Wafer bath imaging

Abstract

An example method for monitoring a bath process includes processing a first wafer by immersing the first wafer in a bath solution, taking a video of the bath solution including the first wafer during a first time interval, analyzing the video based on light intensity captured in frames of the video, and determining a first metric of the bath solution during the first time interval based on analyzing the video.

Description

Technical Field

[0001] Cross - reference to Related Applications This application claims priority and the benefit of the filing date of U.S. Patent Application Publication No. 17 / 349,538, filed on June 16, 2021, which is hereby incorporated by reference in its entirety.

[0002] The present invention generally relates to wafer baths and, in particular, to wafer bath imaging.

Background Art

[0003] In semiconductor manufacturing, wafer baths using circulating fluids are commonly used to simultaneously perform substrate surface pretreatment processes such as surface cleaning, resist stripping, and thin - film etching on batches composed of multiple wafers.

[0004] Across all wafers in a lot, uniformity within and between wafers is important to provide narrow wiring for integrated circuit components such as transistors, capacitors, and resistors. Achieving uniformity within and between wafers in a bath is difficult for large - diameter wafers such as 12 - inch wafers. Therefore, it is important to further develop technologies for bath systems and methods of wafer processing in bath systems with improved process control.

Summary of the Invention

Means for Solving the Problems

[0005] According to one embodiment of the present invention, a method for monitoring a bath process includes processing a first wafer by immersing the first wafer in a bath solution, capturing a video of the bath solution containing the first wafer during a first time interval, analyzing the video based on the intensity of light captured in the video frames, and determining a first metric of the bath solution during the first time interval based on the analysis of the video.

[0006] According to one embodiment of the present invention, a method for detecting wafer bridging includes immersing a plurality of wafers in a bath solution, irradiating the surfaces of the plurality of wafers with a light source, capturing a first image of a first portion of the surfaces of the plurality of wafers, and determining, based on the first image, that wafer bridging has occurred between any one of the plurality of wafers.

[0007] For a more detailed understanding of the present invention and its advantages, refer hereto to the following description, which should be read in conjunction with the accompanying drawings. [Brief explanation of the drawing]

[0008] [Figure 1] This document shows a front view of a bath system according to one embodiment of this application. [Figure 2] A flowchart of a method for monitoring a batch process according to one embodiment is shown. [Figure 3A-3B] A schematic diagram of a pixel-by-pixel analysis for determining a first metric of a bath solution according to one embodiment is shown. Figure 3A shows an analysis of the intensity of light captured by the frame, and Figure 3B shows the measurement of the first metric based on the analysis of the video frame. [Figure 4] This diagram shows a pixel-by-pixel analysis for determining a first metric of the bath solution based on the intensity of light captured in each frame of the video. [Figure 5A-5B]A schematic representation of a zone analysis for determining a first metric of a bath solution according to one embodiment is shown. Figure 5A shows an analysis of the intensity of light captured by the frame, and Figure 5B shows the measurement of the first metric based on the analysis of the video frame. [Figure 6] This diagram shows a flow chart of zone analysis for determining the first metric of the bath solution based on the analysis of the light intensity captured in each frame of the video. [Figure 7A] A schematic diagram of a one-dimensional array analysis for determining a first metric of a bath solution based on the intensity of light captured by a frame is shown in one embodiment. [Figure 7B] An example of a vertical one-dimensional array analysis derived from pixel-by-pixel analysis is shown. [Figure 7C] An example of a vertical one-dimensional array analysis derived from zone analysis is shown. [Figure 7D] A schematic diagram of a one-dimensional array analysis for determining the first metric based on the analysis of video frames is provided. [Figure 8] This diagram shows a one-dimensional analysis flow chart for determining the first metric of the bath solution based on the analysis of the light intensity captured in each frame of the video. [Figure 9] A flowchart of a method for determining a first metric of a bath solution based on the number of pixels exceeding a threshold intensity, according to one embodiment, is shown. [Figure 10] A flowchart of a method for determining a first metric of a bath solution, according to one embodiment, is shown, in which the process parameters of the bath solution are intentionally changed. [Figure 11] A cross-sectional view of a bath system for symmetry analysis having two cameras, according to one embodiment, is shown. [Figure 12A-12C] This invention illustrates the detection of wafer bridging according to one embodiment. Figure 12A shows a top view of the bath after a plurality of wafers have been immersed in the bath solution, Figure 12B shows a top view of the bath after wafer bridging has occurred, and Figure 12C shows a process step for measuring a first analysis of the intensity of light captured by video frames. [Figure 12D] Figure 12D shows example data from the analysis. [Figure 13] A flowchart of a method for detecting wafer bridging according to one embodiment is shown. [Modes for carrying out the invention]

[0009] Differences in wafer bath operation can affect how wafers are processed. For example, changes in the bath process, such as changes in the flow or turbulence of process liquids / gases into the bath, can lead to defects in wafer processing. Therefore, it is important to monitor the wafer bath so that these changes that may occur during processing can be detected.

[0010] Embodiments of this disclosure disclose monitoring a bath process by analyzing frames of video recording (or time-lapse imaging) of the bath solution while processing wafers. The bath solution may be video recorded from various viewpoints, such as from above the bath, below the bath, from the side of the bath, or from an angle of any other object of interest. In various embodiments, the bath process may be monitored by analyzing frames of the video recording of the bath solution and measuring metrics of the bath solution based on the frames. In various embodiments, the metrics may be used to quantify the bath solution in order to detect changes or non-uniformity of the bath solution during or after wafer processing.

[0011] The techniques described herein can be used with a wide variety of processing tools that utilize wafer baths. For example, exemplary processing tools can be used in various processing steps such as cleaning and etching. It should be recognized that the processing tools shown herein are merely examples to which the monitoring techniques may be applied. Therefore, the techniques disclosed herein can be applied to other wafer bath systems and / or other processing tools. Furthermore, these wafer bath systems may be standalone tools or may be incorporated into larger systems.

[0012] FIG. 1 shows a front view of a bath system 100 of a processing tool according to an embodiment of the present application.

[0013] Referring to FIG. 1, a bath system 100 of a processing tool includes a bath 102 configured to process one or more wafers 104 in a bath solution 103. The bath 102 may include wafers 104 immersed in the bath solution 103 and held by a wafer holder 106. The wafer holder 106 may hold a batch of wafers 104 that can be processed simultaneously in the bath solution 103.

[0014] The bath 102 may include a first plurality of flow bars 108. In various embodiments, the bath 102 includes a first plurality of flow bars 108 and a second plurality of flow bars 110. The first plurality of flow bars 108 are configured to supply a gas such as nitrogen (N2) into the bath 102. The second plurality of flow bars 110 may be configured to supply a process chemical including a liquid or a gas such as an etching solution including hydrofluoric acid, phosphoric acid, nitric acid, hydrochloric acid, sulfuric acid, aqua regia, potassium hydroxide, tetramethylammonium hydroxide, ammonium hydroxide, hydrogen peroxide and / or a solvent. The solvent may include, for example, deionized water and / or an organic solvent such as acetone or alcohols together with other additives such as a surfactant for processing the wafer 104 in the bath 102.

