KR101305049B1 - Monitoring system of vacuum and plasma and method thereof - Google Patents

Abstract

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vacuum plasma monitoring system and a method thereof, comprising: a digital hologram sensor for photographing inside a chamber in a vacuum state or a plasma state, and generating a hologram image; And a monitoring diagnostic unit for monitoring whether the vacuum state or the plasma state is normal based on the particle distribution of the present invention. According to the present invention, the particle distribution is calculated from a digital hologram image of the sheath structure in the vacuum state or the plasma state. Patterns can be extracted to monitor vacuum or plasma conditions.

Description

Vacuum plasma monitoring system and its method {MONITORING SYSTEM OF VACUUM AND PLASMA AND METHOD THEREOF}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vacuum plasma monitoring system and method thereof, and more particularly, to measuring a vacuum plasma sheath structure and its particle distribution in a chamber using a digital hologram imaging technique, and based on the vacuum plasma or plasma state. The present invention relates to a vacuum plasma monitoring system and a method for monitoring the normality of a.

Plasma is a key energy source for the deposition and etching of fine thin films for semiconductor and display device manufacturing. Plasma fabrication involves a variety of process and equipment variables, which include radio frequency (RF) source power, bias power, pressure, and gas flow rate.

If the abnormal state of the plasma is not detected, a process and device yield are greatly reduced, and thus, a system for determining whether the plasma state is normal is required.

Sensors that monitor plasma status include Optical Emission Spectroscopy (OES), Langmuir Probe (LP), Color Monitoring System (CMS), and Current-Voltage Probe (IV Probe). Etc. Among them, the light reflection spectrometer (OES) is most widely applied, and the plasma state can be determined by measuring the reflection intensity of the incident light.

The Langmuir probe (LP) measures the basic characteristics of plasma, electron density and plasma potential, the color shift monitoring system (CMS) measures plasma color shift, and the current-voltage probe (IV probe) measures plasma current and voltage. Monitor the plasma state by measuring the variation of.

Electron densities in vacuum and plasma are mainly measured using a Langmuir probe (LP), and a non-intrusive Langmuir probe (LP) sensor is applied to monitor the plasma state.

However, since the Langmuir probe (LP) sensor measures only the average electron density of the plasma, there is a limitation in that the electron distribution at any position in the chamber cannot be measured.

In addition, the Langmuir probe (LP) sensor has a limitation that does not provide information about the charge of the particles in the vacuum before plasma generation.

Electron density has a different distribution depending on the spatial location of the chamber in vacuum and plasma, and varies with process variables. It is necessary to measure the electron distribution at high sensitivity for all process variables or for specific process variables.

The information for monitoring the state of the plasma mainly depends on the reflected light information reflected from the center of the chamber, and the information in the vicinity of the wafer near the lower side of the chamber is not available. This is due to the physical access limitations of the sensors currently used.

In addition, the current plasma monitoring system is limited to monitoring the plasma state after generating the plasma, and does not perform the monitoring function for the vacuum state before the plasma is generated.

However, since the wafer is placed in the chamber before the plasma is generated, monitoring the vacuum condition before the plasma process begins will be necessary to improve the quality of the process. In addition, the vacuum state can play an important role in the plasma state monitoring because it can be a reference state for the subsequent plasma state.

At the bottom of the chamber, there is a sheath structure that has a significant physical impact on process development and analysis.

Sheath is a region having a relatively low electron density compared to a bulk plasma region, and is in contact with a presheath layer having a relatively high electron density. Since the spatial structure of the electron density of the sheath and the friece varies sensitively with process variables, it is required to monitor the plasma state by using the change of the sheath structure.

Background art of the present invention is disclosed in Korean Patent Publication No. 10-2005-0027668 'Diagnostic System and Method for Measuring Pulsed Plasma Characteristics' (2005.03.21).

The present invention has been made to improve the above-described problems, and provides a vacuum plasma monitoring system that can monitor the normality of the vacuum or plasma state by measuring the sheath structure and its particle distribution inside the chamber in a vacuum or plasma state. Its purpose is to.

In addition, an object of the present invention is to provide a vacuum plasma monitoring method for monitoring the plasma state and the vacuum state before the plasma is generated to increase the accuracy of the plasma state monitoring and thereby improve the quality of the plasma process.

Vacuum plasma monitoring system according to an aspect of the present invention comprises a digital hologram sensor for photographing the interior of the chamber in a vacuum or plasma state to generate a holographic image; An image processor which calculates a particle distribution in the chamber from the hologram image; And a monitoring diagnosis unit configured to monitor whether the vacuum state or the plasma state is normal based on the particle distribution in the chamber.

In the present invention, the digital hologram sensor can photograph the sheath structure inside the chamber.

In the present invention, the chamber may include at least one window for receiving or outputting laser light emitted from the digital hologram sensor.

In the present invention, the monitoring diagnosis unit may monitor whether the plasma state is normal after monitoring whether the vacuum state is normal.

The image processor may include a 1-D particle counter that calculates the total number of particles included in a preset gray scale range in the hologram image by applying the following first equation to the hologram image.

………………………… (1)

... ... ... ... ... ... ... ... ... ... (One)

Where GR is a preset gray scale range, g _{(i, j)} is a gray scale value at any pixel location, and α is the number of particles per LSB inherent to the CCD sensor included in the digital hologram sensor.

The image processor may include a 1-D counter for calculating the total number of particles included in the inner region of the chamber by applying the following second equation to the hologram image.

………………………… (2)

... ... ... ... ... ... ... ... ... ... (2)

는 상기 홀로그램 영상의 임의의 픽셀 위치에서의 그레이 스케일 값인 g_(i,j)의 반전된 그레이 스케일 값, α는 상기 디지털 홀로그램 센서에 포함된 CCD 센서 고유의 LSB 당 입자수.Where GR is the preset gray scale range,

Is an inverted gray scale value of g _{(i, j), which} is a gray scale value at any pixel location of the hologram image, and α is the number of particles per LSB inherent to the CCD sensor included in the digital hologram sensor.

