Plasma stability evaluation method and plasma processing apparatus

Abstract

To accurately evaluate plasma stability.SOLUTION: Capturing a moving image of plasma generated inside a processing container included in a processing apparatus, acquiring a reference frame image from the captured moving image, setting an evaluation threshold based on a histogram indicating a distribution of pixel values when a plurality of pixels constituting the reference frame image are gray-scaled, setting an evaluation target area based on the evaluation threshold, gray-scaling each frame image constituting the captured moving image, and applying the evaluation target area to each gray-scaled frame image; An average value of pixel values of all pixels inside an evaluation target region is calculated for each frame image constituting a captured moving image, time-frequency analysis is performed on time-series data of the average value of the pixel values of all the pixels inside the evaluation target region calculated from each frame image to decompose the time-series data into a plurality of frequency components, and a high-frequency component is extracted from the plurality of decomposed frequency components.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present disclosure relates to a plasma stability evaluation method and a plasma processing apparatus. [Background technology]

[0002] In a plasma processing apparatus, a technique for detecting fluctuations in plasma light emission is known in order to perform stable plasma processing. For example, a technique is known in which plasma generated inside a vacuum processing chamber is imaged through an observation window, and the imaged video data is analyzed to determine the degree of fluctuations in plasma light emission (see, for example, Patent Document 1).

[0003] When a high-frequency voltage is applied, the electric field generated inside the vacuum processing chamber can cause the plasma distribution to become uneven, resulting in localized areas of strong electric field. In these areas of strong electric field, the plasma light also becomes locally strong, forming bright spots, which can be considered as an indicator of fluctuations in the plasma light emission.

[0004] Therefore, in the technology described in Patent Document 1, in order to evaluate fluctuations in local plasma light emission, first, each frame image of the captured video data is binarized to calculate the number of bright spots present in the image, and then the calculated numbers of bright spots are summed up to calculate the number of bright spots occurring per unit time, and this number of bright spots occurring per unit time is used as an index of fluctuations in plasma light emission. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2009-64610 Summary of the Invention [Problem to be solved by the invention]

[0006] The techniques disclosed herein provide accurate plasma stability assessment. [Means for solving the problem]

[0007] One aspect of the technology disclosed herein is a plasma stability evaluation method for evaluating the stability of plasma generated inside a processing vessel provided in a plasma processing apparatus, the method comprising the steps of: capturing a video of the generated plasma; acquiring a reference frame image from the captured video; setting an evaluation threshold based on a histogram showing a distribution of pixel values ​​when a plurality of pixels constituting the reference frame image are grayscaled; and setting an evaluation target area based on the evaluation threshold; grayscaling each frame image constituting the captured video and applying the evaluation target area to each of the grayscaled frame images; calculating an average value of pixel values ​​of all pixels inside the evaluation target area for each frame image constituting the captured video; creating time series data of the average value of pixel values ​​of all pixels inside the evaluation target area calculated from each of the frame images; performing time-frequency analysis on the created time series data to decompose the created time series data into a plurality of frequency components; and decomposing the decomposed frequency components into low-frequency components consisting of the frequency components on the low-frequency side and high-frequency components consisting of the frequency components on the high-frequency side based on a predetermined criterion, and extracting the high-frequency components. [Effects of the Invention]

[0008] According to the technology of the present disclosure, accurate plasma stability evaluation can be performed. [Brief explanation of the drawings]

[0009] [Figure 1] 1 is a cross-sectional view schematically showing a configuration of a plasma processing apparatus to which a plasma stability evaluation method according to an embodiment of the technology disclosed herein is applied; [Figure 2] 10 is a flowchart showing an evaluation index acquisition process executed by a control unit of the plasma processing apparatus. [Figure 3]10 is a graph for explaining timing corresponding to a reference frame image. [Figure 4] 10 is a histogram of pixel values ​​of all pixels when the reference frame image is converted into a grayscale image. [Figure 5] FIG. 10 is a diagram for explaining a method for setting an evaluation threshold value; [Figure 6] 10 is a graph showing time-series data of average pixel values ​​in an evaluation target region. [Figure 7] 10 is a graph showing time-series data of average pixel values ​​when the evaluation threshold is changed. [Figure 8] 10 is a graph showing high-frequency components and low-frequency components decomposed from time-series data of average pixel values. [Figure 9] 10 is a graph showing time series data of average pixel values ​​at each level obtained by applying a one-dimensional discrete wavelet transform to time series data of average pixel values ​​created using an evaluation threshold value. [Figure 10] 10 is a graph showing time series data of average pixel values ​​at each level obtained by applying a one-dimensional discrete wavelet transform to time series data of average pixel values ​​created using an evaluation threshold value. DETAILED DESCRIPTION OF THE INVENTION