[0015] The bath system 100 includes a camera 112, for example, a camera having an image sensor. The camera 112 may be any type of sensor known in the art to record images / video, such as a charge-coupled device (CCD) image sensor or a complementary metal-oxide-semiconductor (CMOS) image sensor. The camera 112 may be configured to operate in any visible spectrum, such as grayscale or RGB. The camera 112 may include certain filters that remove or allow certain wavelengths. The camera 112 may be configured to capture video of the bath solution 103 from a single field of view over a first interval during the processing of the wafer 104. In various embodiments, the camera 112 may be positioned in various locations so that the camera 112 can record video of the bath solution 103 from different fields of view. In one embodiment, the video may include a plurality of images captured continuously at a specific frequency or frame rate over a period of time. In another embodiment, the video may include a plurality of images captured separately at different times, such as a time-lapse. Camera 112 may be positioned above the bathtub 102, below the bathtub 102, on either side of the bathtub, or at any other position around the bathtub 102. Camera 112 may also be positioned at any angle to the bathtub 102. In one or more embodiments, multiple cameras may be positioned around the bathtub 102 to monitor different portions of the bath solution 103 from different viewpoints.

[0016] In certain embodiments, the bath system 100 may further include an audio recorder 113, such as an audio recorder 113 that includes a plurality of microphones and recorders. The audio recorder 113 is optional. The plurality of microphones may be spatially arranged around the bathtub 102 to detect spatial variations in the sound generated within the bathtub 102. In another example, the audio recorder 113 may include a plurality of hydrophones within the bathtub 102. The audio recorder 113 may be integrated with the camera 112 in a recording device configured to capture an image / video along with an audio signal. In one embodiment, the recording of the audio signal may be analyzed and used, along with the analysis of the image / video according to the method of the embodiment, for monitoring the bath process. The audio recorder may be a passive audio system or an active audio system. In a passive audio system, the audio recorder is configured to record the sound (i.e., the audio signal) generated in the bath system 100. Examples of sound include the sound of bubbles emerging from the first plurality of flow bars 108 and / or the second plurality of flow bars 110 and the sound of bubbles bursting on the surface of the bath solution 103. Instead, in an active audio system, sound waves may be transmitted from a sound wave generating device of the active audio system as a probing signal to detect objects (e.g., bubbles and wafers 104) in the bath solution 103. Then, the receiver of the active system may collect sound waves reflected by an object, e.g., an object located near the sound wave generating device, or sound waves attenuated by an object, e.g., an object located on the opposite side of the generating device in the bathtub 102. In these embodiments, the parameters of the probing signal may be selected such that the pressure wave from the sound wave does not induce unintended changes in the size and shape of the bubbles, such as cavitation. The sound waves captured by the audio recorder 113 may be analyzed spatially and / or temporally, for example, based on the intensity of the audio. Based on the analysis, an audio-based metric indicating the state of the bath solution 103 may be measured and combined with the video analysis of the method of the embodiment.

[0017] The light source 116 can irradiate at least a portion of the bath solution 103. In various embodiments, the entire bath solution 103 is irradiated by the light source 116. The light source 116 can be positioned anywhere around the bathtub 102. The light source 116 may be located on the opposite side of the bathtub 102 from the camera 112 (e.g., top vs. bottom) or on the same side of the bathtub 102. In one example, the light source 116 may be included in a first plurality of flow bars 108. In another example, reflective elements may be positioned in the first plurality of flow bars 108 to reflect light sent from the light source above the bath solution 103. The method of irradiating the bathtub 102 is not limited by this disclosure.

[0018] A controller 114 may be coupled to the bath system 100 (or a part thereof) for setting and controlling various process parameters of the bath solution 103. The controller 114 may be coupled to any or all components of the bath system 100 to receive information and control each of its components. For example, the controller 114 may be coupled to and control a first plurality of flow bars 108 and a second plurality of flow bars 110. The controller 114 may be configured to change process parameters of the bath solution, such as changing the flow rates of the first plurality of flow bars 108 and / or the second plurality of flow bars 110, changing the level and distribution of bubbles (e.g., N2 bubbles formed by the second plurality of flow bars 110), and changing the temperature of the bath solution 103. The controller 114 may include one or more processors (e.g., microprocessors, microcontrollers, central processing units, etc.), programmable logic devices (e.g., composite programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), etc.) and / or other programmable integrated circuits.

[0019] The controller 114 may be further coupled to the camera 112, which may instruct the camera 112 to record video of the bathtub 102 and to process the output generated by the camera 112. In one embodiment, the controller 114 may also be coupled to an audio recorder 113.

[0020] Monitoring the bath process with camera 112 may involve a wide range of techniques for analyzing and processing frames generated by camera 112 during recording of the bath solution 103. Monitoring the bath solution 103 can provide process feedback that would otherwise be unavailable, potentially leading to process improvement and optimization. Therefore, in various embodiments, temporal recording, including continuous video recording and time-lapse photography, is an efficient data acquisition method that can be performed on a wafer-by-wafer basis. In various embodiments, the temporal recording data can be analyzed to measure and / or control various variables, including the recipe of the bath solution 103, the uniformity of the bath solution 103, and the uniformity of turbulence.

[0021] In a particular embodiment shown in Figure 1, the bath system 100 may include a heating element 109, such as a heating rod, configured to introduce heat into the bath system 100. The bath can be a boiling bath (where the liquid is heated to form bubbles). Embodiments of the present invention are also applicable to boiling processes in which heat can be introduced by the heating element 109 or generated exothermally by the reaction of reactants in the bath solution 103. In such cases, bubbles may be generated in the bath solution 103 due to localized heating by the heating element 109 or an exothermic reaction.

[0022] The heating element 109 may be a heating rod and may be designed to locally heat the bath solution 103. Local boiling of the bath solution 103 causes bubbles to form around the heating element 109. For illustrative purposes, a heating rod is shown in this configuration, but other configurations and other types of heating elements 109 may also be used. The heating element 109 may be integrated with a first plurality of flow bars 108 or a second plurality of flow bars 110, thereby not necessarily being a separate component within the bath system 100.

[0023] The heat required to boil the bath solution 103 can also be generated by exothermic reactions between reactants within the bath solution 103. For example, the temperature of a piranha solution bath (sulfuric acid with hydrogen peroxide added) can be controlled by the rate at which hydrogen peroxide is added to the sulfuric acid. For example, the boiling point of a high-temperature phosphoric acid / water bath used for silicon nitride stripping can be controlled by maintaining a constant ratio of phosphoric acid to water in the bath solution 103.

[0024] In certain embodiments, a heating element 109 may be used instead of the first plurality of flow bars 108 and / or the second plurality of flow bars 110 that release the bubble-forming gas. Similarly, in certain embodiments, the first plurality of flow bars 108 and the second plurality of flow bars 110 may be used instead of the heating element 109. Further embodiments may omit the first plurality of flow bars 108 and / or the second plurality of flow bars 110 and the heating element 109, and the temperature in the bath 102, and therefore the bubbles, may be controlled solely by the flow rate of the reactants.

[0025] Figure 2 is a flowchart illustrating a method 200 for monitoring a bath process according to one embodiment.

[0026] As shown in block 202 and illustrated with reference to Figure 1, the first wafer is processed by immersing the first wafer in the bath solution 103. Although method 200 describes a single wafer, multiple wafers may be immersed in the bath solution 103.

[0027] As shown in block 204 and described with reference to Figure 1, a video of the bath solution 103 containing the first wafer is captured during a first time interval. The first time interval may be less than the total time the first wafer is processed in the bath solution 103. The first wafer may be immersed in the bath solution 103 before or after the first time interval. The video of the bath solution 103 may be a video at a predetermined frame rate such as 24 fps, 30 fps, or 60 fps, or it may be a slower time-lapse image sequence. The video of the bath solution 103 may be captured by illuminating part (or all) of the bath solution 103 and taking a video over the first time interval at a specific viewpoint using a camera 112. The video may contain multiple frames that are temporally separated throughout the entire first time interval, which may be used to monitor the bath process. The time to separate each frame depends on the frame rate of the camera 112.

