Plasma chamber leak detection device and detection method using same

Abstract

The present invention relates to a plasma chamber leak detection device and a chamber leak detection method using same. The plasma chamber leak detection device of the present invention comprises: a signal collection unit for obtaining one or more detection signals in a detection wavelength band and one or more non-detection signals in a non-detection wavelength band; a leak occurrence recognition unit for recognizing a normal state when an abnormal signal is not collected from the detection signal and the non-detection signal, recognizing an abnormal leak occurrence state when an abnormal signal is collected from the detection signal and an abnormal signal is not collected from the non-detection signal, and recognizing an offset noise occurrence normal state when an abnormal signal is collected from both the detection signal and the non-detection signal; and a display unit for providing a determination result of the leak occurrence recognition unit.

Description

Plasma chamber leak detection device and detection method using the same

[0001] The present invention relates to a plasma chamber leak detection device and a chamber leak detection method using the same.

[0002] Self-plasma optical emission spectroscopy (SPOES) is composed of an optical emission spectroscopy (OES) with a plasma generator attached to the front. OES can determine the composition of the gas and the amount of each component in the chamber by spectroscopically analyzing the optical signal of the plasma coming through the light receiving unit. Since OES analyzes using the optical signal of the plasma, if plasma is not generated in the chamber, the gas condition inside the chamber cannot be monitored. Therefore, in order to detect chamber leaks, plasma must be continuously generated, and SPOES includes a plasma generator to improve this problem. Leak detection evaluation is performed based on the optical signal collected in this way.

[0003] Data normalization during the detection evaluation process is a type of data scaling technique. It involves converting the maximum and minimum values ​​of each variable to a specific range to improve the problem of incorrect data analysis results when certain variables are given too much weight. Normalization for data scaling can be described, for example, by the following mathematical equation.

[0004] [Mathematical formula]

[0005]

[0006] The existing SPOES data acquisition method selects a specific single wavelength band, measures the signal at that wavelength over time, and determines that a leak has occurred when the signal intensity increases in the leak detection wavelength band, which can monitor the leak component, and exceeds the set threshold.

[0007] In this way, when detecting a leak with a single wavelength, if offset noise occurs at that wavelength, there may be a problem of misreading it as a leak.

[0008] One object of the present invention is to provide a device capable of effectively detecting chamber leaks and a method for manufacturing the same.

[0009] Another object of the present invention is to provide a chamber leak detection device and detection method capable of preventing false leak detection due to offset noise of a multi-wavelength optical signal.

[0010] Another object of the present invention is to provide a chamber leak detection device and detection method that prevents false detection by converting an optical signal into a two-dimensional image.

[0011] Furthermore, a chamber leak detection device and detection method are provided that detects whether a two-dimensional image has a leak through an artificial intelligence image classification algorithm.

[0012] However, the problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned can be clearly understood by a person having ordinary skill in the relevant technical field from the description below.

[0013] The plasma chamber leak detection device of the present invention comprises: a signal collection unit that obtains at least one detection signal in a detection wavelength band and at least one non-detection signal in a non-detection wavelength band; a leak occurrence recognition unit that recognizes a normal state when no abnormal signal is collected from the detection signal and the non-detection signal; that recognizes an abnormal leak occurrence state when an abnormal signal is collected from the detection signal and no abnormal signal is collected from the non-detection signal; and that recognizes an offset noise occurrence normal state when an abnormal signal is collected from both the detection signal and the non-detection signal; and a display unit that provides a determination result of the leak occurrence recognition unit.

[0014] The above abnormal signal may be a signal that exceeds a preset signal value.

[0015] In an N2 gas atmosphere, the detection wavelength range may include at least one wavelength selected from the group consisting of 234.7 nm, 235 nm, 235.4 nm, 236.1 nm, 236.5 nm, 245.8 nm, 246.1 nm, 246.8 nm, and 258.2 nm, and the non-detection wavelength range may include at least one wavelength selected from the group consisting of 313.6 nm, 334.4 nm, 334.7 nm, 335.1 nm, 366.7 nm, 370.9 nm, 373.3 nm, 393.5 nm, 394.6 nm, and 807.3 nm.

[0016] The above leak occurrence recognition unit may include a 2D image generation unit that generates a 2D image from the detection signal and the non-detection signal; and a 2D image determination unit that determines whether a leak has occurred from the 2D image.