[0010] As described above, the technology described in Patent Document 1 binarizes each frame image and calculates the bright spots present in the image. However, because the binarization calculation result is significantly affected by the threshold value, unless an appropriate threshold value is set, it is not possible to calculate localized strong plasma light emission as a bright spot. In this case, even if localized strong plasma light emission occurs, it may not be determined that fluctuation in the plasma light emission has occurred, and as a result, it is not possible to accurately evaluate the fluctuation in the plasma light emission.

[0011] In contrast, the technology disclosed herein sets a threshold value used to evaluate fluctuations in plasma light emission such that when the pixels constituting the reference frame image are grayscaled, the shape of the area formed by pixels whose pixel values ​​are equal to or greater than the threshold value approximates the shape of the plasma light emission area.

[0012] An embodiment of the technology according to the present disclosure will now be described with reference to the drawings, in which: Fig. 1 is a cross-sectional view schematically showing the configuration of a plasma processing apparatus to which a plasma stability evaluation method according to an embodiment is applied.

[0013] 1 is an inductively coupled plasma processing apparatus. The plasma processing apparatus 10 performs a film formation process to form a protective film or the like on a rectangular substrate, such as a glass substrate G (hereinafter referred to as "substrate G") for an FPD (Flat Panel Display), using plasma generated from a processing gas.

[0014] The plasma processing apparatus 10 includes a rectangular cylindrical processing chamber 11 made of a conductive material, and the processing chamber 11 accommodates a substrate G. The upper portion of the processing chamber 11 is airtightly sealed by a window member 12. Inside the processing chamber 11, a mounting table 13 on which the substrate G is placed is disposed at the lower portion, and the mounting table 13 faces the window member 12. In the processing chamber 11, a processing space U is formed between the mounting table 13 and the window member 12. In the processing space U, plasma is generated from a processing gas, as will be described later.

[0015] An electrostatic chuck (not shown) is provided on the top surface of the mounting table 13. The substrate G placed on the mounting table 13 is attracted and held to the mounting table 13 by the electrostatic chuck. A temperature control mechanism such as a chiller and a heat transfer gas supply mechanism (neither of which are shown) are provided inside the mounting table 13 to control the temperature of the placed substrate G. The mounting table 13 is installed on the bottom surface of the processing chamber 11 via an insulating frame 14.

[0016] A metal frame 15 is provided at the upper end of the side wall of the processing vessel 11, and a side wall portion 16 is installed on the upper surface of the metal frame 15. The side wall portion 16 supports a top plate 17, and the top plate 17 covers the window member 12 from above. A sealing member 18 such as an O-ring is provided between the side wall of the processing vessel 11 and the metal frame 15 to keep the processing space U airtight.

[0017] One side wall of the processing vessel 11 is provided with a loading / unloading port 19 for loading / unloading the substrate G into / from the processing space U, and a gate valve 20 for opening / closing the loading / unloading port 19. A monitoring window 21 is provided on the other side wall of the processing vessel 11, and a camera 22 is provided outside the processing vessel 11 to capture color images of the interior of the processing vessel 11 through the monitoring window 21. The camera 22 transmits the captured image of the interior of the processing vessel 11 to a controller 37 (described later). In this embodiment, the image captured by the camera 22 is not a still image but a video, and this video is a video of the plasma generated in the processing space U. The number of cameras 22 included in the plasma processing apparatus 10 is not limited to one, and may be multiple. In this case, the videos captured by the cameras 22 are used to perform an evaluation index acquisition process (described later).

[0018] The window member 12 is made of a conductor, for example, metal, has a rectangular shape, and is divided into a plurality of segments 23, with a partition member 24 made of an insulator disposed between each of the adjacent segments 23. The partition members 24 separate the adjacent segments 23 and electrically insulate them from one another.