는 하기 제3 수학식에 의해 결정될 수 있다.In the present invention

May be determined by the following third equation.

……………… (3)

... ... ... ... ... ... (3)

).Where n is a bit per pixel of the CCD sensor included in the digital hologram sensor, A _i is a bit value at an arbitrary pixel position (ie, gray scale value g _{(i, j)} ), B _i is an arbitrary pixel Inverted bit value of A _{i at} the position (ie inverted gray scale value)

).

In the present invention, the monitoring diagnosis unit may determine that the vacuum state or the plasma state is normal when the difference value between the total particle count and the reference particle count calculated by the image processor is within a first allowable reference value.

In the present invention, the image processing unit applies a fourth equation to the hologram image to calculate a 2-D particle counter for calculating the particle distribution in the x-axis or y-axis direction included in a preset gray scale range in the hologram image It may include.

………………………… (4)

... ... ... ... ... ... ... ... ... ... (4)

Here, m and n are the total number of pixels in the x- and y-axis directions of the hologram image, Δx and Δy are variables for section separation in the x- and y-axis directions, and N _i is the i-th position in the x-axis direction. Number of particles, N _j is the number of particles at the j-th position in the y-axis direction.

In the present invention, the image processor may calculate the thickness of each layer of the sheath structure using the particle distribution.

In the present invention, the monitoring diagnosis unit may determine that the vacuum state or the plasma state is normal when the difference value between the interlayer thickness and the interlayer reference thickness calculated by the image processor is within an allowable thickness difference.

In the present invention, the image processor may calculate the number of interlayer particles of the sheath structure by applying the following fifth equation to the particle distribution.

………………………… (5)

... ... ... ... ... ... ... ... ... ... (5)

Where a and b are the positions of the two peaks to be monitored and N _j is the number of particles at any j th position between the two peaks a and b on the y-axis.

In the present invention, the monitoring diagnosis unit may determine that the vacuum state or the plasma state is normal when the difference value between the interlayer particle count and the interlayer reference particle count calculated by the image processor is within a second acceptable reference value.

In the present invention, the image processor may include a 3-D counter for calculating the 3-D particle distribution by applying the particle distribution in the z-axis direction.

According to another aspect of the present invention, there is provided a vacuum plasma monitoring system comprising: a digital hologram sensor for photographing inside a chamber in a vacuum state or a plasma state to generate hologram image data; An image processor extracting a current pattern inside the chamber using the holographic image data; And a monitoring diagnosis unit configured to monitor whether the vacuum state or the plasma state is normal based on the current pattern.

In the present invention, the digital hologram sensor can photograph the sheath structure inside the chamber.

In the present invention, the image processing unit may extract a current pattern from a part or all of the hologram image according to the following sixth equation.

………………………… (6)

... ... ... ... ... ... ... ... ... ... (6)

Here, G _p is a gray scale value of a pixel included in a preset gray scale range, and n is a bit per pixel of a CCD sensor included in the digital hologram sensor.

In the present invention, if the difference between the gray scale value at a specific pixel position and the reference gray scale value at the pixel position constituting the current pattern extracted from the image processing unit is within the limit tolerance value, the vacuum state or the plasma state is reduced. It can be judged that it is normal.

In the present invention, the monitoring diagnosis unit may determine that the vacuum state or the plasma state is normal when the ratio of all pixels of the pixel that exceeds the limit allowance is equal to or less than the reference ratio.

According to an aspect of the present invention, there is provided a vacuum plasma monitoring method comprising: receiving, by a monitoring controller, a first hologram image of an interior of a vacuum chamber from a digital hologram sensor; Calculating, by the monitoring controller, a first particle distribution in the chamber from the first hologram image; Determining, by the monitoring controller, whether the vacuum state is normal based on the first particle distribution; If it is determined that the vacuum state is normal, the monitoring control unit receiving a second hologram image of the inside of the chamber in the plasma state from the digital hologram sensor; Calculating, by the monitoring controller, a second particle distribution in the chamber from the second hologram image; And determining, by the monitoring controller, whether the plasma state is normal based on the second particle distribution.

According to another aspect of the present invention, there is provided a vacuum plasma monitoring method comprising: receiving, by a monitoring controller, a first hologram image of an interior of a chamber in a vacuum state from a digital hologram sensor; Extracting, by the monitoring controller, a first current pattern from the first hologram image; Determining, by the monitoring controller, whether the vacuum state is normal based on the first current pattern; If it is determined that the vacuum state is normal, the monitoring control unit receiving a second hologram image of the inside of the chamber in the plasma state from the digital hologram sensor; Extracting, by the monitoring controller, a second current pattern inside the chamber from the second hologram image; And determining, by the monitoring controller, whether the plasma state is normal based on the second current pattern.

In the present invention, the first hologram image and the second hologram image may be images of a sheath structure inside the chamber.

According to the present invention, the vacuum state or the plasma state can be monitored by calculating the particle distribution or extracting the current pattern from the digital hologram image photographing the inside of the chamber in the vacuum state or the plasma state.

In particular, according to the present invention, since the sheath structure inside the chamber that is sensitively changed depending on the process conditions and the particle distribution near the wafer can be monitored, the accuracy of the plasma state monitoring can be improved, thereby improving the quality of the plasma process. It can increase.

In addition, according to the present invention, by monitoring the vacuum state before the plasma is first generated and then monitoring the plasma state, it is possible to monitor whether the plasma state is normal on the basis of the vacuum state.

