TW202346720A - Deposition status monitoring method and substrate processing method for enabling early and easy estimation of the deposition status of a pump - Google Patents

Abstract

An object of the present invention is to provide a technique that enables early and easy estimation of the deposition status of a pump. The solution is a deposition status monitoring method which monitors the deposition status of deposits deposited on a pump connected to a processing container of a substrate processing apparatus that processes a substrate. The deposition status monitoring method includes: a step of acquiring the current value of a motor that rotates the rotating structure of a pump; a step of generating an attenuated waveform as the current value changes over time by supplying a processing gas to the processing container; a step of acquiring peak current values of a plurality of peaks, and bottom current values of a plurality of valleys with the same number of peaks; a step of calculating an estimated convergent current value at which the attenuation waveform of the current value converges by using the average of the peak current values of the plurality of peaks and the bottom current values of the plurality of valleys; and a step of estimating the deposition status of the deposits based on the estimated convergent current value.

Description

Stacking state monitoring method and substrate processing device

This case is about a stacking state monitoring method and a substrate processing device.

Substrate processing equipment used in semiconductor manufacturing uses a turbomolecular pump to exhaust processing gas in a processing container. During exhaust, the turbomolecular pump accumulates deposits such as reaction products generated during substrate processing and by-products that are not beneficial to substrate processing. Patent Document 1 discloses a technology in which a control device monitors a current value supplied to a turbomolecular pump to estimate the accumulation state of deposits in the turbomolecular pump.

In estimating the accumulation state, if the current value of the turbomolecular pump is small, the estimation accuracy will decrease. Therefore, the current value when the processing gas is introduced into the processing container and a load is applied to the turbomolecular pump is used. When the introduction of the processing gas starts, the current value supplied to the turbomolecular pump forms an attenuated waveform that increases rapidly and then repeats the amplitude and then gradually converges. [Prior technical literature] [Patent Document]

[Patent Document 1] Japanese Patent Application Publication No. 2021-179187

(The problem that the invention aims to solve)

This project provides a technology that can estimate the stacking state of a pump early and easily. (Means used to solve problems)

According to one aspect of the present invention, there is provided a stacking state monitoring method for monitoring the stacking state of deposits stacked on a pump of a processing container of a substrate processing apparatus connected to a substrate, including: (a) The process of obtaining the current value of the motor that rotates the rotating structure of the pump; (b) The step of generating an attenuated waveform in response to a temporal change in the current value by supplying a processing gas to the processing container; (c) The step of obtaining the peak current value of the plurality of mountain portions constituting the attenuation waveform and the bottom current value of the valley portions having the same number as the plurality of the aforementioned mountain portions; (d) The step of calculating the estimated convergence current value of the attenuation waveform convergence by averaging the acquired peak current values of the mountain portions and the bottom current values of the valley portions; and (e) A step of estimating the accumulation state of the deposit based on the estimated convergence current value. [Effects of the invention]

According to one form, the accumulation state of the pump can be estimated early and easily.

Hereinafter, the form used to implement this invention will be described with reference to the drawings. In each drawing, the same components are denoted by the same symbols, and repeated explanations may be omitted.

FIG. 1 is a schematic cross-sectional view showing an example of a substrate processing apparatus 1 according to an embodiment. As shown in FIG. 1 , the substrate processing apparatus 1 is an inductively coupled plasma (ICP) processing apparatus that performs various substrate processes on an FPD substrate (hereinafter referred to as a substrate G) formed of a glass material. . Examples of FPD manufactured by processing the substrate G include a liquid crystal display (LCD), electroluminescence (EL), and a plasma display panel (Plasma Display Panel: PDP). In addition, as the material of the substrate G, in addition to glass, synthetic resin, etc. may also be used.

The substrate G may be a support substrate with a circuit patterned on the surface or a support substrate without a circuit. The planar size of the substrate G may be a range of about 1800mm to 3400mm on the long side, and a range of about 1500mm to 3000mm on the short side. In addition, the thickness of the substrate G may be in the range of approximately 0.2 mm to 4.0 mm. Examples of substrate processing performed by the substrate processing apparatus 1 include film formation processing using a CVD (Chemical Vapor Deposition) method, etching processing, and the like. Hereinafter, the substrate processing apparatus 1 which performs a film formation process as a substrate process will be demonstrated as an example.

The substrate processing apparatus 1 is provided with a rectangular parallelepiped box-shaped processing container 10 . The processing container 10 is formed of metal such as aluminum or aluminum alloy. In addition, the processing container 10 may be formed into an appropriate shape according to the shape of the substrate G. For example, when the substrate G is a circular plate or an elliptical plate, the processing container 10 may also be preferably formed into a cylindrical shape or an elliptical cylindrical shape.

The processing container 10 is provided with a rectangular support frame 11 protruding to the inside of the processing container 10 at a predetermined position in the vertical direction. The dielectric plate 12 is supported in the horizontal direction by the support frame 11 . The processing container 10 is divided into an upper chamber 13 and a lower chamber 14 via a dielectric plate 12 . The upper chamber 13 forms an antenna chamber 13a inside. The lower chamber 14 accommodates the substrate G, and forms an internal space 14a inside for performing substrate processing.

The side wall 15 of the lower chamber 14 is provided with a carry-out inlet 17 that is opened and closed by a gate valve 16 . In the substrate processing apparatus 1, when the gate valve 16 is opened, the substrate G is carried in and out through the carry-out entrance 17 by a transport device (not shown).

In addition, the side wall 15 of the lower chamber 14 is grounded (connected to the ground potential) via the ground wire 18 . The square side walls 15 of the lower chamber 14 have an annular sealing groove 19 at the upper end. By arranging the sealing member 20 such as an O-ring in the sealing groove 19, the support frame 11 and the lower chamber 14 hermetically seal the internal space 14a.

