y can be expressed by the following equations, respectively.

E _X = A _X cos (kz−ωt) (Formula 1)
E _Y = A _Y cos (kz−ωt + δ) (Formula 2)
Here, A _X and A _Y c are the amplitudes of the X component and the Y component, k is the wave number (= 2π / λ), λ is the wavelength, ω is the frequency of light, and t is time. Further, δ is a phase difference and is also called retardation.

  The polarization high-speed image sensors (1) 10 and (2) 20 generate the spatial distribution of the phase difference δ, that is, a phase difference image.

  FIG. 3 schematically shows a characteristic configuration of the polarization high-speed image sensors (1) 10 and (2) 20 that generate a phase difference image. The polarization high-speed image sensors (1) 10 and (2) 20 have a pixel array 101 of CCD or CMOS as shown in the lower right of FIG. On each pixel, for example, a photonic crystal polarizer 102 generated by a photonic crystal is provided. Each photonic crystal polarizer 102 is formed with a fine groove-like pattern. In the example shown in FIG. 3, the photonic crystal polarizer 102 constitutes one set of four polarizers whose pattern directions are changed by 45 degrees.

  The transmission axis as a polarizer is determined by the direction of the pattern. Accordingly, the phase difference δ can be obtained by performing a predetermined calculation on the four outputs of the pixels corresponding to the four polarizers having different pattern directions. Since the phase difference δ is calculated every four pixels, a spatial distribution of the phase difference δ, that is, a phase difference image can be obtained as the output of the polarization high-speed image sensors (1) 10 and (2) 20.

  As shown in FIG. 1, the phase difference images output from the two polarization high-speed image sensors (1) 10 and (2) 20 are input to a three-dimensional phase difference image generation unit 30, where a three-dimensional phase difference image is obtained. Is generated.

  FIG. 4 is a diagram schematically illustrating how a three-dimensional phase difference image is generated from two phase difference images. In FIG. 4, the diagram shown in the lower left is a diagram showing the phase difference image (1) output from the polarization high-speed image sensor (1) 10 arranged in the traveling direction of the scribing wheel 200, and the diagram shown in the upper left is It is a figure which shows the phase difference image (2) output from the polarization high-speed image sensor (2) 20 arrange | positioned in the direction orthogonal to the advancing direction of the scribing wheel 200. FIG. The shading patterns of the phase difference images (1) and (2) schematically show the magnitude of the phase difference δ.

  If two phase difference images (1) and (2) are obtained, a three-dimensional phase difference image can be generated by a method such as a stereo method. Note that the three-dimensional phase difference image generated here is obtained by measurement by the polarization high-speed image sensors (1) 10 and (2) 20, and is distinguished from the three-dimensional phase difference image obtained by analysis described later. Therefore, it shall be called a three-dimensional phase difference image (measurement value).

  The three-dimensional phase difference image (measured value) is expressed as a three-dimensional spatial distribution of the phase difference δ. The resolution of this spatial distribution is the number of pixels of the polarization high-speed image sensors (1) 10 and (2) 20. It can be increased by increasing. A three-dimensional phase difference image (measured value) is obtained for each imaging frame of the polarization high-speed image sensors (1) 10 and (2) 20. Therefore, if the frame rate of the polarization high-speed image sensors (1) 10 and (2) 20 is increased, a three-dimensional phase difference image (measurement value) with higher time resolution can be obtained. For example, by constructing the polarization high-speed image sensors (1) 10, (2) 20 based on the technology of a high-speed imaging camera of 50,000 fps (frame per second) or more, high time resolution of 20 microseconds or less is achieved. A three-dimensional phase difference image (measurement value) can be obtained.

  As described above, since a three-dimensional phase difference image (measurement value) with high time resolution is obtained during the cleaving of the glass 201, real-time control of the scribing wheel 200 based on the three-dimensional phase difference image (measurement value) is achieved. It becomes possible. This real-time control is executed by the cleaving control unit 40 shown in FIG. 1 based on the three-dimensional phase difference image (measured value) and the analysis value / experiment database 60.