[0028] After capturing a video of the bath solution 103, different analysis methods may be applied for video analysis (blocks 206-212c) and subsequent output data analysis (blocks 214-222b) according to various embodiments.

[0029] First, a pixel area for video analysis can be determined (block 206). Three analysis methods are described in this disclosure, though not limited to them: pixel-by-pixel analysis (block 208a and Figures 3A-4), zone analysis (block 208b and Figures 5A-6), and one-dimensional array analysis (block 208c and Figures 7A-8). In various embodiments, the intensity of light captured in each frame of the video can be analyzed to measure a first metric of the bath solution 103. In certain embodiments, optional video intensity adjustments can be performed before determining the first metric of the bath solution 103 (blocks 212a, 212b, 212c and Figures 9 and 10). Methods of the embodiments for monitoring can be applied to different temporal stages (block 214). The first metric can be measured dynamically at a first interval (block 216a) or after processing the first wafer (block 216b). The first metric may include raw light intensity, adjusted light intensity, and the number of bubbles derived from the adjusted light intensity using an intensity threshold. In various embodiments, the first metric alone may not be sufficient to define the range of the actual metric of interest, for example, the range of the actual metric of interest due to unintended process variations. Therefore, a second metric may be optionally measured (block 220). The second metric may improve the accuracy and validity of determining defects or faulty processes in monitoring by providing additional useful information about the bath solution 103. In various embodiments, the second metric may include the directional dependence of light intensity, which may also be useful in symmetry analysis (e.g., Figure 11) and wafer bridging detection (e.g., Figures 12A-13). Unless otherwise specified in this disclosure, the metrics may include the first metric, the second metric, or both. Furthermore, additional metrics may also be used.

[0030] The metrics can be used for real-time analysis / control and / or post-bath process analysis (blocks 222a and 222b). For example, in one embodiment, a metric based on light intensity may be used to detect heterogeneity in the bath solution 103. In another example, if the metric is based on the number of bubbles, the metric may be used to quantify parameters such as turbulence or bubble uniformity. Exemplary embodiments of methods for monitoring the bath process according to various embodiments will be discussed in further detail below.

[0031] In various embodiments, pixel-by-pixel analysis can be used to analyze the video based on the intensity of light captured by each frame of the video (for example, block 208a in Figure 2 and Figures 3A to 4).

[0032] Figures 3A and 3B are schematic diagrams of a pixel-by-pixel analysis for determining a first metric of a bath solution according to one embodiment. Figure 3A shows an analysis of the intensity of light captured by a frame, and Figure 3B shows the measurement of the first metric based on the analysis of video frames. Figure 4 is a flowchart of Method 400, which uses a pixel-by-pixel analysis to determine a first metric of a bath solution based on the intensity of light captured in each frame of a video.

[0033] Referring to Figure 3A, video frame 302 may include the bath solution 103 along with hardware of the bath system 100, such as adjacent parts of the bathtub 102. The area of ​​image / video captured within frame 302 may depend on the position, angle, and focus of camera 112. For example, since camera 112 is positioned above the bathtub 102, frame 302 may include the entire bath solution 103 and adjacent parts of the bathtub 102. Additional hardware for the bath system 100 positioned above the bathtub 102, such as a folding lid, may also be included in the field of view of camera 112 and therefore within frame 302. Figure 3A shows one frame 302 of the video, but the analysis performed on frame 302 may be applied to all frames of the video or a subset of frames of the video.

[0034] In various embodiments, each region of the frame being analyzed, for example, the process area 301, may be selected from within frame 302. The same region of the process area 301 may be selected in each of the frames being analyzed. For example, since the entire bath solution 103 is irradiated, the process area 301 in frame 302 may include the entire bath solution 103. The boundaries of the process area 301 may be aligned vertically and horizontally based on hardware detected in each of the frames. For example, the top or bottom of the bath 102 may be used for vertical alignment, and the left or right side of the bath 102 may be used for alignment of the process area 301 in frame 302. The process area 301 may extend to the entire bath solution 103 or to a portion of the bath solution 103. In some embodiments, if there is additional hardware that obstructs the view of the bath solution 103, the process area 301 may be formed around the obstruction. For example, if the opaque edges of a foldable lid that is transparent except for the ends and meets in the center of the bathtub 102 obstruct the view to the central portion of the bath solution 103, the process area 301 may include two sections (e.g., left and right or upper and lower) consisting of the visible portion of the bath solution 103 separated by the opaque edges of the lid.

[0035] Each frame of the video can be divided into an array of pixels 305 based on the sensor size of the camera 112 that captures the video (block 402 in Figure 4). For example, the process area of ​​frame 302 can be divided into an array of pixels 305. Each pixel 304 in the array of pixels 305 can be arranged in rows and columns across frame 302. The size and number of each pixel 304 can be based on the resolution and sensor size of camera 112. Each pixel 304 in Figure 3A is for illustrative purposes only, and the size or number of each pixel 304 is not proportional to the actual size. The number of rows in the array of pixels 305 may or may not be equal to the number of columns across frame 302.

[0036] After dividing each of the frames 302 into a pixel array 305, the light pixel intensity can be calculated for each of the pixel arrays 305 (block 404 in Figure 4). The light pixel intensity can be calculated for each pixel 304 within the process area 301.

[0037] Optionally, the calculated pixel intensity of light at each pixel 304 in frame 302 may be adjusted to remove noise by subtracting the background intensity of the light captured in frame 302 (e.g., block 210 in Figure 2). In one example, the background intensity may be the median of the light intensity captured at each pixel location 304 across all frames or a subset of frames in a given dataset. In this example, the median intensity at each pixel location 304 may then be subtracted from each corresponding location across all frames of interest. This adjustment process may be repeated for each of the array of pixels 305 from each frame captured in the video. In another example, the background intensity of the light may be obtained by using a frame before the start of bubbling / boiling or before the light source 116 is turned on. This step may be designed to identify the largest noise source, which is then removed from the output of camera 112.

[0038] After calculating the light pixel intensity for each of the pixel array 305, a first metric of the bath solution 103 during a first time interval can be measured based on the light pixel intensity for each of the pixel array 305 (block 406 in Figure 4).

[0039] Referring to Figure 3B, multiple frames 300 can be obtained from the video captured by camera 112 over a first time interval. Three frames are shown, namely the first frame 302a, the second frame 302b, and the third frame 302c, but this does not indicate the number of frames that can be obtained over the first time interval.

[0040] In various embodiments, a first metric can be measured by calculating the mean, median, standard deviation, sum, maximum, minimum, or range of light pixel intensity between each of the arrays of pixels 305 corresponding to the same portion of the area of ​​interest in the process area 301 in each of a plurality of frames 300 (e.g., 302a-c). For example, referring again to Figure 3B, the first metric can be calculated between pixel 304a in the first frame 302a, pixel 304b in the second frame 302b, and pixel 304c in the third frame 302c.

[0041] Since the first metric is measured between corresponding pixels across each of the multiple frames 300, the first frame 302a, the second frame 302b, and the third frame 302c can be combined into a single frame 302d. The single frame 302d may include a pixel array 305, each representing a value of the first metric calculated between each of the pixel arrays 305 corresponding to the same portion of the area of ​​interest in the process area 301 in each of the multiple frames 300. For example, pixel 304d in the single frame 302d may include the mean, median, standard deviation, sum, maximum, minimum, or range of the pixel intensity of light for pixels 304a to 304c.

[0042] Alternatively, the video captured over a first time interval can be analyzed using zone analysis by dividing each frame into multiple zones and measuring the zone intensity of each zone (e.g., block 208b in Figure 2 and Figures 5A-6). Zone intensity can be obtained by adding or averaging the pixel intensities from the pixels contained within the zone. Thus, a first metric of the bath solution 103 during the first time interval can be calculated based on the zone intensity of each zone.