[0019] In the plasma processing apparatus 10, the space surrounded by the window member 12, the sidewall 16, and the top plate 17 constitutes an antenna chamber 25. In the antenna chamber 25, an inductive coupling antenna 26 is arranged so as to face the substrate G placed on the mounting table 13 via the window member 12. The inductive coupling antenna 26 is arranged across the areas facing each of the segments 23 as a whole, and is formed in a spiral shape so as to wrap around in the circumferential direction of the window member 12, but the shape of the inductive coupling antenna 26 is not limited to a spiral shape.

[0020] Each divided piece 23 has a number of gas holes 27 formed therein, which open toward the processing space U. A gas supply pipe 28 is connected to each divided piece 23, and the gas supply pipe 28 is connected to a gas supply device 29. A gas diffusion chamber 30 is formed inside each divided piece 23, and the gas supply device 29 introduces processing gas into the gas diffusion chamber 30 via the gas supply pipe 28. The processing gas introduced into the gas diffusion chamber 30 is supplied to the processing space U through the gas holes 27.

[0021] In the plasma processing apparatus 10, a high-frequency power supply 32 is connected to the inductively coupled antenna 26 via a matching box 31. The high-frequency power supply 32 supplies high-frequency power for generating plasma, for example, 13.56 MHz, to the inductively coupled antenna 26. As a result, an eddy current is induced in each of the segments 23 constituting the window member 12, circulating from the upper surface (the inductively coupled antenna 26 side) to the lower surface (the processing space U side). This eddy current forms an inductive electric field in the processing space U. This inductive electric field then excites the processing gas supplied to the processing space U to generate plasma. Note that the window member 12 may be made of a dielectric material rather than a conductor such as metal. In this case, the electric field formed by the inductively coupled antenna 26 passes through the window member 12 and directly acts on the processing gas to generate plasma.

[0022] Furthermore, a high-frequency power supply 34 is connected to the mounting table 13 via a matching box 33. The high-frequency power supply 34 supplies, for example, 3.2 MHz high-frequency bias power to the mounting table 13. This allows various ions in the plasma in the processing space U to be attracted to the substrate G, thereby performing a film formation process to form a protective film on the substrate G.

[0023] Furthermore, in the plasma processing apparatus 10, an exhaust port 35 is formed on the bottom surface of the processing chamber 11. An exhaust device 36 such as a turbo molecular pump or a dry pump is connected to the exhaust port 35. When performing a film formation process, the exhaust device 36 maintains the processing space U at a predetermined pressure lower than atmospheric pressure.

[0024] The plasma processing apparatus 10 further includes a control unit 37. The control unit 37 is composed of a computer having at least a CPU and a memory, and the memory stores recipes (programs) for performing film formation processes and the like, and software for performing a plasma stability evaluation method described below.

[0025] In the plasma processing apparatus 10, when the plasma stability evaluation method is performed, the control unit 37 executes a process for acquiring an index (hereinafter referred to as an "evaluation index") for evaluating the stability of the plasma (fluctuation in light emission). Figure 2 is a flowchart showing the evaluation index acquisition process executed by the control unit 37.

[0026] In the process of FIG. 2, first, plasma is generated in the processing space U, and a moving image of the generated plasma is captured by the camera 22 (step S1).

[0027] However, when evaluating plasma stability based on the plasma light emission mode, if all pixels constituting each frame image constituting a captured video are evaluated, it becomes difficult to grasp the characteristics of the plasma light emission mode that occurs unevenly in each frame image. Therefore, in this embodiment, the target on the frame image when evaluating plasma stability is narrowed down to an area (hereinafter referred to as the "evaluation target area") composed of pixels corresponding to the spatial area where the plasma emits light on the frame image. Correspondingly, in the process of Figure 2, an evaluation target area to be applied to each frame image is set in a reference frame image (hereinafter referred to as the "reference frame image").

[0028] Specifically, first, a reference frame image is acquired (step S2). The reference frame image is an image used to set an evaluation threshold and an evaluation target region, and is a frame image corresponding to the time when the plasma generated in the processing space U has stabilized. In this embodiment, the reference frame image corresponds to the timing when a predetermined time has elapsed since the ignition of the plasma and the control unit 37 starts monitoring parameters related to the plasma (e.g., high-frequency power) (see FIG. 3). The predetermined time varies depending on the content of the plasma processing (e.g., the set values ​​of parameters related to the plasma), but is determined by confirming the time from the ignition of the plasma until the plasma has stabilized through experiments, etc.