1 is a block diagram of a vacuum plasma monitoring system according to an embodiment of the present invention.
2 is a block diagram showing an example of a vacuum plasma monitoring system according to an embodiment of the present invention.
3 is a block diagram illustrating an image processor in a vacuum plasma monitoring system according to an exemplary embodiment of the present invention.
4 is an exemplary diagram of a sheath structure using a digital hologram sensor of a vacuum plasma monitoring system according to an exemplary embodiment of the present invention.
FIG. 5 is an enlarged view of a portion of the frish layer in FIG. 4.
6 is an exemplary diagram illustrating 2-D particle distribution from an image photographed by a digital hologram sensor of a vacuum plasma monitoring system according to an embodiment of the present invention.
7 is an exemplary view of extracting a current pattern from an image photographed by a digital hologram sensor in a vacuum plasma monitoring system according to an embodiment of the present invention.
8 is a flowchart illustrating an operation flow of a vacuum plasma monitoring method according to an embodiment of the present invention.
9 is a flow chart showing the operation flow of the vacuum plasma monitoring method according to another embodiment of the present invention.

Hereinafter, a vacuum plasma monitoring system and a method thereof according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings. In this process, the thicknesses of the lines and the sizes of the components shown in the drawings may be exaggerated for clarity and convenience of explanation. In addition, terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to the intention or convention of a user or an operator. Therefore, the definitions of these terms should be made based on the contents throughout the specification.

1 is a block diagram of a vacuum plasma monitoring system according to an embodiment of the present invention, Figure 2 is a block diagram showing an example of a vacuum plasma monitoring system according to an embodiment of the present invention, Figure 3 is a present invention A block diagram illustrating an image processor in a vacuum plasma monitoring system according to an embodiment of the present invention.

As shown in FIG. 1, a vacuum plasma monitoring system according to an embodiment of the present invention includes a digital hologram sensor 20, a monitoring control unit 100, and an output unit 50, and the monitoring control unit 100 includes an image. It includes a processing unit 30 and the monitoring diagnostic unit 40.

The chamber 10 is a space provided to perform a semiconductor process, and includes a chuck (not shown) on which a wafer is seated, and the inside of the chamber 10 may be in a vacuum state or a plasma state according to a semiconductor process performed. .

As shown in FIG. 2, the chamber 10 may have at least one window for imaging the sheath structure near the chuck or wafer.

In detail, the chamber 10 may include at least one window for receiving or outputting laser light emitted from the digital hologram sensor 20 to be described later.

For example, when the digital hologram sensor 20 captures a hologram image in an off-axis manner, the chamber 10 may include a window 12 and an exterior of the chamber 10 for receiving a laser beam into the chamber 10. It may be provided with a window 11 for exporting.

On the other hand, when the digital hologram sensor 20 takes a hologram image in an on-axis manner, the chamber 10 receives the laser light into the chamber 10 and sends out the reflected light to the outside of the chamber 10. It can be provided with a window.

Such a window may be provided in various sizes at an appropriate position to photograph the chuck or the wafer in the chamber 10.

The digital hologram sensor 20 photographs the inside of the chamber 10 in a vacuum state or a plasma state to generate a hologram image.

The digital hologram sensor 20 includes a light source 21 for irradiating the laser light into the chamber 10, a beam expander 23 for expanding the beam width of the laser light irradiated from the light source 21, and a chamber 10. It may include a charge-coupled device (CCD) sensor 25 in which the light reflected by the particles is stored.

In this case, the reference light refers to light incident to the chamber 10 through the beam expander 23, the object light refers to light reflected by the particles inside the chamber 10, and the three-dimensional image is hologramed in the object light. Included in the form.

An interference pattern is formed between the reference light and the object light, which can be expressed by Equation 1 below.

Where R is the reference light, O is the object light, and R ^* and O ^* are the complex conjugate of the reference light and the object light, respectively.

The interference pattern between the reference light and the object light transmitted to the CCD sensor 25 is a two-dimensional holographic image, and the CCD sensor 25 converts the interference pattern into an electric signal and transmits the interference pattern to the image processing unit 30 of the monitoring controller 100. do.

The monitoring controller 100 monitors a vacuum state or a plasma state by using the hologram image input from the digital hologram sensor 20, and includes an image processing unit 30 and a monitoring diagnosis unit 40.

First, the image processor 30 includes an image restorer 39 to restore a hologram image input from the digital hologram sensor 20.

In detail, the image restorer 39 of the image processor 30 may digitally process the electric signal transmitted from the CCD sensor 25 to restore the 3D hologram image by a numerical method. Such a numerical reconstruction method may be expressed by Equation 2 below.

Here, '*' is a convolution operator, h _z is a free space impulse response function at the depth direction _z , and I _z is an image of an object reconstructed at the depth direction z.

Numerical reconstruction according to Equation 2 described above is a numerical operation method corresponding to hologram reconstruction of an optical method of reconstructing a three-dimensional image of an object by injecting a reference light into the hologram, and digital back propagation using diffraction theory. Fresnel transformation can be used.

Thereafter, the image processor 30 may calculate the particle distribution in the chamber 10 from the reconstructed hologram image, and extract a current pattern from the reconstructed hologram image.

To this end, the image processing unit 30, as shown in Figure 3, 1-D particle counter 31, 2-D particle counter 33, 3-D particle counter for calculating the particle distribution in the chamber 10 At least one of the 35 and the current pattern extractor 37 for extracting the current pattern in the chamber 10 may be included.

The image processor 30 calculates the particle distribution in the chamber 10 by using the 1-D particle counter 31, the 2-D particle counter 33, and the 3-D particle counter 35, and a current pattern. A detailed process of extracting the current pattern inside the chamber 10 using the extractor 37 will be described later.

The monitoring diagnosis unit 40 monitors whether the vacuum state or the plasma state is normal based on the particle distribution or the current pattern in the chamber 10 provided from the image processing unit 30.

In this case, the monitoring diagnosis unit 40 may monitor whether the vacuum state or the plasma state is normal by comparing the particle distribution or the current pattern in the chamber 10 with the particle distribution in the normal state or the current pattern in the normal state.

