TW201447279A - Automated optical inspection device of wafer and a method of inspecting the uniformity of wafer - Google Patents

Abstract

The present invention discloses an automated optical inspection device of wafer and a method of inspecting the uniformity of wafer. More specifically, the present invention discloses a device and method to utilize the data generated in the defect inspecting process to simulate the process of inspection of the uniformity.

Description

Automatic wafer optical inspection device and method for detecting wafer surface uniformity

The present invention relates to an automated optical inspection apparatus and method of detecting wafer surface uniformity. More specifically, the present invention is an apparatus and method for simulating uniformity detection using data generated in a defect detection program.

In order to maintain the quality of existing semiconductor wafers, the quality of the wafer itself is often inspected by optical inspection. Automated Optical Inspection (hereinafter referred to as AOI) is an example. AOI is a general term for high-speed and high-precision optical image detection systems. It uses machine vision as a test standard technology, which is a disadvantage of improving the traditional use of optical instruments for human detection. After the wafer is finished with a processing such as grinding, it is necessary to confirm whether the wafer surface is damaged. Manufacturers often use the defect detection device in AOI to optically detect the surface of the wafer. More specifically, the defect detecting device can capture the surface image of the wafer by the photosensitive element, and then use the algorithm to analyze the defective image taken from the surface of the wafer to identify the damaged portion. Although the basic design has been found in the prior art, the present invention will not be described again.

In addition, in order to perform uniformity detection on the wafer after the crystal yellow process or the etching process, the industry has also developed the use of the area of the wafer to reflect the light intensity to evaluate the uniformity of the wafer surface. Uniformity detecting device. More specifically, the principle of operation is to use a light source to illuminate a certain area of the target wafer, and then use the photosensitive element to measure the light intensity of the area reflected to the photosensitive element. In the process, the photosensitive element detects only the light intensity without imaging a specific image of the area of the wafer.

In general, after performing the semiconductor wafer processing procedure, the manufacturer uses the uniformity detecting device and the defect detecting device to detect the surface of the wafer to ensure that the defects and uniformity of the surface are in compliance with the standard.

In applying the above-described uniformity detecting device and defect detecting device for detecting, the applicant unexpectedly found a simple, reliable, extremely low peripheral cost and improved efficiency of the conventional detecting system.

The applicant's invention will be described below. Briefly, the applicant has proposed a wafer optical detecting device that simulates the uniformity detecting device using data obtained by a conventional defect detecting device, so that the operator can use only one. The defect detecting device achieves the effects of defect detection and uniformity detection to achieve the purpose of omitting the uniformity detecting device. More specifically, the present invention utilizes data from a defect detecting device to perform calculations to simulate a uniformity detecting device. Therefore, during the defect detection and uniformity detection process, the detected wafer will not need to be moved. Thereby, in addition to omitting the hardware construction of the uniformity detecting device, the present invention can reduce the chance of the wafer being damaged or contaminated by movement and the manpower required for moving the wafer.

Furthermore, the applicant found that in the process of simulation, it is necessary to use the thumbnail algorithm to filter out the bright and dark point defects unrelated to the uniformity detection, and then simulate a close to the true uniformity variation result, otherwise the simulation result It will be difficult to present accurately. Accordingly, the present invention omits the uniformity detecting device After the hardware, the present invention still has all the functions of the original uniformity detecting device.

Further, the specific design of the present invention will be further described in detail below.

1‧‧‧ wafer optical inspection device

10‧‧‧Light source

20‧‧‧Loading station

30‧‧‧Optical imaging module

40‧‧‧ storage module

50‧‧‧ Computing Module

60‧‧‧Interface module

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a schematic diagram showing a wafer optical inspection apparatus of the present invention in a specific embodiment.

2A to 2C are schematic diagrams respectively showing a thumbnail algorithm of the wafer optical detecting device of the present invention in a specific embodiment.

As described above, the present invention proposes a wafer optical detecting device (hereinafter referred to as a detecting device or a device of the present invention). Briefly, the wafer optical inspection apparatus of the present invention uses an algorithm to enable the apparatus originally used for detecting defects to have the capability of uniformity detection without requiring substantial adjustment of the hardware equipment.

Before describing the respective constituent elements of the apparatus of the present invention, it should be emphasized that the wafer optical detecting apparatus mentioned in the present invention is improved by the defect detecting apparatus in the conventional automatic optical detecting apparatus, The design of the defect detecting device is well known, and the applicant will only explain the difference between the present invention and the prior art, and the overlapping portions will be selectively omitted.

