Plasma Processing System and Method for Manufacturing a Trained Model

Abstract

The purpose of the present disclosure is to provide a plasma processing system with which the optimum manufacturing conditions for a substrate for actual use can be recognized accurately. This plasma processing system (501) has a plasma processing device (101), a learning model creation device (70), and a manufacturing details inference device (80). In the plasma processing device (101), a Vdc sensor assembly (190) having a plurality of Vdc sensors (s19) is provided directly under the substrate for actual use or inside the substrate for actual use. The plurality of Vdc sensors (s19) is provided in correspondence with a plurality of sites on the formation surface of the substrate for actual use, and each is provided in order to measure the negative-side offset potential (Vdc) at the corresponding site on the substrate for actual use. The learning model creation device (70) creates a trained model (D5) on the basis of Vdc use data (D1), Vpp voltage data (D2), and manufacturing details data (D3).

Description

Technical Field

[0001] The present disclosure relates to a plasma processing system including a plasma processing apparatus that generates plasma by an RF power source.

Background Art

[0002] Plasma processing apparatuses that generate plasma by an RF (Radio Frequency) power source have been used as semiconductor manufacturing apparatuses such as RF plasma thin film manufacturing apparatuses and RF plasma processing apparatuses. As a typical apparatus in a conventional plasma processing apparatus, for example, a parallel plate reactive ion etching apparatus that generates plasma with an RF power source that becomes an alternating current of 13.56 MHz can be mentioned.

[0003] Further, in a conventional plasma processing apparatus, the potential distribution in the plane of the actual substrate to be manufactured (hereinafter abbreviated as "substrate in-plane potential distribution") has a characteristic of becoming complicated. Since the substrate in-plane potential distribution greatly affects manufacturing conditions such as etching conditions, it has been necessary to accurately recognize it. As a technique for measuring the substrate in-plane potential distribution, for example, there is a process monitoring device disclosed in Patent Document 1.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in a conventional process monitoring device, a dedicated substrate for monitoring is used instead of the actual substrate to be manufactured. For this reason, although a conventional process monitoring device is suitable for detecting the presence or absence of abnormalities in the substrate in-plane potential distribution, it is not suitable for optimizing manufacturing conditions such as etching conditions.

[0006] Therefore, even when data analysis processing is performed using a conventional process monitoring device, there is a problem in that it is impossible to recognize the optimum manufacturing conditions for the actual use substrate so that a desired in-plane potential distribution can be obtained when manufacturing the actual use substrate.

[0007] An object of the present disclosure is to provide a plasma processing system that solves the above problems and can accurately recognize the optimum manufacturing conditions for an actual use substrate so that a desired in-plane distribution of negative potential can be obtained when manufacturing the actual use substrate by a plasma processing apparatus.

Means for Solving the Problems

[0008] A plasma processing system according to the present disclosure is a plasma processing system having a plasma processing apparatus that generates plasma in a chamber and a data analysis apparatus, wherein the plasma processing apparatus is provided in the chamber and is connected to an AC power supply. A lower electrode, an upper electrode provided in the chamber opposite to the lower electrode and connected to a reference potential, an actual use substrate disposed on the lower electrode side among the upper electrode and the lower electrode, and a plurality of positions corresponding to a plurality of locations on the formation surface of the actual use substrate. A plurality of potential measurement sensors each for measuring the negative potential at the corresponding location on the actual use substrate, the actual use substrate being a substrate to be manufactured, and measurement potential data indicating a plurality of measurement potentials measured using the plurality of potential measurement sensors is provided to the data analysis apparatus. The data analysis apparatus includes an in-plane potential distribution data creation unit that obtains substrate-corresponding in-plane potential distribution data indicating an in-plane distribution of negative potential in the actual use substrate based on the measurement potential data, and analysis data reflecting the substrate-corresponding in-plane potential distribution data. And a data analysis unit that obtains relevance data indicating the relevance between the manufacturing content data indicating the manufacturing content of the actual use substrate when the measurement potential data is obtained.

Effects of the Invention

[0009] The plasma processing apparatus in the plasma processing system of the present disclosure has a plurality of potential measurement sensors provided corresponding to a plurality of locations on the formation surface of the actual use substrate, which is the substrate to be manufactured. Therefore, based on the measurement potential data indicating the plurality of measurement potentials measured using the plurality of potential measurement sensors, the in-plane distribution of the negative potential corresponding to the actual use substrate can be accurately recognized.