The output unit 50 outputs whether the vacuum state or the plasma state determined by the detection diagnosis unit 40 is normal. The output unit 50 may include a lamp (not shown) or a display panel (not shown) to indicate whether the vacuum state or the plasma state is normal.

4 is an exemplary diagram of a sheath structure using a digital hologram sensor of a vacuum plasma monitoring system according to an embodiment of the present invention, and FIG. 5 is an enlarged view of a portion of a frish sheath layer (B layer) in FIG. 4. It is also.

4A and 4B are exemplary views photographing a sheath structure in a vacuum state and a plasma state using the digital hologram sensor 20, respectively.

Referring to FIG. 4, it can be seen that the bright A layer is a sheath layer having a low particle density, and the dark B layer is a preshes layer having a high particle density.

In addition, referring to FIG. 4, it can be seen that the sheath structure of the vacuum state and the sheath structure of the plasma state are similar, and that the sheath structure of the plasma state can be derived from the sheath structure of the vacuum state.

In addition, it has already been experimentally confirmed that black particles shown in dark colors in FIGS. 4 and 5 have an electrically negative charge.

Referring to FIG. 5, it can be seen that the vacuum or plasma space is filled with black particles, and these black particles are connected to each other to form a circle or filament. In addition, since the negative charge that can be commonly applied to vacuum or plasma is electrons, it can be assumed that such black particles are electrons.

Hereinafter, the image processor 30 calculates the number of particles from the images shown in FIGS. 4 and 5 by using the 1-D particle counter 31 and the monitoring diagnostic unit 40 is in a vacuum or plasma state. Describe the process of monitoring whether or not.

In the image illustrated in FIG. 5, the particle number N may be calculated in units of pixels by using Equation 3 below for an arbitrary gray scale range (GR).

Here, GR means an arbitrary gray scale range which is set in advance, and it is preferable that it is set to a low value range in order to calculate the number of particles of dark color.

Since each pixel of the image has a different color according to the bit value (gray scale value), the number of particles corresponding to an arbitrary gray scale range may be calculated from the hologram image.

In this case, the LSB (Least Significant bit) in each pixel is determined through a test process, and the detailed description is omitted since the test process is generally performed by a company manufacturing a camera.

For example, when each pixel of the CCD sensor 25 is 8 bits, each LSB may include 178.7 (˜179) particles. If each pixel of the CCD sensor 25 is 12 bits, each LSB may include 10 particles.

If each pixel of the CCD sensor 25 is 8 bits, each pixel can be represented by a gray scale value of 0-255, and multiplying this gray scale value by 178.7 (~ 179) described above yields particles at that pixel. You can calculate the number. That is, the number of particles of each pixel may be calculated by Equation 4 below.

Where g _{(i, j)} is the gray scale value at any (i, j) pixel position and α represents the number of particles per LSB inherent to the CCD sensor 25. Therefore, the number of particles included in any gray scale range may be calculated by Equation 5 below.

On the other hand, the number of particles in the image transmitted from the CCD sensor 25 increases as the gray scale value increases.

However, the number of particles in the object region (that is, inside the chamber) captured by the CCD sensor 25 is reversed. That is, since there are many particles in the object region corresponding to the darkened portion of the CCD sensor 25, the gray scale value in the object region should use the value inverted from the gray scale value of the CCD sensor 25.

Therefore, the number of particles of each pixel for the object region may be calculated as in Equation 6 below.

은 g_(i,j)의 반전된 그레이 스케일 값을 의미한다. 그레이 스케일 값이 반전되었다는 의미는 아래의 수학식 7에 의해 정의될 수 있다. here

Means the inverted gray scale value of g _{(i, j)} . Meaning that the gray scale value is inverted may be defined by Equation 7 below.

의 그레이 스케일 값은 수학식 7의 (2)와 같이 정렬되며, (1)과 (2)의 관계는 (3)으로 표현될 수 있다. That is, the gray scale values of g are arranged as in Equation (1),

The gray scale values of are aligned as shown in Equation (2), and the relationship between (1) and (2) may be represented by (3).

값을 의미한다.Here, n denotes a bit per pixel of the CCD sensor 25 included in the digital hologram sensor 20, and A _i denotes a bit value (ie, gray scale value) at an arbitrary (i, j) pixel position. It means, g _{(i, j)} value of the above equation (4 ₎ . Likewise, B _i denotes the inverted bit value (ie, gray scale value) of A _i at any (i, j) pixel location,

Lt; / RTI >

의 그레이 스케일 값은 255에서 0까지 1 간격으로 감소하며 정렬된다.For example, if each pixel of CCD sensor 25 is 8 bits, the gray scale value of g is aligned in increments of 1 from 0 to 255,

The gray scale values of are aligned in decreasing order of one from 255 to zero.

That is, in the CCD sensor 25, the particle number of an arbitrary pixel is calculated by Equation 4, and the particle number of an arbitrary pixel for the photographed object region is calculated using the converted gray scale value as shown in Equation 6. Can be.

On the other hand, the particle information inside the chamber 10 in the vacuum state or the plasma state is generally photographed using the CCD sensor 25. However, in addition to the CCD sensor 25, other various imaging apparatuses such as a digital camera or a web camera can be used, and in this case, the number of particles can be calculated from the captured image. However, the number of particles per LSB will be determined through a separate process.

In summary, the image processor 30 may calculate the number of particles distributed in the object region, that is, distributed in the chamber 10, for an arbitrary gray scale range GR by using Equation 8 below.

Thereafter, the monitoring diagnosis unit 40 determines whether the vacuum state or the plasma state is normal based on the number of particles calculated by the image processor 30.

In this case, the monitoring diagnosis unit 40 compares the particle count calculated by the image processor 30 with reference particle numbers in a normal vacuum state or a plasma state as shown in Equation 9 below to determine whether the vacuum state or the plasma state is normal. You can check it.