[0017] The above 2D image may represent signal intensity according to time and wavelength.

[0018] The above signal strength may be displayed by color change.

[0019] The above 2D image determination unit may determine that, in the 2D image, if it is recognized that there is no significant color change in the detection wavelength band and the non-detection wavelength band, it is determined that the state is normal; if it is recognized that, in the 2D image, there is a significant color change in the detection wavelength band and there is no significant color change in the non-detection wavelength band, it is determined that the state is abnormal leak occurrence; and if it is recognized that, in the 2D image, there is a significant color change in both the detection wavelength band and the non-detection wavelength band, it is determined that the offset noise occurrence is normal.

[0020] It may further include an artificial intelligence algorithm learning unit that learns a process of determining whether an abnormal leak occurrence state or a normal offset noise occurrence state is present from the above 2D image.

[0021] The plasma chamber leak detection device of the present invention may further include an alarm unit that provides an alarm for an abnormal leak occurrence state when the display unit indicates an abnormal leak occurrence state according to a determination result of the leak occurrence recognition unit.

[0022] The plasma chamber leak detection method of the present invention comprises a wavelength signal collection step of obtaining at least one detection signal in a detection wavelength band and at least one non-detection signal in a non-detection wavelength band; a leak occurrence determination step of determining whether a normal state is present from the collected signals; and a result display step of displaying a result of the leak occurrence determination; wherein the leak occurrence determination step recognizes that if an abnormal signal is not collected from the detection signal and the non-detection signal, it is a normal state; if an abnormal signal is collected from the detection signal and no abnormal signal is collected from the non-detection signal, it is recognized as an abnormal leak occurrence state; and if an abnormal signal is collected from both the detection signal and the non-detection signal, it is recognized as a normal state in which offset noise is generated.

[0023] The plasma chamber leak detection method of the present invention may further include, after the wavelength signal collection step, a 2D image generation step of generating a 2D image from the detection signal and the non-detection signal.

[0024] The above 2D image represents signal intensity according to time and wavelength, and may display the signal intensity as a color change.

[0025] The above leak occurrence determination step may be such that, if it is recognized that there is no significant color change in the detection wavelength band and the non-detected wavelength band in the 2D image, it is determined as a normal state; if it is recognized that there is a significant color change in the detection wavelength band and no significant color change in the non-detected wavelength band in the 2D image, it is determined as an abnormal leak occurrence state; and if it is recognized that there is a significant color change in both the detection wavelength band and the non-detected wavelength band in the 2D image, it is determined as a normal state of offset noise occurrence.

[0026] According to one aspect of the present invention, a device capable of effectively detecting a chamber leak in a plasma device and a method for manufacturing the same are provided.

[0027] According to another aspect of the present invention, a chamber leak detection device and detection method are provided that can prevent false leak detection due to offset noise that may appear when determining whether there is a leak from a single wavelength measurement value.

[0028] According to another aspect of the present invention, a chamber leak detection device and detection method are provided that utilize a two-dimensional imaged signal and determine the occurrence of a leak through artificial intelligence learning.

[0029] Figure 1 is a conceptual diagram showing the configuration of a chamber leak detection device.

[0030] FIG. 2 is another example of the present invention, which shows signals in a normal state, a normal state including offset noise, and an abnormal leak occurrence state as two-dimensional images.

[0031] FIG. 3 is a conceptual diagram of a process for converting a collected signal into a two-dimensional image, as another embodiment of the present invention.

[0032] FIG. 4 is a flowchart showing the communication configuration and algorithm progression sequence of MES and FDC according to an additional embodiment of the present invention.

[0033] Hereinafter, embodiments are described in detail with reference to the attached drawings. However, the embodiments may be modified in various ways, and the scope of the patent application is not limited or restricted by these embodiments. It should be understood that all modifications, equivalents, or alternatives to the embodiments are included within the scope of the patent application.