[0010] The data analysis apparatus in the plasma processing system of the present disclosure, after obtaining the substrate-corresponding in-plane potential distribution data based on the measurement potential data from the in-plane potential distribution data creation unit, obtains relevance data indicating the relevance between the analysis data reflecting the substrate-corresponding in-plane potential distribution data and the manufacturing content data by the data analysis unit.

[0011] Therefore, the plasma processing system of the present disclosure can recognize the optimal manufacturing conditions for the actual use substrate so that the desired in-plane distribution of the negative potential can be obtained during the manufacture of the actual use substrate by estimating using the relevance data.

[0012] The object, features, aspects, and advantages of the present disclosure will become clearer from the following detailed description and the accompanying drawings.

Brief Description of the Drawings

[0013]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Embodiments for Carrying Out the Invention

[0014] <Basic Technology> As a plasma processing apparatus that generates plasma by an RF power source, a parallel plate type reactive ion etching (Reactive Ion Etching) apparatus (hereinafter abbreviated as "RIE etching apparatus") that generates plasma by an RF power source that is a high-frequency AC power source of 13.56 MHz can be mentioned. Hereinafter, the principle of the RIE etching apparatus will be described.

[0015] FIG. 17 is an explanatory diagram schematically showing the cross-sectional structure of the plasma processing apparatus 200 of the basic technology. The plasma processing apparatus 200 shown in FIG. 17 is a RIE etching apparatus that generates plasma in the etching chamber 1 and performs etching processing.

[0016] Note that the basic structure of the plasma processing apparatus 200 shown in FIG. 17 is common to most of the basic structures of the plasma processing apparatuses 101 to 104 of Embodiments 1 to 4 described later.

[0017] The etching chamber 1 has a chamber inner housing 1B as a main component. Inside the chamber inner housing 1B, an anode electrode 2 serving as an upper electrode is provided above, and a cathode electrode 3 serving as a lower electrode is provided below. A part of the anode electrode 2 penetrates the upper surface of the chamber inner housing 1B and is provided outside the chamber inner housing 1B, and a part of the cathode electrode 3 penetrates the bottom surface of the chamber inner housing 1B and is provided outside the chamber inner housing 1B.

[0018] An RF power supply 12 is connected to the cathode electrode 3 via a variable capacitor 13, and a negative offset potential Vdc, which is the potential of the cathode electrode 3, is detected by a Vdc measuring voltmeter 15. Note that the RF power supply 12 is also connected to a ground potential 16. The variable capacitor 13 is provided to match the impedance required for RF discharge in the etching chamber 1 and the impedance of the RF power supply 12.

[0019] On the bottom surface of the chamber inner housing 1B, a gas discharge hole 4 is provided below the outer side of the peripheral portion of the cathode electrode 3. An etched substrate 10 serving as an actual use substrate is placed on the cathode electrode 3. Etching processing is performed above the etched substrate 10.

[0020] The anode electrode 2 is connected to a ground potential 16 serving as a reference potential. The anode electrode 2 also serves as a shower head SH for introducing etching gas. The shower head SH for introducing etching gas has a plurality of gas supply holes penetrating the anode electrode 2.

[0021] The inside of the etching chamber 1 is evacuated by a vacuum pump (not shown) from the gas discharge hole 4 until it once becomes a high vacuum region. Then, etching gas is introduced into the etching chamber 1 from a plurality of gas supply holes provided through the anode electrode 2 that also serves as the shower head SH for introducing etching gas.

[0022] In this state, by applying an RF voltage of 13.56 MHz from the RF power supply 12 between the anode electrode 2 and the cathode electrode 3, most of the etching gas becomes plasma. That is, most of the etching gas dissociates into plasma in the etching chamber 1.

[0023] As shown in FIG. 17, the space region in the etching chamber 1 is divided into a plasma region 5 and an ion sheath region 6. In the plasma region 5 that occupies most of the inside of the etching chamber 1, the etching gas is composed of positive ions 7 dissociated into plasma by RF discharge by the RF power supply 12, activated radicals 8 dissociated by the same RF discharge, and gas molecules 9 that did not dissociate into plasma. Here, the activated radicals 8 and the gas molecules 9 have a neutral charge.

[0024] The ion sheath region 6 exists below the plasma region 5 and is a region formed over the surface of the substrate 10 to be etched. The reason for the generation of the ion sheath region 6 is that when a 13.56 MHz RF power supply 12 is applied to the etching gas, the dissociated positive ions 7 are heavy in mass compared to the charge and thus cannot follow the 13.5 MHz high frequency, and the positive ions 7 existing in the plasma region 5 stay in place.