Referring to FIG. 1, FIG. 1 is a functional block diagram of a wafer optical detecting device of the present invention in a specific embodiment. As shown in the figure, in the present embodiment, the wafer optical inspection apparatus 1 of the present invention generally includes a light source 10, a carrier 20, an optical imaging module 30, a storage module 40, an operation module 50, and An interface module 60. Before describing the operation mode of the present invention, the following description will be first made for each component element and module.

The light source 10 of the present invention means that a light is supplied to a wafer to be used on the wafer. Active light source 10 for optical detection. In this example, the light source 10 is at least one external coaxial light source 10 integrated with a light emitting diode. However, the invention is not limited thereto, and the light source 10 may also be an annular light source, an oblique light source or other light-emitting device that can be used for defect detection, depending on its design. In addition, in this example, the light source 10 also includes a plurality of active light sources 10 arranged in a one-dimensional or two-dimensional matrix to simultaneously output light to portions of the wafer.

In addition, in this example, the optical imaging module 30 of the present invention is disposed above the carrier 20 and includes a matrix of semiconductor photosensitive elements (array) and an imaging lens. The imaging lens corresponds to the photosensitive matrix. In this example, the optical imaging module 30 is a CCD lens having a CCD sensor module matrix and a corresponding optical lens. In addition, in this example, the optical imaging module 30 is disposed above the coaxial light source 10 outside the light source 10 to receive reflected light reflected from the wafer via the external coaxial light source 10. However, the optical imaging module 30 of the present invention is not limited to the other end of the light source 10, and has a corresponding angle with the light source 10 for receiving light.

Additionally, the carrier 20 is provided with a carrier surface for carrying the wafer. In this example, the carrying platform 20 is a movable loading platform 20 connected to a power module and having a fixing component, which is controlled by the controller to perform XYZ three-axis movement to drive the wafer and adjust the same. The relative position of the photosensitive elements. However, the carrying platform 20 of the present invention is not limited to being movable, and it may also be a fixed carrier, and the position of the photosensitive member may be moved instead.

The interface module 60 is a hardware, user interface software or a combination thereof for the user to input or output data to the user. In this example, the interface module 60 of the present invention refers to a touch display screen. However, the invention is not limited thereto. The interface refers to a control panel having a physical button or a component such as a keyboard, a mouse, a touch display, and the like.

The storage module 40 of the present invention refers to a device for storing electronic data, such as memory. Any device that stores data, such as a card, a memory, a hard disk, or even a disk matrix of a server connected via a network, is considered to be the storage module 40 of the present invention. When the storage module 40 is located outside the device, the device of the present invention selectively adds a network card or its corresponding component for data exchange.

In this example, the storage module 40 is a hard disk and stores a plurality of program materials. Each program data is treated as a program or a code having a corresponding function. In this example, the storage module 40 includes at least a first predetermined program data and a second predetermined program data.

The first predetermined program data is used to generate a first analysis data and a second analysis data, and the second predetermined program data is used to generate a third analysis data. Briefly, the first analysis data and the second analysis data correspond to a defect detection result, and the third analysis data corresponds to a uniformity detection result. The contents of the first predetermined program data and the second predetermined program data details will be described later.

In another aspect, the computing module 50 of the present invention generally refers to a hardware, software, or a combination thereof that is capable of reacting to programs, instructions, or operations on data to produce results. In this example, the computing module 50 refers to a central processing unit (CPU) and its body. In the present embodiment, the computing module 50 is coupled to the optical imaging module 30, the storage module 40, the interface module 60, the light source 10, and other components for controlling each component or module. Achieve a corresponding purpose. When the device of the present invention is used, the user has to control the computing module 50 by the operating interface to enable the computing module 50 to execute the first predetermined program data and the second predetermined program data. The first analysis data, the second analysis data, and the third analysis data.

In order to explain the principle of operation of the detail of the present invention, the following will be exemplified by the actual application of the present invention. First, in the application, the user firstly operates through the interface module 60. In the meantime, the computing module 50 can read and execute the first predetermined program data and the second predetermined program data stored in the storage module 40.

Specifically, the first predetermined program data is regarded as a corresponding program of the conventional defect detecting device, that is, a part of the wafer is photographed and analyzed by the light source 10 to determine damage or defects on the surface of the wafer. In order to judge the quality of each part of the wafer. The first analysis data and the second analysis data are analysis results of defects around the wafer surface.