Here, N ^r _e denotes the reference particle count calculated in a normal vacuum state or plasma state, N _e means the particle count calculated in the image processor 30, and T _1n indicates that the vacuum state or the plasma state is normal. It means the first acceptance criteria value that can be determined.

In addition, the first acceptance reference value T _1n may be set using an average and a standard deviation of the number of particles measured in a normal vacuum state or a plasma state.

That is, the monitoring diagnosis unit 40 compares the particle count calculated by the image processor 30 with the reference particle count in the normal state, and determines that the vacuum state or the plasma state is normal if the difference is within the first allowable reference value. On the other hand, it is judged that it is not normal if it exceeds the first acceptance standard.

Subsequently, the sensing diagnosis unit 40 may control the output unit 50 to display whether the vacuum state or the plasma state is normal.

Meanwhile, in Equation 9 Instead of N _e and N ^r _e , N and N ^r of Equation 3 may be used.

6 is an exemplary diagram illustrating 2-D particle distribution from an image photographed by a digital hologram sensor of a vacuum plasma monitoring system according to an embodiment of the present invention.

Hereinafter, a process in which the image processor 30 calculates the particle distribution using the 2-D particle counter 33 and a process in which the monitoring diagnosis unit 40 monitors whether the vacuum state or the plasma state is normal will be described with reference to FIG. 6. Will be explained.

The image processor 30 may calculate the particle distribution in the x-axis or y-axis direction from the hologram image captured by the digital hologram sensor 20 using Equation 10 below.

In this case, the size of an arbitrary image is expressed as m × n (pixels), and m and n mean total number of pixels in the x-axis and y-axis directions, respectively. Also, Δx and Δy correspond to variables for section separation in the x- and y-axis directions.

When calculating the number of particles in the x-axis direction, the image size for calculating the number of particles at the i-th position on the x-axis is determined as Δx _i × n to calculate the number of particles included in a given gray scale range GR. It is represented by N _i .

The number of particles at any j-th position in the y-axis direction can be calculated in the same way, which is represented by N _j .

As such, using Equation 10, 2-D particle distribution at an arbitrary (i, j) pixel position can be calculated.

FIG. 6 illustrates a 2-D particle distribution calculated by applying Equation 10 to the sheath structure image of the vacuum state illustrated in FIG. 4A. 6 is a particle distribution calculated in a state in which Δy is fixed to '1' for a gray scale range GR of '0-100' in the y-axis direction. Here, the gray scale range GR and the sizes of Δx and Δy for calculating the particle number distribution may be arbitrarily set.

In FIG. 6, the peak of each layer is denoted by 'a, b, c, d', and the position near the chuck or the wafer is denoted by 'e'. The thickness of each layer and the number of particles in each layer vary according to the process conditions, which can be used for plasma monitoring.

The monitoring diagnosis unit 40 determines whether the vacuum state or the plasma state is normal based on the particle distribution calculated by the image processor 30.

In this case, the monitoring diagnosis unit 40 compares an arbitrary interlayer thickness measured using the particle distribution calculated by the image processor 30 with the interlayer reference thickness in a normal vacuum state or a plasma state as shown in Equation 11 below. Depending on whether the difference is within the allowable thickness difference, it is possible to monitor whether the vacuum state or the plasma state is normal.

That is, the monitoring diagnosis unit 40 compares the interlayer thickness calculated by the image processor 30 with the interlayer reference thickness in a normal state, and determines that the vacuum state or the plasma state is normal when the difference value is within the allowable thickness difference. If it is out of the allowable thickness difference, it is determined that the vacuum state or the plasma state is not normal.

Where D ^r denotes an interlayer reference thickness between arbitrary peaks (b and c, a and b, a and f, etc.) in a normal vacuum state or plasma state, and D uses particle distribution calculated by the image processor 30. It means the interlayer thickness of any peak measured by, and T _d means the allowable thickness difference that can be determined that the vacuum state or the plasma state is normal.

In addition, the allowable thickness difference T _d may be set using the average and the standard deviation of the number of particles measured in the normal vacuum state or the plasma state, similar to the above-described first allowable reference value T _1n , and the process conditions and equipment It can be adjusted in various ways according to the characteristics.

On the other hand, the monitoring diagnostic unit 40 is the number of interlayer particles measured using the particle distribution calculated by the image processing unit 30 as shown in equation (12) below the interlayer reference particle number in a normal vacuum state or plasma state In comparison, it is possible to monitor whether the vacuum state or the plasma state is normal.

That is, the monitoring diagnosis unit 40 compares the number of interlayer particles calculated by the image processor 30 with the number of interlayer reference particles in a normal state, and when the difference value is within the second allowable reference value, the vacuum state or the plasma state is normal. If it is determined that the second limit value is out of the vacuum state or the plasma state, it may be determined that it is not normal.

In this case, the number of interlayer particles may be calculated by Equation 13 below.

Here, N ^r means the number of interlayer reference particles between any peaks (b and c, a and b, a and f, etc.) in a normal vacuum state or plasma state, and N is a particle calculated by the image processor 30. It means the number of inter-layer particles between peaks measured using the distribution, and T _2n means a second threshold value which can be judged that the vacuum state or the plasma state is normal.

The second acceptance reference value T _2n may be set using an average and a standard deviation of the interlayer particle number measured in a normal vacuum state or a plasma state.

Where a and b are positions of two peaks to be monitored, and N _j is the number of particles at any j th position between two peaks a and b on the y-axis.

Meanwhile, although the particle distribution in the y-axis direction is shown in FIG. 6, the particle distribution can be calculated similarly in the x-axis direction. Therefore, the monitoring diagnosis unit 40 may monitor the vacuum state or the plasma state by using the particle distribution in any section on the x-axis.

Next, the process of calculating the particle distribution by the image processor 30 using the 3-D particle counter 35 and the process of monitoring the normal state of the vacuum state or the plasma state will be described.