[0025] On the other hand, since the electrons 11 emitted from the dissociated positive ions 7 are light in mass compared to the charge, they start to reciprocate between the anode electrode 2 and the cathode electrode 3 following the 13.56 MHz RF current. However, the cathode electrode 3 is connected to the RF power supply 12, and a variable capacitor 13 for impedance matching is sandwiched between the cathode electrode 3 and the RF power supply 12. For this reason, the electrons 11 stay in the variable capacitor 13, and as a result, the cathode electrode 3 and the substrate 10 to be etched become negatively charged.

[0026] On the one hand, due to the reasons described above, the plasma region 5 lacks electrons 11 and is positively charged. When considering the offset potential V1 in the etching chamber 1, among the offset potential V1, the negative potential caused by the negative charges carried by the cathode electrode 3 and the substrate to be etched 10 is called the negative-side offset potential Vdc. The negative-side offset potential Vdc greatly dominates the etching conditions as will be described later.

[0027] A Vdc measuring voltmeter 15 is inserted between the cathode electrode 3 and the variable capacitor 13 for impedance matching, and the negative-side offset potential Vdc can be measured by the Vdc measuring voltmeter 15. Similarly, among the offset potential V1, the positive potential caused by the positive charges carried by the plasma region 5 is called the positive-side offset potential Vpp, and like the negative-side offset potential Vdc, it greatly dominates the etching conditions.

[0028] Therefore, a blocking capacitor 17 that blocks the RF current discharged from the RF power supply 12 between the etching chamber 1 and the anode electrode 2, and a Vpp measuring voltmeter 18 serving as an upper electrode voltmeter are inserted. Specifically, the blocking capacitor 17 is inserted between the upper surface of the inner housing 1B of the chamber and the Vpp measuring voltmeter 18, the Vpp measuring voltmeter 18 is connected to the anode electrode 2, and the anode electrode 2 is set to the ground potential 16.

[0029] Then, the upper electrode potential, which is the potential of the anode electrode 2 measured by the Vpp measuring voltmeter 18 serving as the upper electrode voltmeter, is measured. This upper electrode potential becomes the positive-side offset potential Vpp. In this way, the Vpp measuring voltmeter 18 measures the upper electrode potential as the positive-side offset potential Vpp. Here, the differential voltage between the positive-side offset potential Vpp and the negative-side offset potential Vdc is called the DC bias BV.

[0030] Figure 18 is a graph showing the details of the DC bias BV shown on the right side of Figure 17. In the figure, the Y-axis serving as the horizontal axis indicates the height direction of the plasma processing apparatus 200 shown in Figure 17, and the vertical axis indicates the voltage value based on 0V applied from the ground potential 16.

[0031] As shown in the figure, the AC waveform AW, which is the RF output of the RF power supply 12, shifts to the negative side from the plasma region 5 to the ion sheath region 6. Therefore, the offset potential V1, which is the intermediate potential between the maximum value and the minimum value of the AC waveform AW, shifts to the positive side higher than 0V in the plasma region 5, and shifts to the negative side at the location where the etched substrate 10 and the cathode electrode 3 in the ion sheath region 6 exist.

[0032] The positively shifted offset potential V1 in the plasma region 5 becomes the positive offset potential Vpp, and the negatively shifted offset potential V1 at the location where the etched substrate 10 and the cathode electrode 3 exist becomes the negative offset potential Vdc.

[0033] The DC bias BV is the difference value obtained by subtracting the negative offset potential Vdc from the positive offset potential Vpp described above. Hereinafter, an increase in this difference value is expressed as an increase in the DC bias BV, and a decrease in this difference value is expressed as a decrease in the DC bias BV.

[0034] FIG. 19 is an explanatory diagram schematically showing the etching phenomenon occurring in the ion sheath region 6. As shown in the figure, electrons 11 stay inside the etched substrate 10 and are negatively charged. The etched material 21 covered with the resist 20 except for the etching portion attracts the positive ions 7 dissociated by the plasma by the Coulomb force generated by the negatively charged etched substrate 10.

[0035] At this time, not only is the cation 7 attracted by the etching substance 21 itself, but during its movement, it also collides with the activated radical 8 and the neutral gas molecule 9 that has not been dissociated by the plasma, and plays a role in causing these to descend toward the etching substance 21. The neutral gas molecule 9 does not undergo a chemical reaction with the etching substance 21, but instead plays a role in shaving the etching substance 21 through a physical sputtering action. Also, the cation 7 dissociated by the plasma descends intensively toward the etching substance 21 due to the Coulomb force with the etched substrate 10, undergoes a chemical reaction, and evaporates.