The support frame 11 is formed of metal such as aluminum or aluminum alloy. In addition, the dielectric plate 12 is formed of ceramics such as alumina (Al ₂ O ₃ ) or quartz.

The inner side of the support frame 11 is connected to the support frame 11 and is composed of a plurality of elongated members. The shower head 21 that discharges the processing gas into the internal space 14 a also serves as a support beam to support the dielectric plate 12 . The dielectric plate 12 is supported on the top of the shower head 21 . The shower head 21 is formed of metal such as aluminum, and is preferably surface-treated by anodizing. A gas flow path 21a is formed in the horizontal direction inside the shower head 21. Moreover, the shower head 21 has a gas flow path 21a and a plurality of gas discharge holes 21b communicating with the lower surface (inner space 14a) of the shower head 21.

On the upper surface of the shower head 21 is a gas introduction pipe 22 connected to the gas flow path 21a. The gas introduction pipe 22 extends upward in the upper chamber 13 and penetrates the upper chamber 13 , and is connected to the gas supply part 23 provided outside the processing container 10 .

The gas supply part 23 has a gas supply path 24 coupled to the gas introduction pipe 22, and includes a gas supply source 25, a mass flow controller 26, and an on-off valve 27 in order from upstream to downstream of the gas supply path 24. In the film formation process, the process gas is supplied from the gas supply source 25 , the flow rate is controlled by the mass flow controller 26 , and the supply timing is controlled by the opening and closing valve 27 . This processing gas flows from the gas supply path 24 through the gas introduction pipe 22 into the gas flow path 21a, and is discharged to the internal space 14a through each gas discharge hole 21b.

A high-frequency antenna 28 is arranged in the upper chamber 13 forming the antenna chamber 13a. The high-frequency antenna 28 is configured by wiring an antenna wire made of conductive metal such as copper in a loop or spiral shape. Alternatively, the high-frequency antenna 28 may be provided with multiple loop-shaped antenna lines. The terminal of the high-frequency antenna 28 is connected to a power feeding member 29 extending upward in the upper chamber 13 .

The power supply member 29 has an upper end protruding to the outside of the processing container 10 , and the high-frequency power supply unit 30 is connected to the upper end. The high-frequency power supply unit 30 has a power supply line 30a, and the power supply line 30a is connected to the high-frequency power supply 32 via a matching device 31 that performs impedance matching. The high-frequency power supply 32 applies high-frequency power corresponding to the frequency of substrate processing (for example, 13.56 MHz) to the high-frequency antenna 28 . Thereby, the high-frequency antenna 28 forms an induced electric field in the lower chamber 14 .

Furthermore, the processing container 10 is provided with a platform 40 (mounting table) in the lower chamber 14 on which the substrate G carried in from the carry-out port 17 is placed. The platform 40 has a platform body 41, a base 42, a plurality of lifting pins 43, and a plurality of lifting pin lifting mechanisms 44. The substrate G carried into the lower chamber 14 is delivered to each lift pin 43 raised by each lift pin lifting mechanism 44, and is placed on the stage body 41 by lowering each lift pin 43.

The platform body 41 has a rectangular shape in plan view and has a mounting surface 411 having the same planar size as the substrate G. For example, the planar size of the mounting surface 411 may be a range of about 1800 mm to 3400 mm on the long side, and a range of about 1500 mm to 3000 mm on the short side.

A plasma processing space PCS is formed between the mounting surface 411 of the platform body 41 and the shower head 21 . In the plasma processing space PCS, the induced electric field formed by the high-frequency antenna 28 generates plasma to plasmaize the processing gas supplied from the shower head 21 to the internal space 14a. The substrate processing apparatus 1 causes a film-forming precursor (precursor) in the plasma generated in the plasma processing space PCS to be deposited on the substrate G. In addition, a metal plate can be used instead of the dielectric plate 12, and an induced electric field is formed across the metal plate to generate plasma. In this case, the support beam is not needed and the metal plate can also serve as a shower head.

In addition, the stage body 41 is formed of aluminum, aluminum alloy, etc., and is provided with a temperature adjustment mechanism for adjusting the temperature of the substrate G. For example, the temperature control mechanism includes a resistor heating wire 45 inside the stage body 41 , and a temperature control power supply unit 46 that supplies power to the heating wire 45 is provided outside the processing container 10 . Alternatively, the temperature adjustment mechanism may also include a flow path that circulates the refrigerant inside the platform body 41 and a cooler that supplies the refrigerant to the flow path (both are not shown). For example, when the substrate processing apparatus 1 performs substrate processing (film formation processing), the temperature of the mounting surface 411 of the stage 40 is adjusted to approximately 200° C. by heating with the heating wire 45 to maintain the temperature state. Alternatively, the substrate G may be heated by circulating a heated temperature-adjusting medium in the flow path.

The base 42 is made of insulating material and is disposed on the bottom wall 33 of the lower chamber 14 to support the platform body 41 . The base 42 has an opening at the bottom, and fixes and supports the platform body 41 in a state where the platform body 41 is separated from the bottom wall 33 . The base 42 may have a structure that is separable into a lower member that supports the platform body 41 and an upper member that surrounds the side surfaces of the platform body 41 . Furthermore, the stage 40 is provided with a bias power supply unit (not shown) that supplies high-frequency power to form a bias voltage for introducing plasma to the stage 40 side during substrate processing.

Furthermore, the substrate processing apparatus 1 has an exhaust port 33a for exhausting gas in the internal space 14a on the bottom wall 33 of the processing container 10, and includes an exhaust unit 50 connected to the processing container 10 via the exhaust port 33a. In addition, in FIG. 1 , one exhaust port 33 a and one exhaust part 50 are shown as an example, but the substrate processing apparatus 1 may be provided with a plurality of exhaust ports 33 a and exhaust parts 50 .