The image processor 30 may restore the image (x pixel × y pixel) in a predetermined depth direction from the hologram image captured by the digital hologram sensor 20 by applying the Fresnel converter method. Particle distribution of the image can be extracted.

The image processor 30 may calculate the 3-D particle distribution in the chamber 10 by applying the particle distribution in the z-axis direction.

Specifically, the z-axis direction is the lateral direction of the chamber 10, so that the image processor 30 collects the two-dimensional particle distribution in the lateral direction of the arbitrary chamber 10, and this is the entire chamber 10. Application in the lateral direction can collect the particle distribution in the three-dimensional space in the chamber 10.

The monitoring diagnosis unit 40 may monitor the vacuum state or the plasma state by using the 3-D particle distribution in the lateral direction of the part or the entire chamber 10 calculated by the image processor 30. To this end, the equations for calculating the number of particles described above may be applied.

7 is an exemplary diagram in which an image processor extracts a current pattern from a hologram image in a vacuum plasma monitoring system according to an embodiment of the present invention.

Finally, the image processor 30 uses the current pattern extractor 37 in FIGS. 4 and 5. A process of extracting a current pattern from the illustrated image and a process of monitoring whether a vacuum state or a plasma state is normal by the monitoring diagnosis unit 40 will be described.

The current pattern extractor 37 of the image processor 30 extracts an image of a partial region from the original hologram image photographed by the digital hologram sensor 20 or the hologram image reconstructed by the image processor 30. At this time, it is preferable that the size of the image to be analyzed is small for real-time analysis.

Thereafter, the image processor 30 extracts a current pattern by applying Equation 14 below to the extracted partial region image.

Here, G _p means a gray scale value of a pixel included in a preset gray scale range GR, and n means a bit of each pixel of the CCD sensor 25.

For example, when each pixel of the CCD sensor 25 is 8 bits, the pixels included in the set gray scale range GR have a value of '0', and the other pixels have a value of '255'. The image processor 30 may generate a current pattern by connecting pixels having a value of '0'.

Here, '0' and '255' correspond to black and white, respectively. Although the current pattern is shown in black, it can be expressed in other colors. That is, the current patterns may be formed in different colors in different gray scale ranges.

For example, FIG. 7 illustrates a current pattern extracted by applying a gray scale range of '80 to 90 'to the particle distribution illustrated in FIG. 5.

As shown in FIG. 7, it can be seen that the trajectory currents of the particles form various loops, which are interconnected to form a current network.

Since the current pattern when the vacuum state or the plasma state is normal and the current pattern when the state is not normal are different, the monitoring diagnosis unit 40 may monitor whether the vacuum state or the plasma state is normal by using Equation 15 below. Can be.

Here, G ^r (i, j) denotes a reference gray scale value at a specific pixel position constituting a current pattern of an image collected in a normal vacuum state or plasma state, and G (i, j) represents a current pattern extractor 37 Means a gray scale value at a specific pixel position constituting the extracted current pattern. T means a limit allowable value which can be determined that the vacuum state or the plasma state is normal, and may be set using the average and the standard deviation of the current pattern extracted in the normal vacuum state or the plasma state.

That is, the monitoring diagnosis unit 40 determines that the vacuum state or the plasma state is normal when the difference between G ^r (i, j) and G (i, j) is within the limit allowable value, and if the difference is outside the limit allowable value, Since the current flow is not normal, it is determined that the vacuum state or the plasma state is not normal.

In addition, the monitoring diagnosis unit 40 may determine that a failure occurs in the vacuum state or the plasma state when the ratio of all the target pixels of the pixel that exceeds the limit allowance exceeds a preset reference ratio.

For example, the monitoring diagnosis unit 40 may determine that a failure occurs in a vacuum state or a plasma state when a pixel that exceeds a threshold allowance exceeds 5% of all target pixels.

In this case, the reference ratio may be variously adjusted according to the complexity and density of the current pattern to be monitored, and by this monitoring operation, plasma failure monitoring on a particle basis becomes possible. In addition, when the gray scale value is set to a low value, the number of current trajectories is relatively small, thereby improving monitoring performance.

On the other hand, the monitoring diagnostic unit 40 may be utilized for monitoring and diagnosis using a pattern recognition technique for recognizing a complex current pattern. As the pattern recognition technique, computational intelligence models such as neural networks and fuzzy logic may be used.

That is, the monitoring diagnosis unit 40 models a pixel pattern constituting a current for the hologram image photographed in a vacuum or plasma state by using a neural network, and evaluates the performance of the developed model by the following equation (16). Alternatively, it is possible to monitor whether the plasma state is normal.

Here, RMSE ^r means root mean-square error (RMSE) for test data when a model is developed using current patterns collected in a normal vacuum or plasma state. RMSE refers to an error between the model predicted value of the input pattern of the current pattern and the output value of the input pattern constituting the test image.

T _e means an allowable error that can be determined that the vacuum state or the plasma state is normal, and can be adjusted in various ways according to the process conditions and equipment.

In this case, the monitoring diagnosis unit 40 may determine that the vacuum or plasma state is normal when the difference value of the RMSE with respect to the RMSE ^r is within 10%.

In the case of the 3-D current pattern determined at various positions of the chamber, multiple neural networks may be applied, and Equation 16 described above may be applied to each model prediction value. As a result of the application, the space outside the limit can be diagnosed as the location where the vacuum plasma failure occurred.

8 is a flowchart illustrating an operation flow of the vacuum plasma monitoring method according to an embodiment of the present invention, with reference to this will be described the specific operation of the present invention.

First, the monitoring control unit 100 receives a hologram image of the inside of the chamber 10 in a vacuum state from the digital hologram sensor 20 (S110).

Then, the image processing unit 30 of the monitoring control unit 100 calculates the number of particles from the input hologram image (S120).

At this time, the image processing unit 30 of the monitoring control unit 100 is at least one of the 1-D particle counter 31, the 2-D particle counter 33, the 3-D particle counter 35 on the hologram image as described above. You can calculate the number of particles by applying.