[0036] The neutral activated radical 8 dissociated by the plasma also descends toward the etching substance 21 due to a collision with the cation 7, and similarly undergoes a chemical reaction and evaporates. In this way, the etching phenomenon proceeds within the ion sheath region 6, and it can be understood that the DC bias BV, which is the differential voltage between the positive offset potential Vpp and the negative offset potential Vdc, has a great influence on the etching characteristics.

[0037] Figures 20 and 21 are explanatory diagrams showing the influence of the DC bias BV on the etching characteristics. Figure 20 schematically shows the cross-sectional structure when etching is performed in a state where the DC bias BV is relatively low, and Figure 21 schematically shows the cross-sectional structure in a state where etching is performed in a state where the DC bias BV is relatively high.

[0038] As shown in Figures 20 and 21 respectively, the etched substrate 10 is placed on the cathode electrode 3, the etching substance 21 is deposited on the etched substrate 10, and a patterned resist 20 is provided on the etching substance 21.

[0039] The RIE etching apparatus of the basic technology shown in Figure 17 performs an etching process on the etching substance 21 using the resist 20 as a mask with the inside of the etching chamber 1 in a plasma state in the states shown in Figures 20 and 21.

[0040] As shown in FIG. 20, when the DC bias BV is in a relatively low state, the cations 7, activated radicals 8, and neutral gas molecules 9 do not descend vertically onto the material to be etched 21, and the etching reaction direction DE proceeds isotropically. Also, the etching rate with respect to the material to be etched 21 becomes slower.

[0041] On the other hand, as shown in FIG. 21, when the DC bias BV is in a relatively high state, the cations 7, activated radicals 8, and neutral gas molecules 9 descend vertically, the etching reaction direction DE proceeds anisotropically, and the cross-sectional shape of the material to be etched 21 becomes vertically anisotropic. Also, the etching rate with respect to the material to be etched 21 becomes faster.

[0042] Therefore, in the case of an RIE etching apparatus, it can be inferred that the etching process for the material to be etched 21 can be performed with higher accuracy when the DC bias BV is in a relatively high state. That is, in a thin film processing apparatus such as an RIE etching apparatus, it can be inferred that the shape processing such as etching can be performed with higher accuracy when the DC bias BV is in a relatively high state.

[0043] However, it is considered that there is an upper limit to the height of the DC bias BV. This is because increasing the DC bias BV means increasing the absolute value of the negative offset potential Vdc, and there is concern that the underlying material may be damaged.

[0044] Also, an RF plasma thin film manufacturing apparatus, for example, a parallel plate type plasma CVD (Chemical Vapor Deposition) apparatus that generates plasma with a 13.56 MHz RF power source 12, performs a film formation process based on a principle similar to that of an RIE etching apparatus. Hereinafter, the parallel plate type plasma CVD apparatus may be simply abbreviated as a "plasma CVD apparatus".

[0045] In a plasma CVD apparatus, in the plasma processing apparatus of the basic technology shown in FIG. 17, the etching chamber 1 functions as a film formation chamber, and the substrate to be etched 10 is replaced with the substrate for thin film deposition 22.

[0046] Figs. 22 and 23 are explanatory diagrams showing the influence of the DC bias BV on the film formation characteristics. Fig. 22 schematically shows the cross-sectional structure when the film formation process is performed with a relatively low DC bias BV, and Fig. 23 schematically shows the cross-sectional structure when the film formation process is performed with a relatively high DC bias BV.

[0047] As shown in Figs. 22 and 23 respectively, a substrate 22 for thin film deposition to be manufactured is placed on the cathode electrode 3, and after a lower thin film 23 is formed on the substrate 22 for thin film deposition, an upper thin film 24 is formed on the lower thin film 23.

[0048] That is, in the states shown in Figs. 22 and 23, the plasma CVD apparatus functions as a film formation chamber to turn the inside of the etching chamber 1 into a plasma state and performs a film formation process of forming the upper thin film 24 on the lower thin film 23.

[0049] That is, in the states shown in Figs. 22 and 23, the cations 7 and activated radicals 8 that have descended toward the substrate 22 for thin film deposition due to the DC bias BV cause a chemical reaction above the substrate 22 for thin film deposition, and the upper thin film 24 is formed on the lower thin film 23.

[0050] On the other hand, as shown in Figs. 20 and 21, in the RIE etching apparatus, the cations 7 and activated radicals 8 that have descended toward the substrate 10 to be etched due to the DC bias BV cause a chemical reaction with the substance 21 to be etched and evaporate, thereby performing an etching process.