The exhaust port 33a is formed in a perfect circle shape and is provided between the side wall 15 of the processing container 10 and the platform 40. The diameter of the exhaust port 33a also depends on the size of the processing container 10. For example, it is ideally set in the range of about 200 mm to 400 mm. In this embodiment, it is set to 300 mm. In addition, the shape of the exhaust port 33a does not need to be a perfect circle, and may be formed into a shape corresponding to the arrangement position such as a semicircle.

Furthermore, the substrate processing apparatus 1 is provided with a plurality of guide plates 34 on the outer periphery of the stage 40 and between the plasma processing space PCS and the exhaust port 33a. Each guide plate 34 provides conductance to the process gas around the platform 40 to guide the exhaust direction. In addition, the guide plate 34 (and the side wall 15 of the lower chamber 14) is connected to the ground potential and functions as a counter electrode to the high-frequency power for biasing.

The exhaust unit 50 includes an exhaust pipe 51 connected to the exhaust port 33 a and an exhaust pipe provided in the exhaust pipe 51 to exhaust the processing gas (processing gas that does not contribute to the processing of the substrate) in the processing container 10 . Air mechanism 52. Moreover, the exhaust part 50 may be equipped with the exhaust net 35 for preventing components from falling in the connection part (or the exhaust port 33a) of the processing container 10 and the exhaust pipe 51.

The exhaust mechanism 52 depressurizes the internal space 14a of the processing container 10 by exhausting air. In addition, the exhaust mechanism 52 exhausts reaction products (deposits) generated during substrate processing together with the processing gas. Specifically, the exhaust mechanism 52 is provided with an APC (Automatic Pressure Control) valve 53 , a turbo molecular pump (TMP: Turbo Molecular Pump) 54 and a dry pump 55 in this order toward the downstream side in the flow direction of the processing gas of the exhaust pipe 51 . The exhaust mechanism 52 uses the dry pump 55 to pre-evacuate the inside of the processing container 10 and then uses the turbo molecular pump 54 to evacuate the inside of the processing container 10 . In addition, the exhaust mechanism 52 controls the pressure of the internal space 14a by adjusting the opening of the APC valve 53.

The operation of each component of the exhaust mechanism 52 (APC valve 53, turbo molecular pump 54, and dry pump 55) is controlled by the control unit 60. The control unit 60 is a control computer including one or more processors 61, a memory 62, an input/output interface (not shown), and an electronic circuit. In addition, the control unit 60 is connected to a user interface 65 through which the substrate processing device 1 can report information and the user can input information.

The processor 61 is one of a combination of a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), a circuit composed of a plurality of discrete semiconductors, etc. or plural. The memory 62 is an appropriate combination of volatile memory and non-volatile memory (for example, CD, DVD (Digital Versatile Disc), hard disk, flash memory, etc.). In addition, the user interface 65 can be applied to a monitor, a speaker, a warning light, a keyboard, a mouse or a touch panel, etc.

The memory 62 stores a program for operating the substrate processing apparatus 1, process conditions for substrate processing, and the like. The processor 61 reads the program from the memory 62 and executes it, thereby controlling each component of the substrate processing apparatus 1 . For example, in controlling the turbomolecular pump 54 , the control unit 60 calculates the target rotational speed of the motor 549 of the turbomolecular pump 54 so that the pressure in the processing container 10 can reach the target pressure, and outputs the calculated target rotational speed to Driver 56 (see Figure 2). Thereby, the driver 56 supplies electric power corresponding to the target rotation speed to the motor of the turbomolecular pump 54 to control the driving of the turbomolecular pump 54 .

FIG. 2 is a cross-sectional view schematically showing the turbomolecular pump 54 provided in the substrate processing apparatus 1 . The turbomolecular pump 54 is configured by fixing a base 541 and a casing 542 to each other, and accommodates a rotating structure 543 in a space inside the base 541 and a casing 542 . The rotating structure 543 has an umbrella-shaped pump rotor 544 and a transmission shaft 545 that supports the rotation center of the pump rotor 544 .

A plurality of rotary wings 544 a protruding in a direction orthogonal to the axial direction of the transmission shaft 545 (radially outward) are formed on the outer peripheral surface of the pump rotor 544 on the housing 542 side. On the other hand, a plurality of fixed blades 542a are formed on the inner peripheral surface of the housing 542, respectively arranged between the mutually adjacent rotating blades 544a. The rotary blade 544a and the fixed blade 542a form a turbopump region that sucks gas through the opening of the casing 542.

Furthermore, a cylindrical portion 544b having a spiral groove on the outer peripheral surface is formed on the base 541 side of the pump rotor 544 . On the other hand, in the base 541, the inner peripheral surface facing the cylindrical part 544b is sufficiently close to the cylindrical part 544b, and the gas flow path is narrowed. The cylindrical portion 544b and the inner peripheral surface of the base 541 form a spiral pump region that forms an exhaust port 546 that guides gas in the turbopump region to the base 541.

The transmission shaft 545 of the rotating structure 543 is supported in a non-contact manner by a plurality of magnetic bearings 547 provided on the base 541 . Each magnetic bearing 547 is equipped with an electromagnet and a displacement sensor, and the floating position of the transmission shaft 545 is detected by the displacement sensor. Furthermore, the turbomolecular pump 54 is provided with a rotation sensor 548 in the base 541 that detects the number of rotations of the transmission shaft 545 (the number of rotations per second).

Furthermore, the base 541 and the transmission shaft 545 are provided with a motor 549 that drives the transmission shaft 545 to rotate. The motor 549 is formed by a motor stator 549a on the base 541 side and a motor rotor 549b on the transmission shaft 545 side. By this motor 549, the rotating structure 543 rotates around the axis of the transmission shaft 545 at high speed. Thereby, the pump rotor 544 can induce the gas on the opening side in the order of the turbo pump region and the spiral pump region and discharge it from the exhaust port 546 .