[0043] Figures 5A and 5B schematically illustrate a zone analysis for determining a first metric of a bath solution according to one embodiment. Figure 5A shows an analysis of the intensity of light captured by a frame, and Figure 5B shows the measurement of the first metric based on the analysis of video frames. Figure 6 is a flowchart of method 600 using zone analysis to determine a first metric of a bath solution based on an analysis of the intensity of light captured in each frame of a video.

[0044] Figure 5A shows one frame of the video (frame 302), but the analysis performed on frame 302 can be applied to all frames of the video or a subset of all frames of the video.

[0045] Referring to Figures 5A and 6, each frame of video can be divided into multiple zones 506 (block 602 in Figure 6). For example, the process area 301 within frame 302 can be divided into multiple zones 506. In the exemplary embodiment shown in Figure 5A, zone 506 contains nine pixels (3x3). However, the size and number of each of the multiple zones 506 in Figure 5A are for illustrative purposes only and are not proportional to the actual size. Each of the multiple zones 506 may contain a subset of the array of pixels 305. Each of the multiple zones 506 may contain the same number of pixels. The multiple zones 506 may be arranged in rows and columns across the process area 301. The rows and columns of the multiple zones 506 may contain the same number of zones or different numbers of zones.

[0046] After dividing frame 302 into multiple zones 506, the pixel intensity of each light in the pixel array 305 can be measured. Optionally, in the same manner as discussed in Figure 3A, the calculated pixel intensity of the light in each pixel 304 can be adjusted based on the background intensity of the light captured in frame 302.

[0047] Alternatively, each of the calculated pixel intensities of light can be adjusted by subtracting the background intensity of the light captured in each of the multiple zones 506 (e.g., block 210 in Figure 2). For example, the background intensity of the light captured in zone 506 can be measured by calculating the median of the pixel intensities of the light captured by each of the pixels located in zone 506 in multiple frames preceding frame 302. This is because zone 506 in the preceding frames should have captured only the background before bubbles appear in zone 506 in frame 302.

[0048] After calculating the light pixel intensity for each of the pixel arrays 304, the light zone intensity for each of the multiple zones 506 can be calculated (block 604 in Figure 6). The light zone intensity for each of the multiple zones 506 can be measured by calculating the mean, median, standard deviation, sum, maximum, minimum, or range of the light pixel intensity between each of the pixel arrays located in each of the multiple zones 506. This process can be repeated for each frame captured over a first time interval.

[0049] After calculating the light zone intensity for each of the multiple zones 506, a first metric of the bath solution 103 during a first time interval can be measured based on the light zone intensity in each of the multiple zones 506 (block 606 in Figure 6).

[0050] Referring to Figure 5B, similar to Figure 3B, multiple frames 300 can be obtained from video captured by camera 112 over a first time interval.

[0051] A first metric of the bath solution 103 during a first time interval can be measured based on the zone intensity of light for each of the multiple zones 506. The first metric can be measured by calculating the mean, median, standard deviation, sum, maximum, minimum, or range of the zone intensity of light between each of the multiple zones 506 that cover the same portion of the area of ​​interest of the process area 301 in each of the frames 302a to 302c. For example, the first metric can be calculated between zone 506a of the first frame 302a, zone 506b of the second frame 302b, and zone 506c of the third frame 302c.

[0052] Since the first metric of the bath solution 103 is measured between each of the multiple zones 506a to 506c across each of the frames 302a to 302c, each of the frames 302a to 302c can be combined into a single frame 302d. A single frame 302d may contain multiple zones, each representing a value of the first metric of the bath solution 103 calculated across each of the frames 302a to 302c. For example, zone 506d of a single frame 302d may contain the mean, median, standard deviation, sum, maximum, minimum, or range of zone intensities for zones 506a to 506c.

[0053] Alternatively, the video captured over a first time interval can be analyzed using one-dimensional array analysis, by dividing each frame into a one-dimensional array of regions and measuring the regional intensity of each region (e.g., block 208b and Figures 7A-8). Thus, a first metric of the bath solution 103 during the first time interval can be calculated based on the regional intensity of each region.

[0054] Figure 7A schematically shows a one-dimensional array analysis for determining a first metric of the bath solution 103 based on the intensity of light captured by a frame, according to one embodiment. Similar to the previous embodiment, Figure 7A shows one frame of video (frame 302), but the analysis performed on frame 302 can be applied to all frames or a subset of frames of the video.

[0055] Referring to Figures 7A and 8, each frame of the first video can be divided into a one-dimensional array of regions 701 (block 802 in Figure 8). Region 702 represents one region of the one-dimensional array of regions 701. Although the one-dimensional array of regions 701 is shown as a vertical array in Figure 7A, the one-dimensional array of regions 701 can extend in any direction across the process area 301 (e.g., vertical or horizontal). Each of the one-dimensional array of regions 701 may contain a column (or row) of an array of pixels 305.

[0056] After dividing frame 302 into a one-dimensional array 701 of regions, the pixel intensity of each light in the array of pixels 305 can be determined. Optionally, in the same manner as discussed in Figure 3A, the calculated pixel intensity of each light in the array of pixels 305 can be adjusted based on the background intensity of the light captured in frame 302. In certain embodiments, each of the calculated pixel intensities of light can be adjusted by subtracting the background intensity of the light captured in each of the one-dimensional arrays 701 of regions (e.g., block 210 in Figure 2). The background pixel intensity of the light can be calculated in the same manner as described in the previous embodiments. Thus, each of the arrays of pixels in each of the one-dimensional arrays 701 of regions can be adjusted in the same manner as, for example, each of the arrays of pixels in each of a plurality of zones 506.

[0057] After calculating the light pixel intensity for each of the pixel arrays, the light region intensity for each of the one-dimensional arrays 701 of the region can be calculated (block 804 in Figure 8). The light region intensity for each of the one-dimensional arrays 701 of the region can be measured by calculating the mean, median, standard deviation, sum, maximum, minimum, or range of light pixel intensities between each of the arrays of pixels located in each of the one-dimensional arrays 702 of the region. This process can be repeated for each frame captured over a first time interval.

[0058] The one-dimensional array of regions 701 is shown as an array of pixel rows (1 × 9), but the size of each region in the one-dimensional array of regions 701 is not limited and can be any size. In certain embodiments, the one-dimensional array of regions 701 may be defined as the sum of specific zones in a row or column. Thus, the zone intensity of light in each zone within each of the one-dimensional array of regions 701 can be used to calculate the region intensity of light in each of the one-dimensional array of regions 701.

[0059] Therefore, in one embodiment, one-dimensional array analysis may be applied as a secondary analysis step in conjunction with pixel-by-pixel analysis and zone analysis. Alternatively, in one embodiment, one-dimensional array analysis may be applied as a primary analysis step instead of pixel-by-pixel analysis and zone analysis.

[0060] After calculating the light region intensity for each of the one-dimensional arrays 702 of regions, a first metric of the bath solution 103 during a first time interval can be measured based on the light region intensity in each of the one-dimensional arrays 702 of regions (block 806).

[0061] Figure 7B shows an example of a vertical one-dimensional array analysis derived from a pixel-by-pixel analysis. Graph 705 shows the first metric as a function of column number. First, the pixel intensity was measured for each pixel, and the first metric was calculated by vertically adding the pixel intensities for each region of the one-dimensional array 701 of regions. Graph 705 shows how the first metric (e.g., process gas / liquid volume) changes across the locations of the bath solution 103 corresponding to each of the one-dimensional array 701 of regions. In one embodiment, graph 705 may be compared to a graph of a target process (e.g., blocks 222a and 222b in Figure 2) to detect defects in the processing of the first wafer or to provide feedback to the controller for changing process parameters of the bath solution. In another embodiment, graph 705 may be used to determine defects in the processing of the first wafer based on a symmetry analysis between columns equidistant from the center of graph 705 (e.g., Figure 11).