[0051] As shown in Fig. 22, when the DC bias BV is in a relatively low state, the cations 7, activated radicals 8, and neutral gas molecules 9 do not descend vertically onto the lower thin film 23, and the film formation reaction direction DF proceeds isotropically. Therefore, the coverage of the upper thin film 24 with respect to the lower thin film 23 is good, and also, since the upper thin film 24 receives less physical impact from the cations 7, activated radicals 8, and neutral gas molecules 9 during the film formation of the upper thin film 24, the film quality of the upper thin film 24 is good.

[0052] On the one hand, as shown in Fig. 23, when the DC bias BV is in a relatively high state, the cations 7, activated radicals 8, and neutral gas molecules 9 descend vertically onto the lower thin film 23, and since the film formation reaction direction DF proceeds in the vertical direction, the coverage of the upper thin film 24 over the lower thin film 23 deteriorates. In addition, since the physical impact received by the upper thin film 24 during film formation from the cations 7, activated radicals 8, and neutral gas molecules 9 is large, the film quality of the upper thin film 24 deteriorates.

[0053] Therefore, in the case of a plasma CVD apparatus, it is presumed that the film formation process of the upper thin film 24 can be better performed when the DC bias BV is in a relatively low state. However, there is thought to be a limit to the lower limit of the DC bias BV.

[0054] In addition, in the case of a sputtering apparatus using RF plasma that performs a sputtering process which plays an important role in thin film formation, only the chemical reaction in plasma CVD changes to a physical reaction, and similar to dry etching and plasma CVD, the DC bias BV plays a major role in optimizing the film formation processing conditions.

[0055] Thus, it can be understood that the DC bias BV has important significance during thin film formation and shape processing. And as described above, the DC bias BV is the differential voltage between the positive offset potential Vpp and the negative offset potential Vdc.

[0056] Generally, the measured positive offset potential Vpp is several V to several tens of V, while the negative offset potential Vdc is several tens of V to several hundreds of V. And although both the positive offset potential Vpp and the negative offset potential Vdc are expected to have an in-plane distribution, in principle, since there are no potential disturbance factors in the etching chamber 1, it can be expected that the variation in the in-plane distribution of the positive offset potential Vpp at the boundary between the plasma region 5 and the ion sheath region 6 is small.

[0057] On the other hand, it is expected that the in-plane distribution of the negative offset potential Vdc will become relatively large. This is because, as shown in FIGS. 20 and 21, the system composed of the cathode electrode 3, the substrate 10 to be etched, the material 21 to be etched, and the resist 20 has relatively large potential disturbance elements. This point will be described in detail below.

[0058] FIG. 24 is an explanatory diagram showing that the in-plane distribution of the negative offset potential Vdc becomes large. FIG. 24 shows the laminated structure of the cathode electrode 3, the substrate 10 to be etched, and the material 21 to be etched after patterning the resist 20.

[0059] Actually, when an etching process is performed on the material 21 to be etched using the resist 20 as a mask for the structure shown in FIG. 24, the etching content is different between the central part 22c of the substrate of the substrate 10 to be etched and the end part 22e of the substrate.

[0060] For example, the etching rate for the material 21 to be etched on the central part 22c of the substrate is relatively slow, and the etching shape is a shape reflecting isotropic etching. The etching rate for the material 21 to be etched on the end part 22e of the substrate is relatively fast, and the etching shape is a shape reflecting vertical anisotropic etching.

[0061] The reason for this phenomenon is considered to be that the electric field is concentrated at the end part 22e of the substrate and the negative offset potential Vdc increases. This is the basis for estimating that the in-plane distribution of the negative offset potential Vdc affects the etching process content. Note that the increase in the negative offset potential Vdc means that the absolute value of the negative offset potential Vdc at the negative side level becomes larger.

[0062] Similarly, in the resist-dense region R20t, the etching rate is relatively slow and the etching shape becomes closer to isotropic. In the resist-sparse region R20s, the etching rate is relatively fast and the etching shape often becomes vertically anisotropic. The resist-dense region R20t is a region where the density of the resist 20 formed on the material to be etched 21 is relatively high, and the resist-sparse region R20s is a region where the density of the resist 20 formed on the material to be etched 21 is relatively low. That is, the resist-dense region R20t is a region where the covered portion of the resist 20 per unit area is larger compared to the resist-sparse region R20s.