  FIG. 5 is a diagram for explaining the concept of real-time control performed by the cleaving control unit 40. First, the analysis value / experiment database 60 will be described. The analysis value / experiment database 60 includes an analysis value database 601 and an experiment database 602.

  The analysis value database 601 is created by analysis based on the cleaving condition and the material condition. As the cleaving condition, for example, a load on the glass 201 of the scribing wheel 200 and a moving speed of the scribing wheel 200 are set. As material conditions, for example, physical constants such as the thickness and size of the glass 201 and the elastic coefficient of the glass 201 are set. Next, stress / strain analysis using a finite element method or the like is performed based on these cleaving conditions and material condition setting conditions. Thereafter, a three-dimensional phase difference image (analyzed value) is calculated from the stress distribution calculated by the analysis. As is well known, Brewster's law expressed by the following equation exists between the phase difference δ due to polarized light and the stress.

δ = (2π / λ) C · h · (σ1-σ2) (Formula 3)
Here, λ is a wavelength, C is a photoelastic constant, h is a thickness of a transparent flat plate (such as glass 201) receiving a load, and (σ1−σ2) is a main stress difference. By (Equation 3), the spatial distribution of the main stress difference can be obtained from the spatial distribution of the stress calculated by analysis, and the spatial distribution of the phase difference δ (three-dimensional phase difference image (analyzed value)) is calculated. can do.

  A large number of three-dimensional phase difference images (analysis values) can be obtained by analysis with different cleaving conditions and material conditions. The obtained three-dimensional phase difference image (analysis value) is stored in advance in the storage unit 50 as an analysis value database 601 associated with the cleaving condition or material condition.

  On the other hand, the experiment database 602 is a database that associates the cleaving condition with the cleaving quality. Cleaving experiments are performed with various cleaving conditions. And the cutting quality corresponding to each cutting condition is calculated | required and these are preserve | saved beforehand at the memory | storage part 50 as a database. The determination of the cleaving quality may be determined based on some evaluation index, and is not particularly limited. For example, a method may be used in which the presence / absence and amount of cracks in the fractured surface are measured and classified into multi-stage ranks based on the measurement results.

  The cleaving control unit 40 refers to the analysis value database 601 and the experiment database 602 based on the three-dimensional phase difference image (measured value) output from the three-dimensional phase difference image generation unit 30, and performs the optimum cleaving for the scribing wheel 200. Conditions (optimum load, optimum moving speed) are obtained, and the scribing wheel 200 is controlled in real time using the obtained cleaving conditions.

  For example, the three-dimensional phase difference image (measurement value) is compared with the three-dimensional phase difference image (analysis value) in the analysis value database 601, and the three-dimensional phase difference closest to the three-dimensional phase difference image (measurement value) is obtained. The cleaving condition associated with the image (analysis value) is estimated to be the cleaving condition applied to the glass 201 that is currently being cleaved. On the other hand, by referring to the experimental database, it is possible to know a desired cleaving condition for obtaining a desired cleaving quality. Therefore, the cleaving control unit 40 corrects the cleaving condition so as to bring the estimated current cleaving condition closer to the desired cleaving condition, and controls the scribing wheel 200 by feeding back the corrected cleaving condition. Since the three-dimensional phase difference image (measurement value) can be obtained with high time resolution, the hood back loop can be turned in a short time, and the optimum cleaving condition (the optimum load and the optimum moving speed) can be reached in a short time. be able to. Even when the cleaving condition deviates from the optimum state, it can be restored to the optimum state in a short time.

  The above-described control method based on the database is merely an example and is not limited to this, and various optimization methods can be taken as the control method based on the database.

  A control direction using a two-dimensional phase difference image (measured value) generated from one polarization high-speed image sensor and a database is also conceivable with only one imaging direction. In this case, a three-dimensional phase difference image (analysis value) calculated based on the finite element method is integrated in the one imaging direction to obtain a two-dimensional phase difference image (analysis value). An image (analysis value) may be held as an analysis value database.