[0062] Figure 7C shows an example of a vertical one-dimensional array analysis derived from a zone analysis. Graph 709 shows how the first metric changes for each of the one-dimensional arrays 701 of the region. Similar to graph 705 in Figure 7B, graph 709 shows how the first metric (e.g., process gas / liquid volume) changes across the locations of the bath solution 103 corresponding to each of the one-dimensional arrays 701 of the region. Graph 709 can be interpreted and used in the analysis in the same way as discussed above.

[0063] Figure 7D schematically illustrates a one-dimensional array analysis for determining a first metric based on the analysis of video frames. Referring to Figure 7D, multiple frames 300 may be acquired from video captured by camera 112 over a first time interval. The first metric of the bath solution 103 during the first time interval may be measured based on the region intensity of light for each of the multiple regions of a one-dimensional array 701.

[0064] As in other embodiments, the first metric may be measured by calculating the mean, median, standard deviation, sum, maximum, minimum, or range of the optical region intensity between each of the one-dimensional arrays 701 of regions covering the same portion of the area of ​​interest of the process area 301 in each of the frames 302a to 302c. For example, the first metric may be calculated between region 702a of the first frame 302a, region 702b of the second frame 302b, and region 702c of the third frame 302c.

[0065] Since the first metric of the bath solution 103 is measured between each of the one-dimensional arrays 702a to 702c of regions spanning each of frames 302a to 302c, each of frames 302a to 302c can be combined into a single frame 302d. A single frame 302d may contain multiple one-dimensional arrays 702 of regions, each representing a value of the first metric of the bath solution 103 calculated across each of frames 302a to 302c. For example, a region 702d in a single frame 302d may contain the mean, median, standard deviation, sum, maximum, minimum, or range of region intensities between the one-dimensional arrays 702a to 702c of regions.

[0066] In various embodiments, light intensity adjustment can be performed at an optional rate (e.g., blocks 210, 212b, 212c and Figures 9 and 10).

[0067] In certain embodiments, if the video analysis includes grouping of multiple pixels within a video frame (e.g., zone analysis and one-dimensional array analysis), the first metric may be measured based on the number of pixels exceeding an intensity threshold in the zone / array. In various embodiments, the number of pixels exceeding the threshold may represent, for example, the number of bubbles in a zone or one-dimensional array region. If the first metric is measured based on the number of bubbles, the first metric may be used to track the bubble level and distribution of the bath solution 103 over time, which can be used to quantify changes in turbulence or process liquid / gas flow (e.g., chemical state) of the bath solution 103.

[0068] Figure 9 shows a flow chart of a method 900 for determining a first metric of a bath solution based on the number of pixels exceeding a threshold intensity, according to one embodiment. Each pixel 304 exceeding the threshold intensity may be used to indicate a bubble. For example, each bubble may correspond to an N2 bubble generated by a first plurality of flow bars 108 (see Figure 1).

[0069] As shown in block 902, a pixel intensity threshold can be set. This pixel intensity threshold can be set based on various conditions such as the light source 116, or it can be set based on metrics previously applied to the dataset.

[0070] Next, as shown in block 904 and explained with reference to Figure 5A, pixels 304 having an intensity above a threshold can be measured in each frame. In the example of Figure 5A, each of the arrays 305 of pixels in each of the multiple zones 506 can be considered with respect to the pixel intensity threshold criterion.

[0071] Next, as shown in block 906, the falsely detected pixels 304 are removed. After removing the falsely detected pixels 304, the number of pixels 304 in each of the multiple zones 506 in each frame (e.g., the number of bubbles in each zone) can be measured. In one embodiment, falsely detected pixels can be identified based on the size of the detected bubbles. For example, if the intensity is high in only one pixel and all the surrounding pixels are below the intensity threshold, the high-intensity pixel in the center is likely to be a hot pixel and not a bubble, as a bubble would likely be visible in at least several pixels. Therefore, the local neighborhood of the pixel can be analyzed to determine the falsely detected pixels 304, which can then be removed. In addition, based on the local neighborhood of the pixel, a false detection can be determined based on the size of the detected bubble being too large. Analysis of the local neighborhood of the pixel can also eliminate possible data duplication.

[0072] Next, as shown in block 908 and explained with reference to Figure 5B, a first metric of the bath solution 103 may be measured based on the number of pixels 304 above an intensity threshold. The first metric of the bath solution 103 may be measured by calculating the mean, median, standard deviation, sum, maximum, minimum, or range of the number of pixels 304 above an intensity threshold (e.g., number of bubbles) between corresponding zones. For example, the mean, median, standard deviation, sum, maximum, minimum, or range may be calculated between zones 506a to 506c.

[0073] In some embodiments, intentional changes to the bath solution 103, which are part of the bath recipe such as induced boiling, or changes in the process gas / liquid flow, can alter the intensity of light captured in some of the frames captured during the first time interval. Generally, some information about these intentional changes is known. For example, the effect may be greater near the outlet of a fluid or gas or a heating element. Therefore, intentionally induced changes in light intensity can be subtracted before determining the first metric.

[0074] Figure 10 is a flowchart of Method 1000 for determining a first metric of a bath solution 103 when the process parameters of the bath solution 103 are intentionally changed, according to one embodiment of the present application. Although Method 1000 is described in relation to an induced boiling process, changes to other process parameters of the bath solution 103 may be utilized.

[0075] As shown in block 1002, boiling flames (e.g., flames captured during induced boiling) and non-boiling flames (e.g., flames captured before or after induced boiling) can be determined. Boiling and non-boiling flames can be determined based on the duration within a first time interval in which boiling is induced. Boiling can be induced in the bath solution 103 by the use of a heating element 109, such as a heating rod, which can introduce an exothermic reaction due to the flow of reactants or localized heating.

[0076] Next, as shown in block 1004, the contrast of each frame captured in the video can be adjusted. The contrast of each frame can be adjusted by amplifying the range of pixel intensity of the light captured in each frame. The contrast of each frame can be adjusted using image processing techniques known in the art, such as directional filters and Laplacian filters. Advantageously, since the pixel intensity of the light in each frame is subtracted in a subsequent step, the difference in pixel intensity between consecutive frames can be reliably detected by amplifying the range of pixel intensity of the light captured in each frame.

[0077] Next, as shown in block 1006, the contrast-adjusted induced non-boiling frame can be subtracted from the boiling frame. Advantageously, the difference between the light captured in the non-boiling frame is subtracted, and a first metric of the bath solution can be used to determine unintended changes to the bath solution 103.

[0078] Next, as shown in blocks 1008-1012, the threshold intensity analysis process steps described in blocks 902-908 of Figure 9 may be repeated to determine a first metric based on the number of pixels greater than the threshold intensity (e.g., number of bubbles). In other embodiments, after subtracting contrast-adjusted frames, the first metric may be calculated based on light intensity, for example, using method 600 or 800.

[0079] In various embodiments, a first metric of the bath solution 103 may be measured dynamically while the first wafer is being processed (block 216a in Figure 2). The dynamically updated first metric may be further used during the processing of the first wafer for defect detection and / or control of process variables (e.g., block 222a in Figure 2).