[0063] The cause of the above-described phenomenon is called the Micro Loading Effect, and multiple factors have been pointed out. Among them, the following reasons are considered to be major ones. Silicon with a relative dielectric constant of about 12.0 is used as the constituent material of the substrate to be etched 10. On the substrate to be etched 10, etching materials 21 with various relative dielectric constants are formed, and it is considered that the patterned resist 20 with a relative dielectric constant of about 3.0 - 4.0 covers the etching material 21.

[0064] In the resist-dense region R20t, the capacitance locally decreases, and the negative charge remaining on the substrate to be etched 10 decreases, causing a decrease in the negative offset potential Vdc. Conversely, in the resist-sparse region R20s, the capacitance locally improves, and the negative charge remaining on the substrate to be etched 10 increases, so the negative offset potential Vdc increases. The reason for the occurrence of the micro-loading effect can be explained by the above-described phenomenon.

[0065] In addition to the above-described reasons, due to the variation in the negative offset potential Vdc near the substrate lifting pins 29 for transporting the substrate to be etched 10, which are provided along with the cathode electrode 3, phenomena where the etching rate and the etching shape are different also frequently occur. The reason will be described in detail below.

[0066] Figure 25 is an explanatory diagram schematically showing the transport system of the substrate 10 to be etched associated with the cathode electrode 3. Generally, the lifting pins 29 for the substrate for transporting the substrate 10 to be etched are made of an aluminum alloy whose surface is covered with aluminum oxide (Al2O3) like the cathode electrode 3. Therefore, in order to cut off the RF power supply 12 applied to the cathode electrode 3, an insulator 30 is sandwiched between the lifting pins 29 for the substrate and the support shaft 31 for the lifting pins.

[0067] The lower part of the insulator 30 is supported by the support shaft 31 for the lifting pins. The support shaft 31 for the lifting pins is lifted by a lifting motor 32, and the lifting motor 32 is driven by an AC power supply 33. In this way, the driving mechanism of the lifting pins 29 for the substrate is constituted by the AC power supply 33, the lifting motor 32, the support shaft 31 for the lifting pins, and the insulator 30. The lifting pins 29 for the substrate can be lifted and lowered by this driving mechanism.

[0068] When the substrate 10 to be etched is moved to the space above the cathode electrode 3 by a transport arm (not shown) when the lifting pins 29 for the substrate rise, it is supported by the lifting pins 29 for the substrate above the cathode electrode 3. Then, by lowering the lifting pins 29 for the substrate, the substrate 10 to be etched is placed on the cathode electrode 3.

[0069] Therefore, it can be easily understood that the amounts of electrons 11 staying in the cathode electrode 3 and the lifting pins 29 for the substrate are different, which becomes a factor in the variation of the negative offset potential Vdc.

[0070] From the above description, there are variations in the DC bias BV on the substrate 10 to be etched, and the optimization of this DC bias BV is necessary for optimizing the etching conditions. And it can be understood that most of the factors causing the variation of the DC bias BV are occupied by the complex in-plane distribution of the negative offset potential Vdc on the substrate 10 to be etched.

[0071] However, in the RIE etching apparatus, which is a parallel plate type dry etching apparatus of the basic technology, as shown in FIG. 17, the negative offset potential Vdc across the entire cathode electrode 3 is measured by the Vdc voltmeter 15 representing it, and there is a problem in that it is not configured to measure the in-plane distribution of the negative offset potential Vdc in the substrate 10 to be etched.

[0072] In order to solve the above problems, conventionally, as in the technology disclosed in Patent Document 1, a process monitoring device including an etching substrate equipped with various sensors is dedicatedly provided to measure the in-plane distribution of the negative offset potential Vdc.

[0073] However, as described above, as a conventional process monitoring device, a dedicated substrate for monitoring was used instead of the actual substrate to be manufactured. For this reason, it is limited to detecting the presence or absence of abnormalities in the potential distribution in the process monitoring device, and there is a problem that it is not suitable for optimizing manufacturing conditions such as etching conditions for the actual substrate.

[0074] In the embodiments described below, a plasma processing system and a method for manufacturing a learned model that solve the above problems are disclosed.

[0075] <Embodiment 1> FIG. 1 is an explanatory diagram schematically showing a cross-sectional structure of a main part of a plasma processing apparatus 101 included in a plasma processing system 501 according to Embodiment 1 of the present disclosure. The plasma processing apparatus 101 shown in FIG. 1 is a parallel plate type RIE etching apparatus that generates plasma in an etching chamber 1 and performs etching processing.

[0076] (Plasma processing system 501) FIG. 2 is an explanatory diagram schematically showing the configuration of the plasma processing system 500 according to Embodiment 1. As shown in the figure, the plasma processing system 500 includes a plasma processing apparatus 101, a learning model creation apparatus 70, and a manufacturing content inference apparatus 80 as main components.