The driver 56 receives the command of the target rotation speed from the control unit 60 and adjusts the electric power supplied to the motor 549 . At this time, the driver 56 detects the actual rotation speed through the rotation sensor 548 and performs feedback control, thereby causing the motor 549 to rotate until the actual rotation speed is approximately consistent with the target rotation speed. In addition, the control unit 60 receives information on the real current supplied to the motor 549 from the driver 56 or the current sensor 57 provided in the wiring, thereby monitoring the real current of the turbomolecular pump 54 . Then, the control unit 60 estimates the accumulation state of the reaction product (deposit) accumulated in the turbomolecular pump 54 based on the real current information.

FIG. 3 is a block diagram showing functional blocks of the control unit 60 for estimating the accumulation state of the turbomolecular pump 54 . As shown in FIG. 3 , a pump control unit 70 , a current acquisition unit 71 , a memory area 72 , a gas control unit 73 and a deposition state estimation unit 74 are formed inside the control unit 60 .

The pump control unit 70 instructs the driver 56 to execute the target rotation speed for exhausting the processing gas so that the inside of the processing container 10 can be set to the reduced pressure environment set in the substrate processing recipe. Thereby, the driver 56 supplies appropriate power to the motor 549 of the turbomolecular pump 54 to rotate the rotating structure 543 at a substantially constant speed at the target rotation speed. Furthermore, during substrate processing, the turbomolecular pump 54 continues to operate so that a predetermined reduced pressure environment can be maintained in the processing container 10 .

The current acquisition unit 71 acquires the real current information from the driver 56 or the current sensor 57 and stores the motor current data D in the memory area 72 by correlating the time information measured in the control unit 60 with the real current information. The memory area 72 is an area provided in the memory 62 .

The gas control unit 73 controls the operation of the gas supply unit 23 and supplies the processing gas into the processing container 10 . Furthermore, the gas control unit 73 outputs information on the supply status of the processing gas to the accumulation state estimation unit 74 . Examples of the supply state of the processing gas include timing of supply start, type of processing gas, supply amount (flow rate), and the like.

The accumulation state estimating unit 74 estimates the accumulation state of the deposits in the turbo molecular pump 54 based on the current supplied to the motor 549 of the turbo molecular pump 54 . Therefore, the accumulation state estimating unit 74 includes: a current analysis unit 75 that analyzes the motor current data D; an estimation unit 76 that estimates the accumulation state based on the analysis results of the current analysis unit 75; and a unit that urges the user to perform maintenance, etc. based on the estimation results. Notification control unit 77.

The current analysis unit 75 reads the motor current data D stored in the memory area 72 and extracts and processes information necessary for estimating the accumulation state from the motor current data D. Specifically, the current analysis unit 75 reads the motor current data D at the timing of starting the supply of the processing gas based on the information on the supply status of the processing gas from the gas control unit 73 . That is, in estimating the accumulation state, in order to improve the estimation accuracy, a large current value is used when the processing gas is introduced into the processing container 10 and a load is applied to the turbomolecular pump 54 .

FIG. 4 is a graph showing an example of the motor current data D stored in the memory area 72 . In addition, in the graph of FIG. 4 , the horizontal axis represents time, and the vertical axis represents the current value or the flow rate of the processing gas. As shown in Figure 4, the motor current data D is when the supply of the processing gas starts, leaving a certain time difference, so that the current value rises sharply. As the processing gas is supplied, the flow rate of the processing gas in the processing container 10 increases, thereby increasing the flow rate of the processing gas to be exhausted, and the load on the turbomolecular pump 54 that sucks the processing gas also increases. In addition, the motor current data D is formed into an attenuation waveform in which the current value rises sharply and then gradually converges while repeating the amplitude of falling and rising with time a plurality of times.

In addition, in the motor current data D in FIG. 4 , the peak (topmost) of each mountain portion and the bottom (bottom) of each valley (bottom) of the attenuated waveform are flat. This is to detect the current value in the driver 56 of the turbomolecular pump 54 or the current sensor 57 every predetermined sampling period. That is, the motor current data D in FIG. 4 is an interpolated waveform formed such that the current value changes flatly during the sampling period and the flatness of the current value changes suddenly every time sampling is performed.

The sampling period for sampling the real current in the driver 56 or the current sensor 57 is set to a time interval shorter than a half cycle of the attenuation waveform of the motor current data D. This sampling period is ideally set to an appropriate time interval by conducting experiments or simulations in advance. For example, a range of about 1 sec to 10 sec can be cited as the sampling period.

As mentioned above, the current analysis unit 75 uses the motor current data D when the current value increases as the processing gas is supplied to the processing container 10 . Here, the motor current data D draws an attenuation waveform whose amplitude repeats a plurality of times at the timing of starting supply of the processing gas. Therefore, if the current value after the attenuation waveform converges is to be extracted, it takes time to extract it. For example, it takes about 90 seconds for the motor current to stabilize into a stable and smooth waveform (that is, a constant value) after the supply of the processing gas is started. However, in the substrate processing of the substrate processing apparatus 1, there are few steps in which the same flow rate of processing gas is continuously supplied into the processing container 10 for 90 seconds or more. If the flow rate of the processing gas is changed before the attenuation waveform of the motor current data D converges, an inconvenience occurs in which the current value of a stable waveform cannot be obtained. The same applies not only to the flow rate but also when conditions such as the type of process gas are changed.