Thereafter, the monitoring diagnosis unit 40 of the monitoring control unit 100 compares the calculated particle number with the reference particle number in a normal vacuum state and checks whether the difference value is within an allowable reference value (S130).

If the difference value is within the allowable reference value, since the normal vacuum state, the monitoring control unit 100 receives a hologram image of the inside of the chamber 10 in the plasma state from the digital hologram sensor 20 (S140).

Thereafter, the image processing unit 30 of the monitoring control unit 100 calculates the number of particles from the input hologram image (S150).

At this time, as in step S130, the image processing unit 30 of the monitoring control unit 100 is at least one of the 1-D particle counter 31, 2-D particle counter 33, 3-D particle counter 35 on the hologram image. You can calculate the number of particles by applying.

The monitoring diagnosis unit 40 of the monitoring control unit 100 compares the calculated particle number with the reference particle number in a normal plasma state and checks whether the difference value is within an allowable reference value (S160), and the difference value is allowed. If it is within the reference value, the monitoring is terminated because it is a normal plasma state.

On the other hand, in step S130 and step S160, if the difference value between the calculated particle number and the reference particle number is outside the allowable reference value, since the vacuum state or the plasma state is not normal, the monitoring control unit 100 is the output unit 50 By controlling, the failure state of the vacuum state or the plasma state is output and displayed (S170).

9 is a flow chart illustrating an operation flow of the vacuum plasma monitoring method according to another embodiment of the present invention, with reference to this will be described a vacuum plasma monitoring method according to another embodiment of the present invention.

In the above-described embodiment, the image processing unit 30 of the monitoring control unit 100 calculates the number of particles from the hologram image, and the monitoring diagnosis unit 40 monitors whether the vacuum state or the plasma state is normal based on the case. Explained.

However, the image processing unit 30 of the monitoring control unit 100 may extract the current pattern from the hologram image, and the monitoring diagnosis unit 40 may monitor whether the vacuum state or the plasma state is normal.

First, the monitoring controller 100 receives a hologram image of the inside of the chamber 10 in a vacuum state from the digital hologram sensor 20 (S210).

Then, the image processing unit 30 of the monitoring control unit 100 extracts a current pattern from the input hologram image (S220).

In this case, the image processing unit 30 of the monitoring control unit 100 may extract a partial region of the holographic image and binarize to extract a current pattern.

Subsequently, the monitoring diagnosis unit 40 of the monitoring control unit 100 compares the gray scale value of the specific pixel position of the extracted current pattern with the reference gray scale value of the corresponding pixel position in a normal vacuum state, and the difference is a marginal allowable value. Check whether it is within (S230).

If the difference value is within the limit allowable value, since the normal vacuum state, the monitoring control unit 100 receives a hologram image of the inside of the plasma chamber 10 from the digital hologram sensor 20 (S240).

Thereafter, the image processing unit 30 of the monitoring control unit 100 extracts a current pattern from the input hologram image (S250).

At this time, as in step S220, the image processing unit 30 of the monitoring control unit 100 may extract a partial region of the holographic image and binarize to extract a current pattern.

The monitoring diagnosis unit 40 of the monitoring control unit 100 compares the gray scale value of the specific pixel position of the extracted current pattern with the reference gray scale value of the corresponding pixel position in a normal vacuum state and determines whether the difference is within the limit tolerance. If the difference value is within the limit allowable value, the monitoring is terminated because it is a normal plasma state.

On the other hand, in step S230 and S260, if the difference value between the gray scale value of the specific pixel position of the extracted current pattern and the reference gray scale value of the corresponding pixel position is outside the limit tolerance, the vacuum state or the plasma state is not normal. The monitoring control unit 100 controls the output unit 50 to output and display a failure state of a vacuum state or a plasma state (S270).

Meanwhile, in the present embodiment, a system and method for monitoring a vacuum state or a plasma state by using an image of a sheath structure in a vacuum state or a plasma state have been described, but the region to be photographed is not limited to the sheath structure. It may be possible to photograph and monitor particles or particle patterns present in the area.

In the present embodiment, the image processor 30 uses the 1-D particle counter 31, the 2-D particle counter 33, the 3-D particle counter 35, and the current pattern extractor 37, respectively. (10) In the case of calculating the particle distribution or the current pattern therein, for example, the above methods can be applied in various combinations of vacuum or plasma state monitoring.

In addition, the present embodiment has been described on the assumption that the black particles are electrons, but the same may be applied to the case where the black particles are not electrons.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is clearly understood that the same is by way of illustration and example only and is not to be taken by way of limitation, I will understand. Accordingly, the true scope of the present invention should be determined by the following claims.

10: chamber
11,12: window
20: digital hologram sensor
21: light source
23: beam expander
25: CCD sensor
100: supervisory control unit
30: image processing unit
31: 1-D counter
33: 2-D counter
35: 3-D counter
37: current pattern extractor
39: video restorer
40: monitoring diagnosis unit
50: Output section

Claims (22)

A digital hologram sensor for photographing the inside of the chamber in a vacuum state or a plasma state to generate a hologram image;
An image processor which calculates a particle distribution in the chamber from the hologram image generated by the digital hologram sensor; And
And a monitoring diagnosis unit for monitoring whether the vacuum state or the plasma state is normal based on the particle distribution in the chamber calculated by the image processor, wherein the monitoring diagnosis unit monitors whether the vacuum state is normal and then the plasma. Vacuum plasma monitoring system, characterized in that for monitoring the normal state of the state.
The vacuum plasma monitoring system of claim 1, wherein the digital hologram sensor photographs a sheath structure inside the chamber.
3. The vacuum plasma monitoring system according to claim 2, wherein the chamber has at least one window for receiving or outputting laser light emitted from the digital hologram sensor.
delete The method of claim 1,
The image processing unit may include a 1-D particle counter that calculates the total number of particles included in a preset gray scale range in the hologram image by applying the following first equation to the hologram image. .