[0077] The plasma processing apparatus 101 outputs Vdc data D1 and Vpp voltage data D2 to the learning model creation apparatus 70. The learning model creation apparatus 70, which is a data analysis apparatus, generates a learned model D5 based on the Vdc data D1, the Vpp voltage data D2, and the manufacturing content data D3 obtained from the outside, and provides it to the subsequent manufacturing content inference apparatus 80.

[0078] Based on the learned model D5, the manufacturing content inference apparatus 80 obtains, as inference data D7, data indicating the manufacturing content for the actual use substrate so that the in-plane distribution of the negative potential indicated by the desired in-plane distribution data D4 can be obtained.

[0079] The configurations and operations of the plasma processing apparatus 101, the learning model creation apparatus 70, and the manufacturing content inference apparatus 80 will be described in detail later.

[0080] (Plasma Processing Apparatus 101) Hereinafter, with reference to FIG. 1, the configuration of the plasma processing apparatus 101 according to Embodiment 1 will be described. Note that, in the plasma processing apparatus 200 shown in FIG. 17, the basic configuration including the exhaust system such as a vacuum pump using the gas discharge hole 4, the power supply system such as the RF power supply 12 and the variable capacitor 13 for impedance matching, and the transfer system of the actual use substrate corresponding to the substrate to be etched 10 such as the lift pins 29 for the substrate and the lift motor 32 is not shown in FIG. 1. That is, the plasma processing apparatus 101 according to Embodiment 1 also has the basic configuration including the exhaust system, the supply system, and the transfer system as described above, similar to the plasma processing apparatus 200.

[0081] The "actual use substrate" means the "substrate to be manufactured", and the "actual use substrate" is usually used during manufacturing. In the present disclosure, it is characterized in that the "actual use substrate" is also used as the "substrate to be the measurement object of a plurality of measured potentials". That is, in this specification, the "actual use substrate" includes the "substrate used during manufacturing" and the "substrate to be the measurement object of a plurality of measured potentials". Further, the "substrate to be manufactured" includes a "wafer used before mass production" and a "mass-produced product wafer".

[0082] The etching chamber 1 has a chamber inner housing 1B as a main component. Inside the chamber inner housing 1B, an anode electrode 2 serving as an upper electrode is provided above, and a lower electrode is provided below. A part of the anode electrode 2 penetrates the upper surface of the chamber inner housing 1B and is provided outside the chamber inner housing 1B, and a part of the lower electrode penetrates the bottom surface of the chamber inner housing 1B and is provided outside the chamber inner housing 1B.

[0083] In the plasma processing apparatus 101, similar to the cathode electrode 3 of the plasma processing apparatus 200 shown in FIG. 17, an RF power supply 12, which is a high-frequency AC power supply, is applied to the lower electrode via a variable capacitor 13. That is, the plasma processing apparatus 101 of Embodiment 1 also has a lower electrode corresponding to the cathode electrode 3 of the plasma processing apparatus 200.

[0084] In this way, in the plasma processing apparatus 101, the anode electrode 2 serving as the upper electrode and the lower electrode are provided to face each other in the etching chamber 1, and an actual use substrate (not shown) is disposed on the lower electrode side among the anode electrode 2 and the lower electrode. That is, the plasma processing apparatus 101 of Embodiment 1 also has an actual use substrate corresponding to the etched substrate 10 of the plasma processing apparatus 200. The actual use substrate is the substrate to be manufactured as described above.

[0085] Furthermore, a Vdc sensor assembly 190 is provided directly below the actual use substrate or inside the actual use substrate. The Vdc sensor assembly 190 has a plurality of Vdc sensors s19.

[0086] The plurality of Vdc sensors s19 are provided corresponding to a plurality of locations on the formation surface of the actual use substrate, and each is provided to measure the potential at the corresponding location on the actual use substrate.

[0087] The plurality of Vdc sensors s19 function as a plurality of potential measurement sensors for obtaining a plurality of measured potentials. Specifically, the potentials at the plurality of Vdc sensors s19 can be measured by a plurality of voltmeters or the like electrically connected to the plurality of Vdc sensors s19.

[0088] As a configuration in which the main part of the Vdc sensor assembly 190 is provided directly below the actual use substrate, the cathode electrode 3S of Embodiment 2 or the Vdc sensor substrate 19 of Embodiment 4 described later can be considered.

[0089] On the other hand, as a configuration in which a plurality of Vdc sensors s19 are provided in the actual use substrate, the sensor - built - in substrate 10S of Embodiment 3 described later can be considered.