In order to achieve the foregoing effect, the first predetermined program data sequentially includes the sub-program S11 to the sub-program S13, and each sub-program has a corresponding code or a combination thereof.

The subroutine S11 refers to obtaining a first image data corresponding to a first sub-area of the wafer, which refers to illuminating a portion of a wafer with a light source 10 to obtain the Part of the image of the wafer surface. In application, the subroutine S11 further includes a subroutine S111 to a subroutine S113, and the subroutine is a subroutine S111 for controlling the light source 10 to output a first detection ray to a first sub-region of the wafer. The sub-program S112 is configured to control the optical imaging module 30 to extract a first reflected light reflected from the first sub-area of the wafer to generate a first image data, where the first reflected light corresponds to the first A detection light is generated; the subroutine S113 is to store the first image data to the storage module 40.

The subroutine S12 refers to obtaining a second image data corresponding to another portion of the wafer, which is to irradiate another portion of the wafer with a light source 10 to obtain the Part of the image of the wafer surface. On the other hand, the subroutine S12 further includes a subroutine S121 to a subroutine S123. The subroutine S121 is configured to control the light source 10 to output a second detection light to a second sub-area of the wafer; the subroutine S122 controls the optical imaging module 30 to extract the second sub-area from the wafer. Reflecting one of the second reflected rays to generate a second image data, the second reflected light being relatively The second detection light should be used; the sub-program S123 is to store the second image data to the storage module 40. Thereby, the first analysis data and the second analysis data, that is, the defect analysis result of the two regions on the wafer can be obtained.

The sub-program S13 is configured to separately analyze the first image data and the second image data to generate a first analysis data and a second analysis data, where the first analysis data corresponds to the first image data. The second analysis data corresponds to the second image data, which refers to separately analyzing the images of the respective portions to determine whether they have defects, and the analysis result is the first analysis data and the second analysis. The data will then be stored in storage module 40 for subsequent use. The first image data and the second image data are data corresponding to the image on the surface of the wafer.

Then, after the first predetermined program material is completed, the second predetermined program data will be executed. The second predetermined program data is detected relative to the prior art. In order to detect the uniformity of the wafer by using the defect detecting device, the present invention innovatively proposes a method and a program for simulating the uniformity detecting using the data generated in the defect detecting program, and the second predetermined program data will be described below. The contents are explained.

In this example, the second predetermined program data includes a subroutine S21 and a subroutine S23, and the subroutine S21 obtains the first image data and the second image data from the storage module 40; and the subroutine S22 The third image data is generated by the first image data and the second image data. More specifically, the third image data is also integrated with the image image and the brightness correction operation program. The image connection picture refers to a process of connecting and integrating a plurality of images, and the brightness correction operation is a program for correcting and adjusting the brightness of the image; the subroutine S23 is generated according to the third image data. The third analysis data, that is, the first image data and the second image data generated by the foregoing storage module 40 are used for simulation analysis to generate the third analysis data.

More specifically, in the present case, the aforementioned analog analysis uses the integrated color features of multiple images for statistical analysis. In short, the intensity of the reflected light across the surface of the wafer can be calculated by counting the wavelength distribution of each block of the wafer or the difference in color depth, by comparing the intensity of the reflected light of each part, that is, The uniformity distribution of the surface of the wafer is obtained. More specifically, in addition to the use of thumbnail processing, the uniformity distribution is more selectively used to perform numerical processing on the thumbnail image using a grayscale or gradation algorithm to obtain uniformity relative to the surface to be measured. The third analysis of sex.

However, in the actual application, the applicant found that if the original first image data and the second image data were directly counted, it would be impossible to accurately use the data of the difference in wafer surface uniformity.

After the research, it is found that the aforementioned error value is generated in the image data obtained by the first image data and the second image data in the defect detection, which has many bright and dark spots or defects which are not related to the uniformity of the yellow light process or the etching process, so that the data Interfered with statistics. Therefore, it is necessary to use image processing to filter out, in order to obtain a near-true uniformity variation result.

Accordingly, in order to overcome this difference, the Applicant has developed a method for analyzing the uniformity of the brightness by using a moderate adjustment or reduction of the resolution of the image to filter the bright and dark spots or defects, which is referred to as a thumbnail algorithm.

Considering the variety of existing thumbnail algorithms, only a preferred example will be described below. Please refer to FIG. 2 , which is a schematic diagram of a matrix of image data of the wafer of the present invention. Suppose there is a target wafer surface whose original resolution is Δx and Δy. The data matrix S after scanning can be expressed by the following formula: S _{i , j} =[ X _{i , j} , y _{i , j} , G _{i , j} ]

Where i is the pixel coding in the x direction and j is the pixel coding in the y direction.