[0029] The vertical axis of Figure 3 represents the average value of the pixel values ​​of all pixels in each frame image when all pixels constituting each frame image are converted to grayscale, and the horizontal axis represents the order of the numbers of each frame image constituting the video, corresponding to the time axis. Note that the graph in Figure 3 is used for convenience to explain the timing corresponding to the reference frame image, and the graph in Figure 3 is not created in the actual evaluation index acquisition process. In addition, in grayscale conversion, for example, the pixel value for black is 0 and the pixel value for white is 255, and the grayscale is expressed using 256 levels of brightness. In other words, the larger the pixel value, the brighter the image is displayed. However, without being limited to this, the grayscale may be expressed using more or fewer levels of brightness than 256 levels.

[0030] Next, the reference frame image is grayscaled (step S3), and a histogram showing the distribution of pixel values ​​for all pixels constituting the frame image is obtained (step S4). FIG. 4 shows a histogram of the pixel counts of all pixels when the reference frame image is grayscaled, showing not only the number of pixels (frequency) for each pixel value but also the cumulative number of pixels (cumulative frequency) for each pixel value. Here, the frequency is accumulated as the pixel value increases, so the cumulative frequency shows the maximum value (100%) when the pixel value is at its maximum value, i.e., when the pixel value is 255. Then, in the process of FIG. 2, a pixel value corresponding to a specific cumulative frequency is set as a threshold for setting an evaluation target region in the reference frame image (hereinafter referred to as the "evaluation threshold") (step S5). When setting this evaluation threshold, pixel values ​​corresponding to several cumulative frequencies are selected as candidates for the evaluation threshold.

[0031] Here, since the pixel values ​​of pixels corresponding to plasma light emission are expected to be high, candidates for the evaluation threshold are selected from those with high cumulative frequencies. In the example of FIG. 4, a pixel value corresponding to a cumulative frequency of 80% of the maximum value is selected as "threshold 1," a pixel value corresponding to a cumulative frequency of 90% of the maximum value is selected as "threshold 2," and a pixel value corresponding to a cumulative frequency of 95% of the maximum value is selected as "threshold 3." In this embodiment, the control unit 37 selects, according to a predetermined setting, the pixel value corresponding to a cumulative frequency of 80% of the maximum value, the pixel value corresponding to a cumulative frequency of 90% of the maximum value, and the pixel value corresponding to a cumulative frequency of 95% of the maximum value as candidates for the evaluation threshold. However, the number of candidates for the evaluation threshold is not limited to three, and the cumulative frequencies serving as candidates for the evaluation threshold may be cumulative frequencies other than 80%, 90%, or 95% of the maximum value.

[0032] FIG. 5 is a diagram for explaining a method for setting the evaluation threshold. FIG. 5(A) is a reference frame image, and the region where plasma is emitting light (hereinafter referred to as the "plasma emission region") is indicated by a dashed line. FIG. 5(B) shows a candidate region in the reference frame image made up of pixels whose pixel values ​​are equal to or greater than threshold 1, and FIG. 5(C) shows a candidate region in the reference frame image made up of pixels whose pixel values ​​are equal to or greater than threshold 2. Furthermore, FIG. 5(D) shows a candidate region in the reference frame image made up of pixels whose pixel values ​​are equal to or greater than threshold 3. In FIGS. 5(B) to 5(D), the candidate regions are shown as white regions unevenly distributed in the reference frame image, and each candidate region has been subjected to morphology processing. Note that if the boundaries of each candidate region are smooth and there is no noise that appears as discontinuous pixels at the boundary and is not caused by plasma light emission, there is no need to perform morphology processing on each candidate region.

[0033] Then, in the evaluation index acquisition process, the candidate evaluation threshold when the shape of the candidate region most closely resembles the shape of the plasma light-emitting region is set as the evaluation threshold. Furthermore, the candidate region corresponding to the candidate evaluation threshold set as the evaluation threshold, i.e., the region in the reference frame image constituted by pixels whose pixel values ​​are equal to or greater than the evaluation threshold, is set as the evaluation target region (step S5). In the process of FIG. 2, the evaluation target region set in the reference frame image is applied to each frame image. In other words, the evaluation target region remains unchanged in each frame image. In the example of FIG. 5, since the shape of the evaluation target region for threshold 2 most closely resembles the shape of the plasma light-emitting region, threshold 2 is set as the evaluation threshold, and the candidate region corresponding to threshold 2 is applied to each frame image as the evaluation target region.