... ... ... ... ... ... ... ... ... ... (One)
Where GR is a preset gray scale range, g _{(i, j)} is a gray scale value at any pixel location, and α is the number of particles per LSB inherent to the CCD sensor included in the digital hologram sensor.
The vacuum plasma monitoring system of claim 1, wherein the image processor comprises a 1-D counter that calculates the total number of particles included in the inner region of the chamber by applying the following second equation to the hologram image. .

... ... ... ... ... ... ... ... ... ... (2)
Where GR is the preset gray scale range,

Is an inverted gray scale value of g _{(i, j), which} is a gray scale value at any pixel location of the hologram image, and α is the number of particles per LSB inherent to the CCD sensor included in the digital hologram sensor.
The method of claim 6, wherein

The vacuum plasma monitoring system, characterized in that determined by the following equation.

... ... ... ... ... ... ... (3)
Where n is a bit per pixel of the CCD sensor included in the digital hologram sensor, A _i is a bit value at an arbitrary pixel position (ie, gray scale value g _{(i, j)} ), B _i is an arbitrary pixel Inverted bit value of A _{i at} the position (ie inverted gray scale value)

).
The method according to any one of claims 5 to 7,
And the monitoring diagnosis unit determines that the vacuum state or the plasma state is normal when the difference value between the total particle count and the reference particle count calculated by the image processor is within a first allowable reference value.
The method of claim 2,
The image processing unit may include a 2-D particle counter that calculates a particle distribution in the x-axis or y-axis direction included in a preset gray scale range in the hologram image by applying a fourth equation to the hologram image. Vacuum plasma monitoring system.

... ... ... ... ... ... ... ... ... ... (4)
Here, m and n are the total number of pixels in the x- and y-axis directions of the hologram image, Δx and Δy are variables for section separation in the x- and y-axis directions, and N _i is the i-th position in the x-axis direction. Number of particles, N _j is the number of particles at the j-th position in the y-axis direction.
The method of claim 9,
And the image processor calculates the thickness of each layer of the sheath structure using the particle distribution.
The method of claim 10,
And the monitoring diagnosis unit determines that the vacuum state or the plasma state is normal when the difference value between the interlayer thickness and the interlayer reference thickness calculated by the image processor is within an allowable thickness difference.
The method of claim 9,
And the image processor calculates the number of interlayer particles of the sheath structure by applying a fifth equation to the particle distribution.

... ... ... ... ... ... ... ... ... ... (5)
Where a and b are the positions of the two peaks to be monitored and N _j is the number of particles at any j th position between the two peaks a and b on the y-axis.
13. The method of claim 12,
And the monitoring diagnosis unit determines that the vacuum state or the plasma state is normal when the difference value between the interlayer particle number and the interlayer reference particle number calculated by the image processor is within a second acceptable reference value.
The method of claim 9,
The image processing unit is a vacuum plasma monitoring system comprising a 3-D counter for calculating the 3-D particle distribution by applying the particle distribution in the z-axis direction.
A digital hologram sensor for photographing the inside of the chamber in a vacuum state or a plasma state to generate hologram image data;
An image processor extracting a current pattern inside the chamber by using the hologram image data generated by the digital hologram sensor; And
And a monitoring diagnosis unit configured to monitor whether the vacuum state or the plasma state is normal based on the current pattern extracted by the image processing unit, wherein the monitoring diagnosis unit monitors whether the vacuum state is normal and then normalizes the plasma state. Vacuum plasma monitoring system, characterized in that for monitoring whether or not.
16. The vacuum plasma monitoring system according to claim 15, wherein the digital hologram sensor photographs the sheath structure inside the chamber.
16. The method of claim 15,
The image processing unit is a vacuum plasma monitoring system, characterized in that for extracting a current pattern from a part or all of the holographic image according to the sixth equation.

... ... ... ... ... ... ... ... ... ... (6)
Here, Gp is a gray scale value of a pixel included in a preset gray scale range, and n is a bit per pixel of a CCD sensor included in the digital hologram sensor.
18. The method of claim 17,
If the difference between the gray scale value at a specific pixel position constituting the current pattern extracted by the image processor and the reference gray scale value at the pixel position is within a limit tolerance value, the monitoring diagnosis unit may be in a normal state. Vacuum plasma monitoring system, characterized in that judging.
19. The method of claim 18,
And the monitoring diagnosis unit determines that the vacuum state or the plasma state is normal when the ratio of all pixels of the pixel that is outside the limit allowance is equal to or less than a reference ratio.
Receiving, by the monitoring controller, a first hologram image of the inside of the chamber in a vacuum state from the digital hologram sensor;
Calculating, by the monitoring controller, a first particle distribution in the chamber from the first hologram image;
Determining, by the monitoring controller, whether the vacuum state is normal based on the first particle distribution;
If it is determined that the vacuum state is normal, the monitoring control unit receiving a second hologram image of the inside of the chamber in the plasma state from the digital hologram sensor;
Calculating, by the monitoring controller, a second particle distribution in the chamber from the second hologram image; And
And the monitoring control unit determines whether the plasma state is normal based on the second particle distribution.
Receiving, by the monitoring controller, a first hologram image photographing the inside of the vacuum chamber from the digital hologram sensor;
Extracting, by the monitoring controller, a first current pattern from the first hologram image;
Determining, by the monitoring controller, whether the vacuum state is normal based on the first current pattern;
If it is determined that the vacuum state is normal, the monitoring control unit receiving a second hologram image of the inside of the chamber in the plasma state from the digital hologram sensor;
Extracting, by the monitoring controller, a second current pattern inside the chamber from the second hologram image; And
And the monitoring control unit determining whether the plasma state is normal based on the second current pattern.
22. The method according to claim 20 or 21,
And the first hologram image and the second hologram image are images of a sheath structure inside the chamber.

Images