In addition, G _{i , j} is the reflected light intensity value measured by the position of X _{i , j} and Y _{i , j} after the thumbnail. After the original image is subjected to the thumbnail operation, the resolution is reduced to Δx' and Δy', and the processed data matrix S' can be expressed by the following formula: _{S'p , q} = [ x ' _{p , q} , y ' _{p , q} , G ' _{p , q} ]

Where p is the pixel coding in the x direction after the thumbnail, and q is the pixel coding in the y direction after the thumbnail. G ' _{p , q} is the average reflected light intensity value calculated from x ' _{p , q} and y ' _{p , q} position after the thumbnail, so G ' _{p , q} , x ' _{p , q} and y ' _{p , q} The representative formula is as follows: x ' _{p , q} = p × △ x; y ' _{p , q} = q × Δ y;

In addition, the representative expressions of m and n are as follows:

Based on this, the average value of the pattern uniformity is obtained ( ) and the standard deviation (σ _{G '} ) is calculated by G ' _{p , q to} obtain the following formula:

By using the thumbnail algorithm, the bright and dark point defects which are not related to the uniformity of the yellow light process or the etching process can be filtered out, and the result of the near-true uniformity variation is obtained. The resolution of the thumbnails can be adjusted by measuring reproducibility requirements or by customer shipping specifications. For the effect part, please refer to the figure depicted in Figure 3. As shown by the black line in the figure, the uniformity calculation is performed using the original image of the wafer, that is, the first image data and the second image data. The winner. It can be seen from the figure that in addition to the intermediate region, it still has a plurality of peaks representing bright spot defects and dark spot defects. The gray line and the white line in the figure are respectively corresponding to the intensity of the reflected light of the second-order thumbnail algorithm, which is visible, except that the middle area has become smoother. In addition, the peak value has also been filtered, thereby filtering out bright and dark point defects that are not related to the uniformity of the yellow light process or the etching process, so as to obtain an effect close to the true uniformity variation result. In addition, the resolution of the thumbnails can be adjusted by measuring reproducibility requirements or by wafer level requirements.

Furthermore, the microstructure of the yellow light process or the etching process is usually a micron or submicron size structure, and the reflectance of the structure varies correspondingly with the wavelength of the light source. Therefore, the light source of the present invention may be a single wavelength light source, A white light source or multiple sets of single-wavelength light sources with switchable wavelengths, and the source wavelength is selected as needed to obtain a suitable sensitivity. Even through the beam splitter group and the filter, corresponding to two or more sets of photosensitive elements, multiple sets of wavelength detection images can be obtained at one time, taking into account various defect and uniformity inspection requirements.

In addition, the light source may be a UV light source plus a fluorescent conversion element, and the light source of different wavelengths may be obtained by replacement of the fluorescent conversion element, and the sensitivity and accuracy are dynamically enhanced corresponding to different pattern structures or defect types. More specifically, the light source has a UV LED, and in order to meet different light requirements, the user has to apply the UV LED surface to the phosphor layer to have an optical wavelength corresponding to the wavelength of the light to have another wavelength. The optical components are replaced to adjust the wavelength and characteristics of the output light, and the aforementioned UV LED refers to a bare crystal or a packaged wafer, which is not limited by the present invention.

In summary, the present invention differs from the prior art in that the present invention proposes an apparatus and method for detecting the uniformity of a wafer using a defect detecting device. More specifically, the present invention utilizes data generated in a defect detection program to simulate a method and program for uniformity detection so that conventional defect detecting devices can further detect wafer uniformity.

It is to be understood that the present description is only one of the many example methods of the present invention, and in the actual use of the present invention, similar or equivalent to the method and apparatus described in the present specification can be used. Any method or means for it. Furthermore, one or more of the numbers mentioned in the specification include the number itself.

It should be understood that the present disclosure discloses certain methods and processes for performing the disclosed functions, and is not limited to the order described in the specification, unless the specification is explicitly excluded or the procedures and steps are necessarily causally related. Otherwise, the previous end of the execution time is free to adjust according to the requirements of the user. Further, since the properties of the respective elements of the present invention are considered to be similar to each other, the descriptions and reference numerals between the respective elements apply to each other. It should be noted that the components, modules, devices, components and other components mentioned in this specification are not limited to hardware that is actually independent of each other. It can also be individually or integrated with software, firmware or The hardware is presented in a way that is stated first.