[0034] Furthermore, the evaluation threshold in step S5 is set not by the user but by control unit 37, which uses software to calculate the similarity between the shape of the plasma light-emitting region and each candidate region, determines the value at which the similarity is greatest as the value that most closely resembles the shape of the plasma light-emitting region, and selects the candidate evaluation threshold at that time as the evaluation threshold. Examples of similarity include the overlap area between the plasma light-emitting region and each candidate region, or the degree of agreement between the outline of the plasma light-emitting region and the outline of each candidate region, and the similarity is calculated based on a predetermined method. While the user may simply determine the degree of similarity between the plasma light-emitting region and each candidate region, it is preferable to calculate the similarity using software from the standpoints of consistency, accuracy, etc.

[0035] Next, each frame image constituting the captured video is grayscaled (step S6), and the evaluation target area set in the reference frame image is applied to each grayscaled frame image (step S7).

[0036] Next, for each frame image, the average value of the pixel values ​​of all pixels inside the applied evaluation target region is calculated (step S8).The calculated average pixel values ​​of the evaluation target region are then arranged in numerical order for each frame image.In other words, the calculated average pixel values ​​of the evaluation target region are arranged in chronological order to create time-series data of the average pixel values ​​of the evaluation target region (step S9).

[0037] Fig. 6 is a graph showing time-series data of the average pixel values ​​of the evaluation target region. In the graph of Fig. 3, the vertical axis was the average pixel value of all pixels constituting each frame image, but in the graph of Fig. 6, the vertical axis is the average pixel value of the evaluation target region of each frame image. The evaluation target region is composed of strongly luminescent pixels whose pixel values ​​are equal to or greater than the evaluation threshold, or strongly luminescent pixels and their neighboring pixels. Therefore, the average pixel values ​​of the graph of Fig. 6 are generally higher than the average pixel values ​​of the graph of Fig. 3.

[0038] 6, as the frame image number increases over time, for example, when the frame image number exceeds approximately 830, the average pixel value does not become irregularly distorted, but begins to fluctuate regularly. Fluctuations in plasma light emission are often accompanied by regular fluctuations in light emission, and since fluctuations in light emission are expressed by fluctuations in pixel values, the regular fluctuations in the average pixel value are thought to correspond to fluctuations in plasma light emission. Therefore, the applicant confirmed whether changes in the regular fluctuations in the average pixel value occur when the evaluation threshold is changed, i.e., when the evaluation target area is changed.

[0039] Specifically, thresholds 1, 2, and 3 were used as evaluation thresholds, and candidate regions (Figures 5B to 5D) corresponding to each evaluation threshold were used as evaluation target regions for each grayscaled frame image. The average pixel values ​​of all pixels within each evaluation target region were calculated. The average pixel values ​​of the evaluation target regions corresponding to thresholds 1, 2, and 3 were then arranged in the order of the frame image numbers to create time-series data of the average pixel values ​​of the evaluation target regions. The average pixel values ​​of all pixels constituting each grayscaled frame image were also calculated, and the calculated average pixel values ​​were arranged in the order of the frame image numbers to create time-series data of the average pixel values ​​for comparison. Calculating the average pixel values ​​of all pixels constituting each frame image is nothing more than setting a pixel value of 0 as the evaluation threshold and calculating the average pixel values ​​of all pixels in a region defined by multiple pixels with a pixel value of 0 or greater. Therefore, hereinafter, calculating the average pixel values ​​of all pixels constituting each frame image will be referred to as the "threshold 0" case. In the following description, the average value of pixel values ​​in the evaluation target region will be simply referred to as the "average pixel value."

[0040] Figure 7 is a graph showing time-series data of the average pixel value when the evaluation threshold is changed. As shown in Figure 7, regardless of the threshold used, once the frame image number exceeds approximately 830, the average pixel value stops being irregular and begins to fluctuate regularly. Figure 7 compares the regular fluctuations of these average pixel values.

[0041] It was confirmed that the amplitude W1 of the fluctuation in the average pixel value in the case of "Threshold 1" was approximately 3, which was larger than the amplitude W0 of the fluctuation in the average pixel value in the case of "Threshold 0", which was approximately 2. It was also confirmed that the amplitude W2 of the fluctuation in the average pixel value in the case of "Threshold 2" was approximately 5, which was larger than the amplitude W1 of the fluctuation in the average pixel value in the case of "Threshold 1". In other words, it was found that the greater the cumulative frequency of the pixel value corresponding to the threshold, the larger the amplitude of the fluctuation in the average pixel value. It was also confirmed that the amplitude W3 of the fluctuation in the average pixel value in the case of "Threshold 3" was approximately 5, which was almost the same as the amplitude W2 of the fluctuation in the average pixel value in the case of "Threshold 2".