Therefore, the substrate processing apparatus 1 of this embodiment is configured to predict the current value at which the attenuation waveform converges even during a period when the motor current data D has an unstable attenuation waveform. Specifically, the current analysis unit 75 extracts two points each of the peak current value and the bottom current value of the amplitude of the attenuated waveform at the timing of starting the supply of the processing gas, and calculates the average value using the current values of the total four points.

FIG. 5 is an explanatory diagram showing the process of calculating an average value from the motor current data D. FIG. As shown in Figure 5, the attenuation waveform of the motor current data D is along the time elapsed direction, in accordance with the first mountain portion 101, the first valley portion 102, the second mountain portion 103, the second valley portion 104, and the third mountain portion. 105. The sequential amplitude of the third valley 106. Furthermore, if we only focus on each peak of the motor current data D, the attenuation waveform is basically in the order of the peak of the first peak 101, the peak of the second peak 103, the peak of the third peak 105, ... The current value decreases. However, the driver 56 or the current sensor 57 detects the current for each sampling period regardless of the timing of starting supply of the processing gas. Therefore, the height of the peak value (current value) between adjacent mountain portions depends on the sampling timing. ) will vary. Similarly, if we only focus on the valleys of the motor current data D, the attenuation waveform is basically in the order of the bottom of the first valley 102, the bottom of the second valley 104, the bottom of the third valley 106,... The current value rises. However, depending on the timing of sampling, the order of peak heights (current values) between adjacent valley portions may also differ in each valley portion.

The current analysis unit 75 of this embodiment extracts the peak current values of the two mountain portions (the first mountain portion 101 and the second mountain portion 103) immediately after the attenuation waveform of the motor current data D and the two valley portions. The bottom current value of (the first valley portion 102 and the second valley portion 104). The first mountain portion 101 is a wave when the current value increases significantly as the supply of the processing gas starts. The first valley portion 102 is adjacent to and continuous with the first mountain portion 101 , and is a wave when the current value drops significantly relative to the first mountain portion 101 . The second mountain portion 103 is adjacent to and continuous with the first valley portion 102, and is a wave when the current value increases significantly relative to the second valley portion 104. The second valley portion 104 is adjacent to and continuous with the second mountain portion 103 , and is a wave when the current value drops significantly relative to the second mountain portion 103 .

Then, the current analysis unit 75 calculates an average value (hereinafter referred to as estimated convergence current value ES) from the four extracted current values, and regards this estimated convergence current value ES as the current value when the attenuation waveform converges. In addition, the calculation of the average value of the four extracted current values may be an additive average or a multiplicative average.

Here, regarding the attenuation waveform of the current value of the turbomolecular pump 54, the applicant of the present application, through repeated experiments, estimated the convergence current by averaging the peak current values of a plurality of mountain portions and the bottom current values of the same number of valley portions. Value ES, confirm the converged current value that approximates the attenuation waveform. That is, the attenuation waveform of the current of the turbomolecular pump 54 as the supply of the processing gas starts is balanced by the load applied to the turbomolecular pump 54 and the reaction force (torque) applied by the driver to suppress the load. , so it can be said that the converged current value is used as the reference up and down amplitude. Therefore, the estimated converged current value ES that averages the peak current values of the plurality of peaks and the bottom current values of the same number of valleys is a converged current value that represents the attenuation waveform with sufficient accuracy. This means that even if the peak current value of the first peak portion 101 is lower than the peak current value of the second peak portion 103 due to a deviation in the sampling period of the current value, substantially the same estimated convergence current value ES can be obtained. Because the bottom current value of the first valley part 102 is higher than the bottom current value of the second valley part 104 in a manner that matches the peak current value of the first mountain part 101 and the peak current value of the second mountain part 103, the result is if Taking the average, it approximates the estimated convergence current value ES.

In addition, regarding the calculation timing of the estimated convergence current value ES, the current analysis unit 75 is set just after acquiring the set peak current values of the two mountain portions and the bottom current values of the two valley portions. For example, if the current acquisition unit 71 acquires current values up to the third mountain portion 105, the peak current values of the two mountain portions and the bottom current values of the two valley portions of the attenuated waveform can be extracted. Therefore, the current analysis unit 75 immediately calculates the estimated convergence current value ES from the four previous current values at the timing of obtaining the current value of the third mountain portion 105, thereby obtaining an approximate attenuation value in a sufficiently short time. The estimated converged current value ES is the current value after the waveform has converged.

In addition, the number of peaks from which the peak current value is extracted in the attenuation waveform is not limited to two, and may be three or more as long as the conditions of the processing gas such as the flow rate are changed. Of course, the number of valleys from which the bottom current value is extracted is not limited to two, as long as it is the same number as the number of mountains. The greater the number of peaks and troughs from which the estimated converged current value ES is calculated, the closer it can be expected to calculation of the estimated converged current value ES that is close to the post-converged current value of the attenuated waveform. Furthermore, the current analysis unit 75 does not need to use the peak or valley portion immediately after the start of the attenuation waveform. For example, the current analyzing unit 75 may use the peak current values of the second peak portion 103 and the third peak portion 105 and the second valley portion 104 . The estimated convergence current value ES is obtained with each bottom current value of the third valley portion 106 . Alternatively, the current analysis unit 75 may monitor the attenuation waveform of the motor current data D and calculate the estimated convergence current value ES by excluding mountain portions or valley portions where abnormal amplitude occurs. For example, when the peak current value of the second mountain portion 103 is abnormally low, the current analysis unit 75 does not extract the second mountain portion 103 and the second valley portion 104, but uses the first mountain portion 101 and the third mountain portion 105. Each peak current value and each bottom current value of the first valley portion 102 and the third valley portion 106 are sufficient. Furthermore, the current analysis unit 75 may use the bottom current value of the first valley portion 102 and the second peak current value when the value of the first mountain portion 101 is significantly larger than a predetermined value (current value obtained in the past). The estimated convergence current value ES is calculated from the peak current value of the portion 103, the bottom current value of the second valley portion 104, and the peak current value of the third mountain portion 105.