The features and spirit of the present invention will be more apparent from the detailed description of the preferred embodiments. On the contrary, the intention is to cover various modifications and equivalents within the scope of the invention as claimed. Therefore, the scope of the patented scope of the invention should be construed in the broadest

1‧‧‧ wafer optical inspection device

10‧‧‧Light source

20‧‧‧Loading station

30‧‧‧Optical imaging module

40‧‧‧ storage module

50‧‧‧ Computing Module

60‧‧‧Interface module

Claims (11)

An automatic wafer optical detecting device for performing optical detection on a plurality of sub-areas of a wafer, the wafer optical detecting device comprising: a light source; a carrying platform provided with a carrying surface for carrying the wafer; An optical imaging module is disposed above the carrying platform; a storage module stores: a first predetermined program data, and the following subroutine includes: acquiring a first image data, the first image data Corresponding to a first sub-area of the wafer; obtaining a second image data corresponding to a second sub-area of the wafer; analyzing according to the first image data to generate Having a first analysis data; and performing analysis according to the second image data to generate a second analysis data; and a second predetermined program data, comprising: the following sub-programs: reading the first image data and the a second image data; integrating the first image data and the second image data to generate a third analysis data; and an operation module, and the optical imaging module, the The memory module and the light source are coupled to each other; an interface module is coupled to the computing module and has an operation interface; wherein, in application, the user controls the operation module by using the operation interface And causing the computing module to execute the first predetermined program data and the second predetermined program data to obtain the first analysis data, the second analysis data, and the third analysis data, the first analysis data and the second analysis The data is corresponding to the defects of the first sub-area and the second sub-area of the wafer, and the third analysis data is corresponding to the wafer Information on the uniformity of the first sub-region and the second sub-region. The wafer optical inspection apparatus of claim 1, wherein the subroutine for acquiring a first image data further comprises the following subroutine: controlling the light source to the first subregion of the wafer Outputting a first detection light; controlling the optical imaging module to extract a first reflected light reflected from the first sub-area of the wafer to generate a corresponding first image data, the first reflected light system Corresponding to the first detection light; and storing the first image data to the storage module. The wafer optical inspection device of claim 2, wherein the subroutine for acquiring a second image data further comprises: controlling the light source to output the second sub-region of the wafer Having a second detection light; controlling the optical imaging module to extract a second reflected light reflected from the second sub-area of the wafer to generate a second image data, wherein the second reflected light corresponds to the Second detecting light; and storing the second image data to the storage module. The wafer optical inspection apparatus of claim 3, wherein the step of integrating the first image data and the second image data to generate a third analysis data comprises the following procedure: from the storage module Acquiring the first image data and the second image data; and integrating the first image data and the second image data to generate a third image data by using a image connection or brightness correction operation; and according to the third image The data is generated to have the third analysis data. The wafer optical inspection apparatus of claim 4, wherein the step of generating a third analysis data based on the third image data comprises the following procedure: The thumbnail image is subjected to a thumbnail image processing by using a thumbnail algorithm; and the thumbnail image is numerically processed by using a grayscale or color scale algorithm to obtain the third analysis data. The wafer optical inspection device of claim 1, wherein the optical imaging module comprises: a matrix of semiconductor photosensitive elements for receiving the first reflected light to generate a corresponding first image And receiving the second reflected light to generate a corresponding second image data; and an imaging lens, the second reflected light is concentrated on the surface of the semiconductor photosensitive element. The wafer optical inspection device of claim 1, wherein the light source is an external coaxial light source. The wafer optical inspection apparatus of claim 1, wherein the light source comprises a single wavelength light source, a white light source or a plurality of sets of single wavelength light sources of switchable wavelengths. The wafer optical detecting device according to claim 1, comprising a UV light source and an alternative fluorescent conversion element, wherein the fluorescent conversion element is selected from a plurality of fluorescent conversion elements, the plurality of The fluorescent conversion elements respectively have a wavelength. The optical imaging module according to claim 1 may include a beam splitter group and a filter, and corresponding to two or more sets of photosensitive elements to simultaneously obtain multiple sets of wavelength detection images. A method for detecting uniformity of a wafer surface includes the steps of: acquiring a first image data corresponding to an image of a first sub-area of the wafer; and obtaining a second image data, The second image data corresponds to an image of a second sub-area of the wafer; The first image data and the second image data are integrated to generate a third analysis data, the third analysis data being relative to the uniformity of the first sub-region and the second sub-region of the wafer.