[0080] For example, defects in the processing of the first wafer may be detected by identifying changes in the first metric at specific locations in the bath solution 103 (e.g., changes in the first metric between pixels 304a-304c in Figure 3B, zones 502a-502c in Figure 5B, or the one-dimensional array 702a-702c of the region in Figure 7D). Alternatively, non-uniformity across the entire bath solution may be measured based on a single generated frame 302d. For example, the bath solution 103 may be analyzed based on the differences in the first metric between pixel 304d and pixel 308 (Figure 3B), between zone 506d and zone 520 (Figure 5B), and between the one-dimensional array 702d of the region and region 720 (Figure 7D). As will be understood by those skilled in the art, slight changes in the bath solution 103 may have only a negligible effect on the processing of the first wafer or may be due to noise in the intensity of light captured in frame 302. Therefore, in one or more embodiments, a defect may be detected if the change in the first metric is greater than or equal to a threshold defect detection value. The threshold defect detection value may need to be determined first, for example, by collecting the first metric and comparing it with the actual defects in wafer processing. Based on the identified change in the first metric of the bath solution, the cause of the defect may be determined. Causes of defects include, but are not limited to, non-uniformity of the bath solution 103, non-uniformity in the hardware of the bath solution such as non-uniform flow rates or composition of the process liquid / gas supplied through the first plurality of flow bars 108 or the second plurality of flow bars 110, or too little / too much or non-uniform turbulence.

[0081] In another example, process parameters of the bath solution 103 may be changed over a second time interval while a first wafer is being processed, based on identified changes in a dynamically measured first metric of the bath solution 103. For example, process parameters of the bath solution 103 that may be changed include, but are not limited to, the mixture of process gas / liquid supplied from the first plurality of flow bars 108 and / or the second plurality of flow bars 110, the bubbling rate of the second plurality of flow bars 110, changes in the flow rates of the first plurality of flow bars 108 and / or the second plurality of flow bars 110, and changes in the temperature of the bath solution 103.

[0082] In other embodiments, a first metric of the bath solution 103 may be measured after processing a first wafer (block 216a in Figure 2). The first metric may then be used for defect detection, feedback of process variables for processing a second wafer, and / or correlation of process parameters of the bath solution 103 (e.g., block 222b in Figure 2).

[0083] If a first metric of the bath solution is measured after processing the first wafer, the first metric can be used to analyze specific locations in the bath solution 103 over time (e.g., pixels 304a-304c in Figure 3B, zones 506a-506c in Figure 5B, or the corresponding one-dimensional arrays 702a-702c in the region in Figure 7D), or to analyze a single frame 302d generated from each frame of a video. For example, the bath solution 103 can be analyzed based on the difference in the first metric between pixel 304d and pixel 308 (Figure 3B), between zone 506d and zone 520 (Figure 5B), and between region 702d and region 720 (Figure 7D).

[0084] In various embodiments, a second metric of the bath solution 103 may be measured to provide an additional dimension for the analysis of the bath solution 103 (block 220 in Figure 2). The second metric may be measured independently of the first metric and then used for output data analysis. In certain embodiments, the second metric may be derived from the first metric.

[0085] The second metric can advantageously improve the detection of defects or faulty processes in monitoring the bath solution 103. In various embodiments, the second metric may include the directional dependence (directivity) of light intensity. Any predetermined pattern, orientation, or structural features of the wafer in the bath and / or bath solution 103 can be used to measure the second metric. The second metric can be measured and applied to analysis in the same manner as described above for the first metric and will not be described in repeated detail. For example, the second metric can be measured horizontally, vertically, or in any direction by pixel-by-pixel analysis (block 208a and Figures 3A-4), zone analysis (block 208b and Figures 5A-6), or one-dimensional array analysis (block 208c and Figures 7A-8).

[0086] According to one embodiment, defect detection based on the monitoring method of the embodiment is described as follows. In the first example, again referring to Figures 3B, 5B and 7D, defects may be detected after processing a first wafer based on the metric uniformity of the bath solution 103 across a single frame 302d. The uniformity across a single frame 302d can be measured by symmetry analysis.

[0087] In one embodiment, as shown in Figure 3B, a vertical symmetry line 306 may be formed across a single frame 302d, and metrics measured for a one-dimensional array of pixels, zones, or regions equidistant from the vertical symmetry line 306 may be compared. For example, pairs of pixels equidistant from the vertical symmetry line 306 (e.g., pixels 304d and 309 in Figure 3B) may be compared. If the metrics differ significantly between one-dimensional arrays of pixels, zones, or regions, defects may be detected.

[0088] In another embodiment, as shown in Figure 3B, a horizontal symmetry line 312 may be formed across a single frame 302d, and defects may be determined based on a comparison between one-dimensional arrays of pixels, zones, or regions equidistant from the horizontal symmetry line 312. For example, pairs of pixels equidistant from the horizontal symmetry line 312 (e.g., pixels 304d and 310 in Figure 3B) may be compared.

[0089] Alternatively, defects in the processing of the first wafer can be detected by comparing the metrics of the bath solution 103 from the processing of the first wafer and the processing of the second wafer. In other words, the change in the metrics over time in the processing of the first wafer can be compared with the change in the metrics over time in the processing of the second wafer, or a single frame 302d can be compared with a further single frame generated from the processing of the second wafer. Therefore, if a significant difference is found between the metrics, a defect in the processing of the first or second wafer can be determined.

[0090] In yet another embodiment, the bath solution metric may also be used for tool-to-tool matching (block 222b in Figure 2). The metric may be compared between different bath solutions from different processing tools (e.g., a first bath solution and a second bath solution). Thus, based on the difference between the metrics, the process parameters of the bath solution 103 may be modified to match the metrics for subsequent wafer processing. For example, the process parameters of the bath solution 103 that may be modified include, but are not limited to, the mixture of process gas / liquid supplied from the first plurality of flow bars 108 and / or the second plurality of flow bars 110, the bubbling rate of the second plurality of flow bars 110, changes in the flow rates of the first plurality of flow bars 108 and / or the second plurality of flow bars 110, and changes in the temperature of the bath solution 103.

[0091] In one or more embodiments, changes in metric between corresponding pixels 304 (Figure 3A), zones 506 (Figure 5A), or one-dimensional arrays of regions 701 (Figure 7A) formed over consecutive frames (e.g., pixels 304a-304c in Figure 3B, zones 506a-506c in Figure 5B, or one-dimensional arrays of regions 702a-702c in Figure 7D) are fed back for subsequent processing. In other words, based on the changes in metric detected after processing the first wafer, process parameters of the bath solution 103 may be modified to improve the uniformity of the metric of the bath solution 103 for processing the second wafer. Process parameters that may be modified include, but are not limited to, those discussed above.

[0092] If the metric is measured after the first wafer has been processed, the metric may be associated with process metrics such as surface roughness, thickness of the film etched, uniformity of film thickness across the wafer, variation of surface roughness across the wafer, and other parameters.

[0093] In various embodiments, heterogeneity of the bath solution 103 can be detected using a symmetry analysis between two different viewpoints of the bath solution 103 over a first time interval.

[0094] Figure 11 shows a cross-sectional view of a bath system having two cameras according to one embodiment.

[0095] Referring to Figure 11, two cameras may be positioned on opposite sides of the bath solution 103, for example, with the first camera 1104 on the left and the second camera 1106 on the right, as shown in the figure. Figure 11 also shows a symmetry line 1102 that can be used for symmetry analysis. The symmetry line 1102 may be defined as a horizontal, vertical, or any other angle across the bath solution 103.