[0042] As described above, the regular fluctuations in the average pixel value are thought to correspond to fluctuations in the plasma light emission. Furthermore, the greater the amplitude of the fluctuations in the average pixel value, the easier it is to grasp signs of change, making it preferable for use in evaluating fluctuations in the plasma light emission. Therefore, from the perspective of ease of evaluating fluctuations in the plasma light emission, the greater the cumulative frequency of pixel values ​​corresponding to the threshold, and it is preferable to use "threshold 2" or "threshold 3" as the evaluation threshold. Considering the approximation between the shape of the evaluation target region and the shape of the plasma light emission region in the reference frame image, and the ease of evaluating fluctuations in the plasma light emission, it is most preferable to use "threshold 2" as the evaluation threshold in this embodiment. Note that using "threshold 3" reduces the number of pixels constituting the evaluation target region compared to when "threshold 2" is used, thereby reducing the accuracy of the average pixel value, so it is preferable to use "threshold 2" as the evaluation threshold.

[0043] However, even if time series data of average pixel values ​​is created using "Threshold 2" as the evaluation threshold, the time series data may be considered to be a superposition of waves of multiple average pixel values ​​with different periods (frequencies) and convergence times. In that case, in the created time series data of average pixel values, the superposition of waves of multiple average pixel values ​​with different periods and convergence times may make the amplitude of waves of average pixel values ​​with periods that more strongly reflect fluctuations in plasma light emission less noticeable.

[0044] Therefore, in the evaluation index acquisition process, time-frequency analysis is performed on the time series data of average pixel values ​​created based on the evaluation target region (step S10). Here, a one-dimensional discrete wavelet transform is performed on the time series data of average pixel values ​​created based on the evaluation target region to acquire multiple frequency components, and the time series data of average pixel values ​​is decomposed into multiple components, for example, high-frequency components and low-frequency components. Specifically, the time series data is decomposed into time series data of average pixel values ​​of low-frequency components consisting of a composite wave of low-frequency components from level 0 to level 2 (FIG. 8(A)), and time series data of average pixel values ​​of high-frequency components consisting of a composite wave of high-frequency components from level 3 to level 5 (FIG. 8(B)). At this time, as shown in FIG. 8(A), there is almost no regular fluctuation in the average pixel values ​​in the time series data of average pixel values ​​consisting of a composite wave of low-frequency components (low-frequency components). On the other hand, as shown in FIG. 8(B), the time series data of the pixel value averages, which are composed of a composite wave (high-frequency component) of frequency components on the high-frequency side, emphasizes the regular fluctuations of the pixel value averages. Here, based on a general standard, the low-frequency components are composed of levels 0 to 2, and the high-frequency components are composed of levels 3 to 5, but the method of decomposition into low-frequency and high-frequency components is not limited to this. For example, the boundary (standard) of the frequency components for decomposition into low-frequency and high-frequency components may be appropriately determined. In this case, the low-frequency components may be set so as to include at least level 0, the lowest frequency component, and the high-frequency components may be set by frequency components on the high-frequency side that are not included in the low-frequency components. The boundary of the frequency components for decomposition into low-frequency and high-frequency components may be set in advance, for example, by obtaining frequency components in which regular fluctuations appear through experiments or the like, and setting the boundary so that the high-frequency components include frequency components at or above the frequency of the frequency components in which regular fluctuations appear.

[0045] Therefore, in the evaluation index acquisition process, after one-dimensional discrete wavelet transform in step S10, time series data of average pixel values ​​consisting of high-frequency components is extracted in order to evaluate fluctuations in plasma light emission (step S11).Then, the amplitude of regular fluctuations in the time series data of average pixel values ​​consisting of high-frequency components is acquired as the evaluation index.

[0046] The plasma stability evaluation method then determines whether the acquired amplitude exceeds a predetermined criterion, and if it does, determines that fluctuation in the plasma light emission is occurring, and if it does not, determines that fluctuation in the plasma light emission is not occurring.