[0034] The terms used in the examples are for illustrative purposes only and should not be construed as limiting. Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as "comprise" or "have" are intended to specify the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, but should be understood to not preclude the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0035] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art to which the embodiments pertain. Terms defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and shall not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0036] In addition, when describing with reference to the attached drawings, the same components will be given the same reference numerals regardless of the drawing numbers, and redundant descriptions thereof will be omitted. When describing an embodiment, if it is determined that a detailed description of a related known technology may unnecessarily obscure the gist of the embodiment, the detailed description thereof will be omitted. In addition, when describing a component of an embodiment, terms such as first, second, A, B, (a), (b), etc. may be used. These terms are only to distinguish the component from other components, and the nature, order, or sequence of the component is not limited by the terms. When a component is described as being "connected," "coupled," or "connected" to another component, it should be understood that the component may be directly connected or connected to the other component, but another component may also be "connected," "coupled," or "connected" between each component.

[0037] Components included in one embodiment and components with common functions will be described using the same names in other embodiments. Unless otherwise stated, the descriptions given in one embodiment can be applied to other embodiments, and detailed descriptions will be omitted to the extent of overlap.

[0038]

[0039] Figure 1 is a conceptual diagram showing the configuration of a chamber leak detection device.

[0040] Referring to FIG. 1, a chamber leak detection device (100) for detecting in real time the occurrence of a leak, which is one of the main causes of product defects, in a chamber of a semiconductor manufacturing device such as a plasma generating device, may include a plasma chamber (110), a self-plasma optical emission spectroscopy (SPOES) (120), a manufacturing execution system (MES) (130), and a fault detection and classification (FDC) (140).

[0041] A plasma chamber (110) and a self-plasma optical emission spectrometer (SPOES) (120) may be connected to monitor the gas state of the plasma chamber in real time. For example, the self-plasma optical emission spectrometer (SPOES) (120) may be connected through a chamber window. Data measured by the self-plasma optical emission spectrometer (SPOES) (120) may be transmitted to a manufacturing execution system (MES) (130), and multi-wavelength data selected from the MES (130) may be transmitted to a fault detection and classification unit (FDC) (140) through two-way communication.

[0042] One of the signal collection methods of the self-plasma optical emission spectrometer (SPOES) is to collect signals of a specific single wavelength, and the signals of that wavelength are collected over time. When a leak occurs, the signal intensity increases in the leak detection wavelength range where the leak component can be monitored, and if it exceeds the set threshold, it is determined to be a leak.

[0043] However, if a leak is determined using a signal of a single wavelength in this way, the detection resolution is reduced due to fluctuation components caused by noise, and in particular, if an offset noise signal is generated at the corresponding wavelength even though no actual leak has occurred, the detection system described above may misjudge this as a leak occurring.

[0044] In consideration of this, the present invention utilizes signals of multiple wavelengths, particularly including a detection wavelength and a non-detected wavelength, to distinguish between i) a normal state, ii) a normal state in which offset noise occurs but no leak occurs, and iii) an abnormal leak occurrence state. In other words, by simultaneously monitoring the detection wavelength and the non-detected wavelength, the accuracy of determining the occurrence of a leak is increased.

[0045] The plasma chamber leak detection device of the present invention comprises: a signal collection unit that obtains at least one detection signal in a detection wavelength band and at least one non-detection signal in a non-detection wavelength band; a leak occurrence recognition unit that recognizes a normal state when no abnormal signal is collected from the detection signal and the non-detection signal; that recognizes an abnormal leak occurrence state when an abnormal signal is collected from the detection signal and no abnormal signal is collected from the non-detection signal; and that recognizes an offset noise occurrence normal state when an abnormal signal is collected from both the detection signal and the non-detection signal; and a display unit that provides a determination result of the leak occurrence recognition unit.

[0046] The above abnormal signal may be a signal that exceeds a preset signal value.

[0047] In an N2 gas atmosphere, the detection wavelength range may include at least one wavelength selected from the group consisting of 234.7 nm, 235 nm, 235.4 nm, 236.1 nm, 236.5 nm, 245.8 nm, 246.1 nm, 246.8 nm, and 258.2 nm, and the non-detection wavelength range may include at least one wavelength selected from the group consisting of 313.6 nm, 334.4 nm, 334.7 nm, 335.1 nm, 366.7 nm, 370.9 nm, 373.3 nm, 393.5 nm, 394.6 nm, and 807.3 nm.

[0048] The above detection wavelength range and non-detection wavelength range are exemplary in an N2 gas atmosphere, and the present invention is not limited thereto.