Returning to FIG. 3 , the estimating unit 76 estimates the accumulation state of the deposits accumulated in the turbomolecular pump 54 based on the flow rate of the processing gas obtained from the gas control unit 73 and the estimated convergence current value ES calculated by the current analysis unit 75 . The accumulation state may be a value quantified as the amount of deposits in the turbomolecular pump 54 , or may be a value converted during the continuous operation of the pump. For example, the control unit 60 stores in the memory area 72 map information or a function that correlates the flow rate of the process gas and the estimated convergence current value ES based on the information on the accumulation state of the deposits of the turbomolecular pump 54 . Then, upon receiving the flow rate of the processing gas and the estimated convergence current value ES, the estimating unit 76 refers to the memory area 72 to extract or calculate the accumulation state. As described above, since the estimated converged current value ES is a converged current value that is close to the attenuation waveform, the estimating unit 76 can estimate the accumulation state of the turbomolecular pump 54 with high accuracy.

In addition, the notification control unit 77 outputs the status information of the turbomolecular pump 54 to the user interface 65 based on the accumulation status estimated by the estimation unit 76, and notifies the user. For example, the notification control unit 77 displays the estimated accumulation state (amount of accumulation, continuous operation period, maintenance prediction time, etc.) as it is on the display unit of the user interface 65 as status information of the turbomolecular pump 54 . Alternatively, the notification control unit 77 may have a maintenance threshold value for comparison with the amount of deposits, and when the amount of deposits is equal to or above the maintenance threshold value, a notification is issued to urge the user to maintain the turbomolecular pump 54 . In addition, the estimated convergence current value ES itself may be treated as a numerical value indicating the accumulation state. For example, a maintenance threshold value may be set as the current value, and when the estimated convergence current value ES is equal to or higher than the current value set as the maintenance threshold value, the maintenance threshold value may be set. Supervise maintenance reports.

The substrate processing apparatus 1 of this embodiment is basically configured as above, and its operation (stacking state monitoring method) will be described below with reference to FIG. 6 . FIG. 6 is a flowchart showing the processing flow of the stacking state monitoring method.

The control unit 60 of the substrate processing apparatus 1 first opens the gate valve 16 and uses the transport device to transport the substrate G into the internal space 14 a through the carry-out entrance 17 as preparation for substrate processing. Then, the control unit 60 raises each lifting pin 43 of the lifting pin lifting mechanism 44 to receive the substrate G from the transport device, and lowers each lift pin 43 after the transport device retreats to place the substrate G on the platform 40 . Face 411.

After the substrate G is placed, the control unit 60 of the substrate processing apparatus 1 starts substrate processing, and in conjunction with this, starts the stacking state monitoring method. The control unit 60 first operates the exhaust mechanism 52 of the exhaust unit 50 to exhaust the gas in the internal space 14 a of the processing container 10 . At this time, the pump control unit 70 of the control unit 60 instructs the driver 56 to execute the target rotation number for depressurizing the processing container 10 to a predetermined internal pressure, thereby causing the turbomolecular pump 54 (motor) to operate under the current adjustment of the driver 56 549) Rotation drive (step S1). The motor 549 is continuously controlled to rotate at a target rotation number by the driver 56 to become constant.

During rotational driving of the turbomolecular pump 54 , the driver 56 or the current sensor 57 detects the actual current supplied from the driver 56 to the motor 549 . Then, the current acquisition unit 71 of the control unit 60 acquires the real current transmitted from the driver 56 or the current sensor 57 (step S2).

When the internal space 14a of the processing container 10 forms a predetermined reduced pressure environment, the gas control unit 73 controls the gas supply unit 23 to supply the processing gas into the processing container 10 (step S3). Thereby, the inside of the processing container 10 is in a state in which the flow rate of the processing gas sharply increases (see time point t0 in FIG. 5 ). The processing gas is discharged to the plasma processing space PCS through the gas discharge hole 21b of the shower head 21. Along with the ejection of the processing gas, the substrate processing apparatus 1 supplies high-frequency power of, for example, 13.56 MHz from the high-frequency power supply 32 to the high-frequency antenna 28, thereby forming a uniform induction in the plasma processing space PCS via the dielectric plate 12. electric field. By the induced electric field thus formed, the processing gas will be plasmatized in the plasma processing space PCS to generate high-density inductively coupled plasma. As a result, the substrate processing apparatus 1 can perform substrate processing (film formation processing) for forming a predetermined film on the substrate G.

In addition, the processing gas supplied to the plasma processing space PCS and which does not contribute to the processing of the substrate and the reaction products or by-products (deposits) generated by the substrate processing are sucked through the exhaust part 50, thereby The air is exhausted from the exhaust port 33a through the exhaust pipe 51.

In response to the supply of processing gas into the processing chamber 10 for this substrate processing, the current acquisition unit 71 of the control unit 60 acquires a real current in a decaying waveform that repeats the amplitude after the motor current rises sharply. Upon receiving the timing of starting supply of the processing gas from the gas control unit 73 , the accumulation state estimating unit 74 reads and monitors the motor current data D after the timing. Then, the current analysis unit 75 of the accumulation state estimating unit 74 extracts the peak current values of the two mountain portions (the first mountain portion 101 and the second mountain portion 103) and the two valley portions (the first valley portion) from the motor current data D. portion 102, second valley portion 104) (step S4).

The current analysis unit 75 obtains the peak current values of a set number (two) of mountain portions and the same number (two) of bottom current values of valley portions, and then calculates the estimated convergence current value ES by averaging them (step S5 ). For example, the current analysis unit 75 calculates the estimated convergence current value ES immediately after acquiring a set number of current values (see time point t1 in FIG. 5 ).