[0047] If it is determined that fluctuations in the plasma light emission are occurring, the control unit 37 either interrupts the film formation process in the plasma processing device 10 or displays a warning or the like on a display unit (not shown) provided in the plasma processing device 10.

[0048] 9 and 10 are graphs showing time series data of average pixel values ​​at each level obtained by applying a one-dimensional discrete wavelet transform to time series data of average pixel values ​​created based on the evaluation target region. Fig. 9(A) is time series data of average pixel values ​​at level 0, Fig. 9(B) is time series data of average pixel values ​​at level 1, and Fig. 9(C) is time series data of average pixel values ​​at level 2. Fig. 10(A) is time series data of average pixel values ​​at level 3, Fig. 10(B) is time series data of average pixel values ​​at level 4, and Fig. 10(C) is time series data of average pixel values ​​at level 5.

[0049] Comparing the time series data of average pixel values ​​in Figures 9(A) to 10(C), a regular fluctuation in the time series data of average pixel values ​​occurs only in the time series data of average pixel values ​​of frequency components on the high frequency side of level 3 or higher. In other words, a regular fluctuation in the time series data of average pixel values ​​corresponding to fluctuations in plasma light emission occurs only in the time series data of average pixel values ​​of frequency components on the high frequency side. Therefore, if a regular fluctuation occurs in the time series data of average pixel values ​​of frequency components on the high frequency side of level 3 or higher, it can be determined that fluctuations in plasma light emission are occurring. In this case, it can be said that whether fluctuations in plasma light emission are occurring can be determined by decomposing the time series data of average pixel values ​​according to frequency (period) and checking whether regular fluctuations occur in the time series data of average pixel values ​​of frequency components on the high frequency side. In this case, the occurrence of fluctuations in plasma light emission can be determined using the frequency (period), and therefore the period of the time series data of average pixel values ​​corresponds to the evaluation index.

[0050] According to this embodiment, the candidate evaluation threshold when the shape of the candidate region in the reference frame image constituted by pixels whose pixel values ​​are equal to or greater than the candidate evaluation threshold is most similar to the shape of the plasma light-emitting region is set as the evaluation threshold, and the region in the reference frame image constituted by pixels whose pixel values ​​are equal to or greater than the evaluation threshold is set as the evaluation target region. This evaluation target region is then used to determine the fluctuation in the plasma light emission. This makes it possible to prevent the fluctuation in the plasma light emission from being determined using pixels far from the plasma light-emitting region, thereby enabling accurate plasma stability evaluation.

[0051] Although the preferred embodiments of the present disclosure have been described above, the present disclosure is not limited to the above-described embodiments, and various modifications and changes are possible within the scope of the gist of the present disclosure.

[0052] In this embodiment, the reference frame image is acquired from a video captured of the plasma whose stability is to be evaluated (whether or not fluctuation in light emission occurs is determined). However, the reference frame image may also be acquired from a video captured of a plasma different from the plasma whose stability is to be evaluated. In this case, a frame image corresponding to a timing when a predetermined time has elapsed since the ignition of the plasma different from the plasma whose stability is to be evaluated is acquired as the reference frame image.

[0053] In this embodiment, the amplitude of the regular fluctuations in the time-series data of the average pixel values ​​or the period of the time-series data of the average pixel values ​​is used as the evaluation index, but a combination of both may be used as the evaluation index. In this case, the time-series data of the average pixel values ​​is decomposed for each period, and it is confirmed whether or not regular fluctuations occur in the time-series data of the average pixel values ​​of the high-frequency components. Furthermore, it is determined whether or not the amplitude of the regular fluctuations exceeds the amplitude used as the judgment criterion. This allows for more accurate determination of the occurrence of fluctuations in the plasma light emission.

[0054] Furthermore, in this embodiment, camera 22 captures the plasma moving image in color, but camera 22 may capture the plasma moving image in monochrome. In this case, grayscale images can be obtained directly from the moving image, eliminating the need to grayscale each frame image of the moving image and reducing the load on control unit 37.

[0055] In this embodiment, the control unit 37 of the plasma processing apparatus 10 executes the evaluation index acquisition process and determines whether fluctuations in the plasma light emission have occurred. However, an information processing apparatus other than the plasma processing apparatus 10, such as an external server or client PC, may execute the evaluation index acquisition process and determine whether fluctuations in the plasma light emission have occurred.