  In addition, a control method based only on the behavior of a two-dimensional phase difference image (measurement value) or a three-dimensional phase difference image (measurement value) without using a database is also conceivable. For example, a two-dimensional phase difference image (measurement value) or a specific point in a three-dimensional phase difference image (measurement value) or an observation phase difference δ on a specific line segment is compared with a predetermined threshold, and the observation phase difference δ When the value exceeds the threshold value, the applied load and the phase speed of the scribing wheel 200 may be controlled so as to return the value to fall within the threshold value.

  So far, the scribing wheel 200 has been described as an example of the cleaving means, but the cleaving means using laser light can be controlled in the same manner. In this case, the intensity of the laser light and the moving speed of the laser light source are controlled.

  In the above description, the hard and brittle material to be cleaved has been described as being glass 201. However, the present invention is not limited to this, and the present invention can be applied to dicing of a wafer material such as SiC.

  As described above, according to the cleaving quality measurement control device 1 and the cleaving quality measurement control method according to the embodiment, the control target of cleaving processing such as a scribing wheel and a laser beam is appropriately automatically controlled, and a high quality cleaving control is performed. A cross section can be obtained.

  Note that the present invention is not limited to the above-described embodiments as they are, and can be embodied by modifying the constituent elements without departing from the scope of the invention in the implementation stage. In addition, various inventions can be formed by appropriately combining a plurality of components disclosed in the embodiment. For example, some components may be deleted from all the components shown in the embodiment. Furthermore, the constituent elements over different embodiments may be appropriately combined.

DESCRIPTION OF SYMBOLS 1 Cleaving quality measurement control apparatus 10, 20 Polarization high-speed image sensor 30 Three-dimensional phase difference image generation part 40 Cleavage control part 50 Storage part 60 Analytical value / experiment database 200 Scribing wheel 201 Glass

Claims (4)

In the cleaving quality measurement control device for controlling a predetermined cleaving means to cleave a hard and brittle material,
A polarization high-speed image sensor that measures the spatial distribution of the polarization characteristics of the hard and brittle material at high speed and in real time during the cleaving of the hard and brittle material;
A storage unit for storing a cleaving condition for the hard and brittle material and a database in which a spatial distribution of polarization characteristics of the hard and brittle material is related,
By referring to the database based on the spatial distribution of the polarization characteristics measured by the polarization high-speed image sensor, an optimum cleaving condition for the cleaving means is obtained, and the cleaving means is determined in real time using the cleaving condition obtained. A cleaving control unit controlled by
A cleaving quality measurement control device characterized by comprising: The hard and brittle material is glass;
The cleaving means is a scribing wheel that cleaves the glass,
The cleaving condition is the speed of the scribing wheel and the applied load of the scribing wheel to the glass,
The cleaving control unit controls the speed of the scribing wheel and the applied load of the scribing wheel on the glass in real time.
The cleaving quality measurement control apparatus according to claim 1. The polarization characteristic is a phase difference between orthogonal two-direction components of polarized light,
The spatial distribution of the polarization characteristics is a three-dimensional distribution of the phase difference.
The cleaving quality measurement control device according to claim 1 or 2. In the cleaving quality measurement control method for cleaving hard and brittle materials by controlling a predetermined cleaving means,
During the cleaving of the hard and brittle material, the polarization high-speed image sensor is used to measure the spatial distribution of the polarization characteristics of the hard and brittle material at high speed and in real time.
A database in which the cleaving condition for the hard and brittle material and the spatial distribution of the polarization characteristics of the hard and brittle material are referred to based on the spatial distribution of the polarization characteristics measured by the polarization high-speed image sensor, and Find the optimal cleaving condition for the cleaving means
Control the cleaving means in real time using the obtained cleaving condition,
A cleaving quality measurement control method characterized by the above.

Images