Then, the estimating unit 76 estimates the accumulation state of the deposits in the turbomolecular pump 54 based on the flow rate of the processing gas and the calculated estimated convergence current value ES (step S6). Thereby, the estimating unit 76 can accurately estimate the accumulation state of the deposits in the turbomolecular pump 54 before the attenuation waveform converges. Therefore, the control unit 60 can well recognize the accumulation state of the turbomolecular pump 54 before, for example, the flow rate of the processing gas for substrate processing is changed (see time point t2 in FIG. 5 ).

Furthermore, the estimating unit 76 compares the estimated amount of deposits with the maintenance threshold value to determine whether maintenance of the turbomolecular pump 54 is necessary (step S7). The estimating unit 76 advances to step S8 when the amount of deposits is equal to or greater than the maintenance threshold value (step S7: YES), and skips step S8 when the amount of deposits is less than the maintenance threshold value (step S7: NO). .

In step S8 , the notification control unit 77 notifies the user of information urging maintenance of the turbomolecular pump 54 through the user interface 65 . Thereby, the user can perform maintenance on the turbomolecular pump 54 at an appropriate time.

As described above, the substrate processing apparatus 1 and the accumulation state monitoring method of the present embodiment estimate the accumulation state of the deposits toward the turbomolecular pump 54 using the attenuation waveform of the current value supplied to the motor 549 of the turbomolecular pump 54 . For this reason, the control unit 60 can estimate the accumulation state within a short period of time since the supply of the processing gas to the processing container 10 is started. For example, even if the flow rate of the processing gas fluctuates according to the substrate processing, the control unit 60 can estimate the deposition state stably and accurately.

The technical ideas and effects of the present invention described in the above embodiments are described below.

The first aspect of the present invention is a deposition state monitoring method for monitoring the accumulation state of deposits accumulated in the pump (turbo molecular pump 54) of the processing container 10 of the substrate processing apparatus 1 connected to the processing substrate G, and has: (a) The process of obtaining the current value of the motor 549 that rotates the rotating structure 543 of the pump; (b) The process of generating an attenuated waveform in response to a time change in the current value by supplying the processing gas to the processing container 10; (c) Obtain the peak current value of the plurality of mountain portions (the first mountain portion 101 and the second mountain portion 103) constituting the attenuated waveform and the same number of valley portions (the first valley portion 102 and the second valley portion) as the plurality of mountain portions. The process of determining the bottom current value of part 104); (d) The process of calculating the estimated convergence current value ES for attenuation waveform convergence by averaging the peak current values of the plurality of mountain portions and the bottom current values of the plurality of valley portions; and (e) A step of estimating the accumulation state of the deposit based on the estimated convergence current value ES.

Based on the above, the accumulation state monitoring method is to calculate the estimated convergence current value ES using the attenuation waveform of the current value of the pump (turbomolecular pump 54), whereby the accumulation state of the pump can be estimated early and easily. That is, the average estimated convergence current value ES of the plurality of mountain portions (the first mountain portion 101 and the second mountain portion 103) and the plurality of valley portions (the first valley portion 102 and the second valley portion 104) of the attenuation waveform is The converged current value with respect to the attenuated waveform is sufficiently approximated. Therefore, the accumulation state monitoring method can estimate the accumulation state in a manner that is substantially the same as the converged current value without waiting for the attenuation waveform to converge. For example, the accumulation state can be well grasped before the flow rate of the processing gas is changed during substrate processing.

Furthermore, in the step (c), the peak current values of the plurality of mountain portions (the first mountain portion 101 and the second mountain portion 103) and the plurality of valley portions (the first valley portion 102) immediately after the start of the attenuation waveform are extracted. , the bottom current value of the second valley 104). Thereby, the accumulation state monitoring method can estimate the accumulation state without consuming time after the processing gas is supplied to the processing container 10 and an attenuation waveform is generated in the current value.

Moreover, the step of (d) is to obtain the peak current value of the set number of mountain portions (first mountain portion 101, second mountain portion 103) and valley portions (first valley portion 102, second valley portion 104) just after ) after the bottom current value. Thereby, the accumulation state monitoring method can estimate the accumulation state immediately after repeating the mountain portions and valley portions for a predetermined number, thereby shortening the time spent in estimation.

Furthermore, in the step (c), the peak current values of the two mountain portions (the first mountain portion 101 and the second mountain portion 103) immediately after the start of the attenuation waveform and the two valley portions (the first valley portion) are extracted. portion 102, second valley portion 104). Thereby, the accumulation state monitoring method can estimate the accumulation state in a shorter time after the processing gas is supplied to the processing container 10 .

Furthermore, in the step (c), the values of the valley portions (the first valley portion 102 and the second valley portion 104) adjacent to the mountain portions (the first mountain portion 101 and the second mountain portion 103) from which the peak current value is obtained are obtained. Bottom current value. Thereby, the accumulation state monitoring method can calculate the estimated convergence current value ES more accurately.

Furthermore, (f) is provided: a step of determining whether maintenance of the pump (turbomolecular pump 54) is necessary based on the accumulation state of the deposits estimated in the step (e). Thereby, the accumulation state monitoring method can appropriately notify the user of the maintenance timing of the pump based on the estimated accumulation state.

In addition, a pump (turbo molecular pump 54) is connected to the processing container 10 and exhausts the internal space 14a of the processing container 10, thereby introducing the processing gas supplied to the internal space 14a into the inside of the pump. Accordingly, the accumulation state monitoring method is configured such that even if reaction products (deposits) generated in the processing container 10 are accumulated on the pump that performs exhaust of the processing container, the accumulation state of the pump can be monitored stably.