[0056] Furthermore, in this embodiment, the evaluation index acquisition process has been described as being performed in a plasma processing apparatus 10 that generates plasma by inductive coupling. However, the plasma processing apparatus in which the evaluation index acquisition process is performed is not limited to this. For example, a plasma processing apparatus that generates plasma by capacitive coupling, a plasma processing apparatus that generates plasma by microwaves, or another method may be used. Furthermore, the process performed on the glass substrate G in the plasma processing apparatus 10 is not limited to a film formation process, and may be an etching process, an ashing process, or other process using plasma. Furthermore, although the plasma processing apparatus 10 includes a rectangular cylindrical processing chamber 11, the shape of the processing chamber 11 is not limited to a rectangular cylindrical shape and may be another shape, such as a cylindrical shape. Furthermore, when the processing chamber 11 is cylindrical, the plasma processing apparatus 10 performs a film formation process or the like on a circular substrate, such as a silicon semiconductor wafer, instead of the glass substrate G. [Explanation of symbols]

[0057] 10. Plasma processing device 11 Processing container 37 Control Unit

Claims

1. A plasma stability evaluation method for evaluating stability of plasma generated inside a processing vessel provided in a plasma processing apparatus, comprising: taking a video of the generated plasma; a step of acquiring a reference frame image from the captured video, setting an evaluation threshold based on a histogram showing a distribution of pixel values ​​when a plurality of pixels constituting the reference frame image are grayscaled, and setting an evaluation target area based on the evaluation threshold; A step of grayscaling each frame image constituting the captured moving image, and applying the evaluation target area to each of the grayscaled frame images; calculating an average value of pixel values ​​of all pixels inside the evaluation target area for each frame image constituting the captured moving image; creating time-series data of average values ​​of pixel values ​​of all pixels inside the evaluation target region calculated from each of the frame images; a step of subjecting the created time series data to time-frequency analysis to decompose it into a plurality of frequency components; and a step of decomposing the decomposed plurality of frequency components into low-frequency components consisting of the frequency components on the low-frequency side and high-frequency components consisting of the frequency components on the high-frequency side based on a predetermined criterion, and extracting the high-frequency components.

2. a shape of a region formed by pixels whose pixel values ​​are equal to or greater than the evaluation threshold when the reference frame image is grayscaled is most similar to a shape of the plasma light emission region; 2. The plasma stability evaluation method according to claim 1, wherein an area constituted by pixels whose pixel values ​​are equal to or greater than the evaluation threshold in the grayscaled reference frame image is set as the evaluation target area.

3. 2. The plasma stability evaluation method according to claim 1, wherein the reference frame image is a frame image corresponding to a timing when a predetermined time has elapsed since the ignition of the plasma.

4. 2. The plasma stability evaluation method according to claim 1, wherein the reference frame image is a frame image in a moving image of a plasma different from the plasma whose stability is to be evaluated, the frame image corresponding to a timing at which a predetermined time has elapsed since the ignition of the plasma different from the plasma whose stability is to be evaluated.

5. 2. The plasma stability evaluation method according to claim 1, wherein the plasma stability is evaluated based on at least one of the amplitude of fluctuations of the high frequency component and the period of the created time series data.

6. The plasma stability evaluation method according to claim 5 , wherein the fluctuation in the plasma light emission corresponds to a fluctuation in the high frequency component.

7. A plasma processing apparatus for performing plasma processing on a substrate, a processing vessel in which plasma is generated; a control unit; The control unit taking a video of the generated plasma; a step of acquiring a reference frame image from the captured video, setting an evaluation threshold based on a histogram showing a distribution of pixel values ​​when a plurality of pixels constituting the reference frame image are grayscaled, and setting an evaluation target area based on the evaluation threshold; A step of grayscaling each frame image constituting the captured moving image, and applying the evaluation target area to each of the grayscaled frame images; calculating an average value of pixel values ​​of all pixels inside the evaluation target area for each frame image constituting the captured moving image; creating time-series data of average values ​​of pixel values ​​of all pixels inside the evaluation target region calculated from each of the frame images; a step of subjecting the created time series data to time-frequency analysis to decompose it into a plurality of frequency components; and a step of decomposing the decomposed frequency components into low-frequency components consisting of the frequency components on the low-frequency side and high-frequency components consisting of the frequency components on the high-frequency side based on a predetermined criterion, and extracting the high-frequency components.

Images