[0049]

[0050] The above leak occurrence recognition unit may include a 2D image generation unit that generates a 2D image from the detection signal and the non-detection signal; and a 2D image determination unit that determines whether a leak has occurred from the 2D image.

[0051] As a way to solve the problem of misjudgment in judging the occurrence of a leak based on a single wavelength signal, we propose converting a multi-wavelength signal into a 2D image as one way to implement a leak detection method based on a multi-wavelength signal.

[0052] The above 2D image may represent the detection signal wavelength band and the non-detected signal wavelength band together in one image.

[0053]

[0054] The above 2D image may represent signal intensity according to time and wavelength. It may be 2D imaged so that signal intensity is displayed according to time and wavelength.

[0055]

[0056] The above signal intensity may be indicated by a color change. The color change may be a change in color or a change in color intensity.

[0057]

[0058] The above 2D image determination unit may determine that, in the 2D image, if it is recognized that there is no significant color change in the detection wavelength band and the non-detection wavelength band, it is determined that the state is normal; if it is recognized that, in the 2D image, there is a significant color change in the detection wavelength band and there is no significant color change in the non-detection wavelength band, it is determined that the state is abnormal leak occurrence; and if it is recognized that, in the 2D image, there is a significant color change in both the detection wavelength band and the non-detection wavelength band, it is determined that the offset noise occurrence is normal.

[0059] FIG. 2 is another example of the present invention, which shows signals in a normal state, a normal state including offset noise, and an abnormal leak occurrence state as two-dimensional images.

[0060] FIG. 2 is a graph illustrating, according to one embodiment, features that may appear when wavelength signals of a steady state, a steady state with offset noise, and an abnormal leak occurrence state acquired from a self-plasma optical emission spectroscopy (SPOES) are converted into 2D images.

[0061] The normal state 2D image on the left of Fig. 2 shows no particularly distinguishable changes in the image.

[0062] In the 2D image of the abnormal leak occurrence state on the right side of Fig. 2, it is confirmed that the color changes as a result of the leak in the detection wavelength range, and no color change appears in the non-detected wavelength range. In other words, since the color change in the detection wavelength range is confirmed to be due to the leak, it can be seen that it is an abnormal leak occurrence state.

[0063] In the normal state 2D image with offset noise in the center of Figure 2, color changes appear not only in the detection wavelength range but also in the non-detected wavelength range. This can be seen as not being the result of actual leak occurrence, but rather as the result of signal change due to offset noise. In this case, despite signal occurrence, the normal state can be determined to be due to offset noise, not leak occurrence.

[0064] That is, if the wavelength signal is imaged in 2D as in Fig. 2, the wavelength signals in the normal state, the normal state including offset noise, and the abnormal leak occurrence state can be accurately distinguished.

[0065] FIG. 3 is a conceptual diagram of a process for converting a collected signal into a two-dimensional image, as another embodiment of the present invention.

[0066] Referring to Figure 3, it can be confirmed that a leak is detected in the wavelength range where the leak is detected. Although the signal-to-noise ratio of the measured data varies for each wavelength range, it can be confirmed that the presence or absence of a signal increase can be the same.

[0067]

[0068] It may further include an artificial intelligence algorithm learning unit that learns a process of determining whether an abnormal leak occurrence state or a normal offset noise occurrence state is present from the above 2D image.

[0069] FIG. 4 is a flowchart showing the communication configuration and algorithm progression sequence of a manufacturing execution system (MES) and a fault detection and classification unit (FDC) according to one embodiment.

[0070] The configuration of FIG. 4 is a flowchart showing the order of progress in a case where various additional configurations are included in addition to the essential configurations of the present invention, as a possible embodiment, and the present invention is not limited thereto.

[0071] FDC can be applied to leak detection evaluation. The leak detection evaluation process may include an AI image classification algorithm and a process for generating an alarm when a leak is detected. As described above, when data acquired from SPOES is converted into a 2D image, the results are classified into three types. Based on the characteristics of each type, the wavelength signal is classified into a normal state, a normal state with offset noise, and an abnormal leak state.

[0072] Referring to Figure 4, if an abnormal leak condition is determined to have occurred according to the detection algorithm, an abnormal leak condition alarm can be transmitted to the MES via communication. When an abnormal leak condition alarm is transmitted, the operator can confirm the occurrence of a leak through the MES.