[0096] After a first wafer (or more wafers) is immersed in the bath solution 103, a first camera 1104 may capture video of the left side of the bath solution 103 during a first time interval. Similarly, a second camera 1106 may capture video of the right side of the bath solution 103 over the first time interval. Thus, a new set of frames may be generated from the video from the second camera 1106, which can then be analyzed for metrics in the same manner as described above. Therefore, if the difference between metrics in corresponding frames from two videos equidistant from the symmetry line 1102 exceeds a certain threshold, a defect may be detected in the wafer processing. Symmetry analysis can be performed on any pair of two different process areas of interest. For example, in one embodiment, a left-to-right analysis may be performed as shown in Figure 11, but other embodiments may use other symmetries in the process chamber. Furthermore, more than one camera may be used in the symmetry analysis performed, but this is not mandatory. For example, one camera may provide one set of video frames, which can then be divided into zones equidistant from the symmetry line 1102 for symmetry analysis.

[0097] In various embodiments of this application, bridging between wafers 104 in a bath solution can be detected using video captured by a camera 112 over a first time interval (for example, blocks 222a and 222b in Figure 2).

[0098] Figures 12A to 12D are schematic diagrams of a process for detecting wafer bridging according to one embodiment of the present application. Figure 12A shows a top view of the bath after multiple wafers have been immersed in the bath solution, Figure 12B shows a top view of the bath after wafer bridging has occurred, Figure 12C shows a process step for determining a first analysis of the intensity of light captured by video frames, and Figure 12D shows an example of the analysis. Figure 13 is a flowchart of a method 1300 for detecting wafer bridging according to one embodiment.

[0099] Referring to Figures 12A and 13, multiple wafers 104 are immersed in a bath solution 103 (block 1302 in Figure 13). The multiple wafers 104 are separated by a distance 1202 defined by a substrate holder or boat that holds the wafers 104. Before bridging, the distance 1202 between adjacent wafers 104 is approximately the same. After the wafers enter the bath to be processed, the surface of the wafers may be illuminated with a light source (block 1304 in Figure 13), and video of the wafers may be captured (block 1306). However, in some cases, bridging may occur before the wafers 104 are placed in the bath 102. Wafer bridging or fusion can cause misprocessing of the bridged wafer because the surface of the bridged portion of the wafer may not be exposed to the bath solution and may not be processed like the other wafers.

[0100] As shown in Figure 12B, when wafers are processed, they may fuse or bridge with other wafers in the process area 301 for various reasons. Wafer bridging or fusion results in a variation in the distance 1202 separating the wafers (Figure 12B). Therefore, wafer bridging can be detected by analyzing and detecting the variation in the distance 1202 from the initial value.

[0101] Analyzing video of wafer 104, as shown in Figure 13 as Method 1300, can help determine when and where bridging may occur. Video of a specific location in the wafer is first recorded and stored (block 1306 in Figure 13). The time to perform the bridging detection analysis is determined (block 1308 in Figure 13). The area of ​​the image to perform the bridging detection analysis (e.g., 1210 in Figure 12C) is also determined (block 1308 in Figure 13).

[0102] In certain embodiments, before detecting wafers, the data may optionally be realigned to aid in any rotation present in the data and to better detect wafers (block 1310). Then, an average value is calculated across this area (block 1312), from which wafers and spaces may be detected (block 1314). This detection allows for the calculation of the distances between wafers and spaces, along with the number of wafers that may be detected (block 1316). These can be used to determine whether or not wafer bridging has occurred (block 1318).

[0103] Figure 12C schematically shows an example of the method for detecting the wafer bridging described above, according to one embodiment. The wafer area to be analyzed may be selected as a region of the process area 301 and may include a one-dimensional array 1208 of regions. For example, the array 1208 of regions may include a row of arrays of pixels perpendicular to the wafer 104, as shown by line 1210.

[0104] Figure 12D shows an example of analysis. Graph 1220 shows the average pixel values ​​across a one-dimensional array of regions. In Graph 1220, points with high pixel values ​​represent wafer locations, and points with low pixel values ​​represent the absence of wafers (i.e., space between wafers). These pixel value points can be analyzed to calculate, for example, the distance between wafers (Graph 1230) or the distance between spaces (Graph 1240). In Graphs 1230 and 1240, outliers are identified at the beginning and end of the one-dimensional array of regions, as indicated by square boxes. A threshold can be set to determine whether a given variation indicates wafer bridging.

[0105] In another embodiment, the number of wafers may be counted based on the peaks (or valleys) of pixels observed in Figure 12D. If the number of peaks is less than the number of wafers actually loaded into the bathtub (10⁴), wafer bridging may be determined to have occurred.

[0106] Herein, exemplary embodiments of the present invention are summarized. Other embodiments may be understood from the entirety of this specification and the claims filed herein. Reference figures are provided below for illustrative purposes only, and various examples may be carried out differently and should not be construed as being limited to these examples alone.

[0107] Example 1. A method for monitoring a bath process, comprising: processing a first wafer by immersing the first wafer in a bath solution (103); taking a video of the bath solution (103) containing the first wafer during a first time interval; analyzing the video based on the intensity of light captured in the video frames (302); and determining a first metric of the bath solution (103) during the first time interval based on the analysis of the video.

[0108] Example 2. The method according to Example 1, wherein analyzing a video includes selecting a region of the video to be analyzed and analyzing the same region in subsequent frames of the video following that frame (302).

[0109] Example 3. The method according to one of Example 1 or 2, wherein analyzing a video comprises dividing each frame (302) of the video being analyzed into an array of pixels based on the sensor size of a camera (112) that captures the video, and calculating the pixel intensity of light for each of the arrays of pixels, and determining a first metric of the bath solution (103) in a first time interval comprises determining a first metric for each of the arrays of pixels based on the pixel intensity of light.

[0110] Example 4. The method according to any one of Examples 1 to 3, wherein analyzing a video includes dividing each frame (302) of the video being analyzed into multiple zones and calculating the zone intensity of light in each of the multiple zones, and determining a first metric of the bath solution (103) during a first time interval includes determining the first metric based on the zone intensity of light in each of the multiple zones.

[0111] Example 5. The method according to any one of Examples 1 to 4, wherein analyzing a video comprises dividing each frame (302) of the video being analyzed into a one-dimensional array of regions, and calculating the region intensity of light for each of the arrays of regions, and determining a first metric of the bath solution (103) during a first time interval comprises determining a first metric for each of the arrays of regions based on the region intensity of light.

[0112] Example 6. The method according to any one of Examples 1-5, wherein analyzing a video involves subtracting the background light intensity obtained for each frame (302) of the video being analyzed from the light intensity captured in that frame (302).

[0113] Example 7. The first metric is measured dynamically during the processing of the first wafer, as described in any one of Examples 1-6.

[0114] Example 8. The first metric is measured after processing the first wafer, according to any one of Examples 1-7.

[0115] Example 9. The first metric of the bath solution (103) is based on the number of pixels exceeding the intensity threshold, as described in any one of Examples 1-8.

[0116] Example 10. The method according to any one of Examples 1 to 9, wherein the first metric of the bath solution (103) includes the number of bubbles in the bath solution (103).

[0117] Example 11. The method according to any one of Examples 1 to 10, further comprising measuring a second metric of the bath solution (103) during a first time interval based on the analysis of a video, wherein the second metric is based on the directionality of the intensity of light captured in each frame (302) of the video being analyzed.

[0118] Example 12. The method according to any one of Examples 1 to 11, further comprising detecting defects in processing a first wafer.

[0119] Example 13. The method according to any one of Examples 1 to 12, wherein detecting a defect involves identifying that a first metric of the bath solution (103) changes during a first time interval.

[0120] Example 14. The method according to any one of Examples 1 to 13, wherein detecting a defect includes measuring a second metric of the bath solution (103) during processing of a second wafer and determining that the first metric is different from the second metric.