Furthermore, the second aspect of the present invention is a substrate processing apparatus 1 for processing a substrate G, and includes a processing container 10 for processing the substrate G; and a pump (turbine molecular pump) connected to the processing container 10 for exhausting the internal space 14a of the processing container 10. Pump 54); and control part 60, The control unit 60 controls: (a) The process of obtaining the current value of the motor 549 that rotates the rotating structure 543 of the pump; (b) The process of generating an attenuated waveform in response to a time change in the current value by supplying the processing gas to the processing container 10; (c) Obtain the peak current values of the plurality of mountain portions (the first mountain portion 101 and the second mountain portion 103) constituting the attenuation waveform and the valley portions (the first and second mountain portions) adjacent to the plurality of mountain portions and having the same number as the plurality of mountain portions. The process of determining the bottom current value of the valley portion 102 and the second valley portion 104); (d) The process of calculating the estimated convergence current value ES for attenuation waveform convergence by averaging the peak current values of the plurality of mountain portions and the bottom current values of the plurality of valley portions; and (e) A step of estimating the accumulation state of the deposit based on the estimated convergence current value ES. In this case, the substrate processing apparatus 1 can estimate the accumulation state of the pump early and easily.

The substrate processing apparatus 1 and the stacking state monitoring method of the embodiment disclosed this time are all examples and are not restrictive. The embodiments can be modified and improved in various ways without departing from the scope of the accompanying patent application and its gist. The matters described in the above plural embodiments may be taken in other configurations to the extent that they are not inconsistent, and may be combined to the extent that they are not inconsistent.

The substrate processing device 1 and the stacking state monitoring method of the present invention can be applied to various types of devices regardless of whether or not plasma is used in substrate processing. For example, in any of Atomic Layer Deposition (ALD) devices, Capacitively Coupled Plasma (CCP), Inductively Coupled Plasma (ICP), Radial Line Slot Antenna (RLSA), Electron Cyclotron Resonance Plasma (ECR), and Helicon Wave Plasma (HWP) Applicable to all types of devices.

1:Substrate processing device 10: Handle the container 54:Turbo molecular pump 543:Rotation structure 549: Motor 101: 1st Mountain Division 102:1st Valley 103:The 2nd Mountain Division 104: 2nd Valley

[Fig. 1] is a schematic cross-sectional view showing an example of a substrate processing apparatus according to an embodiment. [Fig. 2] is a cross-sectional view schematically showing a turbomolecular pump provided in the substrate processing apparatus. 3 is a block diagram showing functional blocks of a control unit for estimating the accumulation state of a turbomolecular pump. [Fig. 4] is a graph showing an example of motor current data memorized in the memory area. [Fig. 5] is an explanatory diagram showing the process of calculating an average value from motor current data. [Fig. 6] is a flowchart showing the processing flow of the accumulation state monitoring method.

101: 1st Mountain Division

102:1st Valley

103:The 2nd Mountain Division

104: 2nd Valley

105:The 3rd Mountain Division

106:3rd Valley

D: Motor current data

ES: Estimated convergence current value

t0,t1,t2: time point

Claims (8)

A deposition state monitoring method for monitoring the accumulation state of deposits accumulated in a pump of a processing container of a substrate processing apparatus connected to a substrate, having: (a) The process of obtaining the current value of the motor that rotates the rotating structure of the pump; (b) A step of generating an attenuated waveform in response to a temporal change in the current value by supplying a processing gas to the processing container; (c) The step of obtaining the peak current value of the plurality of mountain portions constituting the attenuation waveform and the bottom current value of the valley portions with the same number as the plurality of the aforementioned mountain portions; (d) The step of calculating the estimated convergence current value of the attenuation waveform convergence by averaging the acquired peak current values of the mountain portions and the bottom current values of the valley portions; and (e) A step of estimating the accumulation state of the deposit based on the estimated convergence current value. The stacking state monitoring method according to claim 1, wherein in the step (c), a plurality of peak current values of the mountain portions and a plurality of bottom current values of the valley portions immediately after the start of the attenuation waveform are extracted. . The stacking state monitoring method according to Claim 2, wherein the step (d) is performed immediately after acquiring a set number of peak current values of the mountain portions and bottom current values of the valley portions. The stacking state monitoring method according to claim 3, wherein in the step (c), peak current values of two peaks immediately after the start of the attenuation waveform and bottoms of two valleys are extracted. current value. The stacking state monitoring method according to any one of claims 1 to 4, wherein in the step (c), the bottom current value of the valley portion adjacent to the mountain portion where the peak current value is obtained is obtained. . The accumulation state monitoring method according to any one of claims 1 to 4, further comprising (f): a step of determining whether maintenance of the pump is necessary based on the accumulation state of the accumulation estimated in the step (e). . The accumulation state monitoring method according to any one of claims 1 to 4, wherein the pump is connected to the processing container to exhaust the internal space of the processing container and to supply the gas to the internal space. The aforementioned processing gas is introduced into the inside of the pump. A substrate processing device is a substrate processing device that processes a substrate, and is characterized by: A processing container for processing the aforementioned substrate; A pump connected to the aforementioned processing container for exhausting the internal space of the aforementioned processing container; and control department, The aforementioned control section controls: (a) The process of obtaining the current value of the motor that rotates the rotating structure of the pump; (b) A step of generating an attenuated waveform in response to a temporal change in the current value by supplying a processing gas to the processing container; (c) The step of obtaining the peak current value of the plurality of mountain portions constituting the attenuation waveform and the bottom current value of the valley portions with the same number as the plurality of the aforementioned mountain portions; (d) The step of calculating the estimated convergence current value of the attenuation waveform convergence by averaging the acquired peak current values of the mountain portions and the bottom current values of the valley portions; and (e) A step of estimating the accumulation state of the deposit based on the estimated convergence current value.

Images