[0073] For example, the 2D image converted data contains both detected and undetected wavelengths simultaneously. Leak judgment learning algorithms using artificial intelligence may include image classification algorithms such as CNN and DNN. In particular, CNN is a neural network model that uses convolutional preprocessing in deep learning. It can classify images by building up a hierarchy of features while maintaining spatial / local information by receiving raw input as is. To explain in more detail, the convolutional preprocessing task (convolution filter) repeatedly applies the same filter to all areas of the input image to find patterns, convolves the input image and the filter to obtain the result, and uses an activation function (such as sigmoid or ReLU) to create a new image map. It is generally composed of three layers: a convolution layer, a ReLU layer, and a pooling layer, and each layer has the following roles.

[0074] Convolutional layer: The input image passes through convolutional filters. Each filter activates a specific feature in the image.

[0075] ReLU layer: Only activated features are passed to the next layer.

[0076] Pooling layer: Reduces the number of parameters by performing nonlinear downsampling.

[0077]

[0078] When data is collected by converting to 2D images, clear characteristics can be identified between the normal state, the normal state with offset noise, and the abnormal leak occurrence state data. In the case of the abnormal leak occurrence state data, the signal increases in the detection wavelength range, the normal state data with offset noise shows simultaneous increases in the detection wavelength range and the undetected wavelength range signals, and the normal state data shows no change. These three characteristics can be labeled and trained using a CNN. By dividing the total data into 3 / 4 of the training data and 1 / 4 of the validation data, the training and detection accuracy can be compared.

[0079]

[0080] By comparing the verification accuracy obtained through the process, an appropriate learning model can be selected.

[0081]

[0082] The plasma chamber leak detection device of the present invention may further include an alarm unit that provides an alarm for an abnormal leak occurrence state when the display unit indicates an abnormal leak occurrence state according to a determination result of the leak occurrence recognition unit.

[0083] The above abnormal leak occurrence status alarm may be at least one selected from the group consisting of a light signal, an audio signal, and a vibration signal.

[0084]

[0085] The plasma chamber leak detection method of the present invention comprises a wavelength signal collection step of obtaining at least one detection signal in a detection wavelength band and at least one non-detection signal in a non-detection wavelength band; a leak occurrence determination step of determining whether a normal state is present from the collected signals; and a result display step of displaying a result of the leak occurrence determination; wherein the leak occurrence determination step recognizes that if an abnormal signal is not collected from the detection signal and the non-detection signal, it is a normal state; if an abnormal signal is collected from the detection signal and no abnormal signal is collected from the non-detection signal, it is recognized as an abnormal leak occurrence state; and if an abnormal signal is collected from both the detection signal and the non-detection signal, it is recognized as a normal state in which offset noise is generated.

[0086] The plasma chamber leak detection method of the present invention may be performed by the plasma chamber leak detection device of the present invention.

[0087]

[0088] The plasma chamber leak detection method of the present invention may further include, after the wavelength signal collection step, a 2D image generation step of generating a 2D image from the detection signal and the non-detection signal; wherein the 2D image represents signal intensity according to time and wavelength, and may display the signal intensity as a color change.

[0089]

[0090] The above leak occurrence determination step may be such that, if it is recognized that there is no significant color change in the detection wavelength band and the non-detected wavelength band in the 2D image, it is determined as a normal state; if it is recognized that there is a significant color change in the detection wavelength band and no significant color change in the non-detected wavelength band in the 2D image, it is determined as an abnormal leak occurrence state; and if it is recognized that there is a significant color change in both the detection wavelength band and the non-detected wavelength band in the 2D image, it is determined as a normal state of offset noise occurrence.

[0091]

[0092] Although the embodiments have been described above, those skilled in the art will appreciate that various technical modifications and variations can be applied based on the above. For example, appropriate results can still be achieved even if the described techniques are performed in a different order than described, and / or components of the described systems, structures, devices, circuits, etc. are combined or combined in a different manner than described, or are replaced or substituted with other components or equivalents.

[0093] Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims described below.