[0121] Example 15. The method according to any one of Examples 1 to 14, further comprising changing the process of the bath solution (103) during a second time interval in order to process a first wafer in the bath solution (103).

[0122] Example 16. The method according to any one of Examples 1 to 15, further comprising comparing a first metric of bath solution (103) with a second metric of another bath solution (103) of a different processing tool, and adjusting the process parameters of bath solution (103) to match the first metric of bath solution (103) with the second metric of the other bath solution (103).

[0123] Example 17. The method according to any one of Examples 1 to 16, further comprising changing the process of the bath solution (103) during a second time interval in order to process a second wafer.

[0124] Example 18. The method according to any one of Examples 1 to 17, wherein the analysis of the video includes measuring the uniformity of the bath solution (103) during a first time interval based on the intensity of light captured in each frame (302) of the video being analyzed.

[0125] Example 19. The method according to any one of Examples 1 to 18, further comprising associating a first metric of the bath solution (103) with a process metric of a processing tool that holds the bath solution (103).

[0126] Example 20. The method according to any one of Examples 1 to 19, further comprising recording audio of the bath solution (103) during a first time interval, analyzing the audio based on its intensity, and measuring an audio-based metric of the bath solution (103) during the first time interval based on the analysis of the audio.

[0127] Example 21. A method for detecting wafer bridging, comprising: immersing a plurality of wafers (104) in a bath solution (103); irradiating the surfaces of the plurality of wafers (104) with a light source (116); taking a first image of a first portion of the surfaces of the plurality of wafers (104); and determining, based on the first image, that wafer bridging has occurred between any one of the plurality of wafers (104).

[0128] Example 22. The method according to Example 21, further comprising taking a second image of a second portion of the surface of a plurality of wafers (104), and determining that wafer bridging has occurred, comprising comparing the first image with the second image.

[0129] Example 23. The method according to one of Examples 21 or 22, wherein the first image and the second image are different frames of a video.

[0130] Example 24. The method according to any one of Examples 21-23, further comprising analyzing a first image to measure the distance between adjacent wafers (104) and determining whether wafer bridging has occurred, which includes determining whether any of the distances between adjacent wafers (104) is below a threshold.

[0131] Example 25. The method of any one of Examples 21-24, further comprising analyzing a first image to count the number of wafers in a plurality of wafers (104), and determining whether wafer bridging has occurred, which includes determining that the count is less than the actual number of wafers (104) in the plurality of wafers (104).

[0132] While the present invention has been described with reference to exemplary embodiments, this specification is not intended to be constrained. Those skilled in the art will be able to see various modifications and combinations of the exemplary embodiments, as well as other embodiments of the invention, by reference to this specification. Therefore, the appended claims are intended to encompass all such modifications or embodiments.

Claims

1. A method for monitoring the bath process, This method is The first wafer is processed by immersing it in a bath solution, The steps include taking a video of the bath solution containing the first wafer during a first time interval, The steps include: analyzing the video based on the intensity of light captured in the frames of the video; Based on the step of analyzing the video, the steps include determining a first metric of the bath solution during the first time interval, It has, Furthermore, the process includes a step of determining a second metric of the bath solution during the first time interval based on the step of analyzing the video, The second metric is a method based on the directionality of the light intensity captured in each frame of the video being analyzed.

2. The step of analyzing the aforementioned video is: The steps include selecting the region of the video to be analyzed, The steps include analyzing the same region in another frame of the video following the aforementioned frame, The method according to claim 1, comprising:

3. The step of analyzing the aforementioned video is: The steps include dividing each frame of the video to be analyzed into an array of pixels based on the sensor size of the camera that captures the video, For each of the aforementioned array of pixels, the step of calculating the pixel intensity of light, It has, The step of determining the first metric of the bath solution during the first time interval is: The method according to claim 1, further comprising the step of determining the first metric for each of the array of pixels based on the pixel intensity of the light.

4. The step of analyzing the aforementioned video is: The steps include dividing each frame of the video to be analyzed into multiple zones, A step of calculating the zone intensity of light in each of the aforementioned multiple zones, It has, The method according to claim 1, wherein the step of measuring the first metric of the bath solution during the first time interval comprises the step of determining the first metric based on the zone intensity of the light in each of the plurality of zones.

5. The step of analyzing the aforementioned video is: The steps include dividing each frame of the video to be analyzed into a one-dimensional array of regions, For each of the array of regions, the steps include calculating the region intensity of light, It has, The method according to claim 1, wherein the step of determining the first metric of the bath solution during the first time interval comprises the step of determining the first metric for each of the array of regions based on the region intensity of the light.

6. The method according to claim 1, wherein the step of analyzing the video comprises subtracting the background intensity of the light obtained for each frame from the intensity of the light captured for each frame of the video being analyzed.

7. The method according to claim 1, wherein the first metric is dynamically determined during the processing of the first wafer.

8. The method according to claim 1, wherein the first metric is determined after processing the first wafer.

9. The method according to claim 1, wherein the first metric of the bath solution is based on the number of pixels exceeding an intensity threshold.

10. The method according to claim 1, wherein the first metric of the bath solution includes the number of bubbles in the bath solution.

11. The method according to claim 1, further comprising the step of detecting defects when processing the first wafer.

12. The method according to claim 11, wherein the step of detecting the defect includes the step of identifying that the first metric of the bath solution changes during the first time interval.

13. The step of detecting the defect is: The steps include determining a second metric of the bath solution during the processing of the second wafer, A step of determining that the first metric is different from the second metric, The method according to claim 11, comprising:

14. The method according to claim 1, further comprising the step of changing the process of the bath solution for processing the first wafer in the bath solution during a second time interval.

15. moreover, The steps include comparing the first metric of the bath solution with the second metric of another bath solution of a different processing tool, The steps include adjusting the process parameters of the bath solution to match the first metric of the bath solution with the second metric of the other bath solution, The method according to claim 1, comprising:

16. The method according to claim 1, further comprising the step of changing the process of the bath solution for processing the second wafer during a second time interval.

17. The method according to claim 1, wherein the step of analyzing the video comprises determining the uniformity of the bath solution during a first time interval based on the intensity of the light captured in each frame of the video being analyzed.

18. The method according to claim 1, further comprising the step of relating the first metric of the bath solution to the processing metric of a processing tool that holds the bath solution.

19. moreover, The steps include capturing the audio of the bath solution during the first time interval, A step of analyzing the audio based on the intensity of the audio, Based on the step of analyzing the audio, the steps include determining the audio-based metric of the bath solution during the first time interval, The method according to claim 1, comprising:

20. A method for detecting wafer bridging, This method is The steps include immersing multiple wafers in a bath solution, The steps include irradiating the sides of the plurality of wafers with a light source, The steps include taking a first image of the first portion of the side surface of the plurality of wafers, A step of determining, based on the first image, that wafer bridging has occurred between any one of the plurality of wafers, A method having

21. Furthermore, the process includes the step of taking a second image of the second portion of the side surface of the plurality of wafers, The method according to claim 20, wherein the step of determining that wafer bridging has occurred comprises the step of comparing the first image with the second image.

22. The method according to claim 21, wherein the first image and the second image are different video frames.

23. Furthermore, the process includes the step of analyzing the first image and determining the distance between adjacent wafers of the plurality of wafers, The method according to claim 20, wherein the step of determining whether wafer bridging has occurred includes the step of determining whether any of the distances between adjacent wafers of the plurality of wafers is less than a threshold.

24. Furthermore, the process includes the step of analyzing the first image and counting the number of wafers in the plurality of wafers, The method according to claim 20, wherein the step of determining whether wafer bridging has occurred includes the step of determining that the count is less than the actual number of wafers in the plurality of wafers.

Images