Claims

1. A signal collection unit that obtains one or more detection signals in a detection wavelength band and one or more non-detection signals in a non-detection wavelength band; If no abnormal signal is collected from the above detection signal and the above non-detection signal, it is recognized as normal, If an abnormal signal is collected from the above detection signal and an abnormal signal is not collected from the above non-detection signal, it is recognized as an abnormal leak occurrence state, A leak occurrence recognition unit that recognizes an offset noise occurrence as a normal state when an abnormal signal is collected from both the above detection signal and the non-detection signal; and A display unit that provides the judgment result of the above leak occurrence recognition unit; including, Plasma chamber leak detection device.

2. In paragraph 1, The above abnormal signal is a signal that exceeds a preset signal value. Plasma chamber leak detection device.

3. In paragraph 1, In an N2 gas atmosphere, The above detection wavelength range includes at least one wavelength selected from the group consisting of 234.7 nm, 235 nm, 235.4 nm, 236.1 nm, 236.5 nm, 245.8 nm, 246.1 nm, 246.8 nm and 258.2 nm, The above non-detectable wavelength range includes at least one wavelength selected from the group consisting of 313.6 nm, 334.4 nm, 334.7 nm, 335.1 nm, 366.7 nm, 370.9 nm, 373.3 nm, 393.5 nm, 394.6 nm and 807.3 nm. Plasma chamber leak detection device.

4. In paragraph 1, The above leak occurrence recognition unit, A 2D image generation unit that generates a 2D image from the above detection signal and the above non-detection signal; and A 2D image judgment unit that judges whether a leak has occurred from the above 2D image; A plasma chamber leak detection device comprising:

5. In paragraph 4, The above 2D image represents signal intensity according to time and wavelength. Plasma chamber leak detection device.

6. In paragraph 5, The signal strength is displayed as a color change, Plasma chamber leak detection device.

7. In paragraph 4, The above 2D image judgment unit, In the above 2D image, if it is recognized that there is no significant color change in the detection wavelength range and the non-detection wavelength range, it is judged to be in a normal state, In the above 2D image, if it is recognized that there is a significant color change in the detection wavelength range and no significant color change in the non-detection wavelength range, it is determined that an abnormal leak has occurred. In the above 2D image, if it is recognized that there is a significant color change in both the detection wavelength and the non-detection wavelength, it is judged that the offset noise generation is normal. Plasma chamber leak detection device.

8. In paragraph 4, It further includes an artificial intelligence algorithm learning unit that learns the process of determining whether an abnormal leak occurrence state or a normal offset noise occurrence state is present from the above 2D image. Plasma chamber leak detection device.

9. In paragraph 1, If the display unit displays the abnormal leak occurrence status based on the judgment result of the above leak occurrence recognition unit, An alarm unit further comprising: an alarm unit providing an alarm for an abnormal leak occurrence status; Plasma chamber leak detection device.

10. A wavelength signal collection step for obtaining one or more detection signals in a detection wavelength band and one or more non-detection signals in a non-detection wavelength band; A leak occurrence determination step for determining whether the collected signal is in a normal state; and A result display step for displaying the above leak occurrence determination result; Including, The above leak occurrence determination step is: If no abnormal signal is collected from the above detection signal and the above non-detection signal, it is recognized as normal, If an abnormal signal is collected from the above detection signal and an abnormal signal is not collected from the above non-detection signal, it is recognized as an abnormal leak occurrence state, When an abnormal signal is collected from both the above detection signal and the non-detection signal, the offset noise is recognized as a normal state. Plasma chamber leak detection method.

11. In paragraph 10, After the above wavelength signal collection step, A 2D image generation step for generating a 2D image from the detection signal and the non-detection signal; further comprising; Plasma chamber leak detection method.

12. In paragraph 11, The above 2D image represents signal intensity according to time and wavelength, The signal strength is displayed as a color change, Plasma chamber leak detection method.

13. In paragraph 11, The above leak occurrence determination step is: In the above 2D image, if it is recognized that there is no significant color change in the detection wavelength range and the non-detection wavelength range, it is judged to be in a normal state, In the above 2D image, if it is recognized that there is a significant color change in the detection wavelength range and no significant color change in the non-detection wavelength range, it is determined that an abnormal leak has occurred. In the above 2D image, if it is recognized that there is a significant color change in both the detection wavelength and the non-detection wavelength, it is judged that the offset noise generation is normal. Plasma chamber leak detection method.

Images