[0090] That is, the Vdc sensor assembly 190 shown in FIG. 1 is a component of the plasma processing apparatus 101, including the cathode electrode 3S of Embodiment 2, the sensor - built - in substrate 10S of Embodiment 3, and the Vdc sensor substrate 19 of Embodiment 4 described later.

[0091] The Vdc sensor assembly 190 is mainly used for creating a learned model. In order to measure the in - plane distribution of the negative - side offset potential Vdc on the actual use substrate disposed in the ion sheath region 6 of the etching chamber 1, the plurality of Vdc sensors s19 are provided corresponding to a plurality of locations on the formation surface of the actual use substrate in plan view.

[0092] That is, the in - plane distribution of the negative - side offset potential Vdc on the actual use substrate can be comprehensively measured by the plurality of Vdc sensors s19 in the Vdc sensor assembly 190.

[0093] A plurality of measured potentials obtained from the Vdc sensor assembly 190 are obtained as Vdc sense data D11 described later, and the Vdc sense data D11 becomes measurement potential data. Further, position information of the actual use substrate corresponding to the plurality of measured potentials is obtained as position information data D12 described later. The position information data D12 is data that complements the Vdc sense data D11 which is measurement potential data. The combination of the Vdc sense data D11 and the position information data D12 becomes the Vdc data D1.

[0094] Each Vdc sensor s19 means a component capable of measuring, with a voltmeter, the potential difference between a corresponding location on the actual use substrate disposed in the ion sheath region 6 and the ground potential serving as the reference potential. Note that the type of the voltmeter is not limited.

[0095] The anode electrode 2 is connected to the ground potential 16 serving as the reference potential. The anode electrode 2 also serves as a shower head SH for introducing an etching gas. The shower head SH for introducing an etching gas has a plurality of gas supply holes (not shown) for supplying the etching gas into the etching chamber 1 through the anode electrode 2.

[0096] Note that instead of the shower head SH for introducing an etching gas, gas supply holes may be provided other than the anode electrode 2. For example, a modification such as providing gas supply holes on the side surface of the inner housing 1B of the chamber is conceivable.

[0097] The inside of the etching chamber 1 is once evacuated to a high vacuum region by a vacuum pump (not shown) through the gas discharge hole 4. Thereafter, the etching gas is introduced into the etching chamber 1 from the plurality of gas supply holes of the anode electrode 2 that also serves as the shower head SH for introducing an etching gas, and an RF voltage of 13.56 MHz is applied from the RF power supply 12 between the anode electrode 2 and the lower electrode, so that most of the etching gas becomes plasma.

[0098] Similar to the plasma processing apparatus 200, the plasma processing apparatus 101 has a blocking capacitor 17 and a Vpp measuring voltmeter 18. The upper electrode potential obtained from the anode electrode 2 can be measured as the positive offset potential Vpp by the Vpp measuring voltmeter 18. The Vpp voltage data D2 indicating the measured positive offset potential Vpp is output to the subsequent learning model creation device 70. This Vpp voltage data D2 becomes the upper electrode measurement data.

[0099] In principle, since it is considered that the in-plane distribution at the boundary between the plasma region 5 and the ion sheath region 6 is small, the positive offset potential Vpp is measured by the Vpp measuring voltmeter 18 in the same manner as the plasma processing apparatus 200 of the basic technology. That is, the positive offset potential Vpp is measured by measuring the potential difference between the blocking capacitor 17 and the ground potential 16 in the circuit from the inner housing 1B of the chamber through the blocking capacitor 17 to the ground potential 16 with the Vpp measuring voltmeter 18.

[0100] On the other hand, the in-plane distribution of the negative potential on the actual use substrate is obtained as a plurality of measured potentials obtained from the plurality of Vdc sensors s19 of the Vdc sensor assembly 190. That is, a complex in-plane portion of the negative offset potential Vdc on the actual use substrate can be recognized from the plurality of measured potentials.

[0101] In this way, the plasma processing apparatus 101, which is a main component of the plasma processing system 501 of the first embodiment, is configured. The Vdc data D1 includes Vdc sense data D11 indicating a plurality of measured potential data obtained from the Vdc sensor assembly 190 in the plasma processing apparatus 101, and position information data D12 indicating the position information of the plurality of measured potential data on the actual use substrate. This Vdc data D1 is output to the subsequent learning model creation device 70 as shown in FIG. 2.

[0102] The upper electrode potential measured by the voltmeter 18 for Vpp measurement in the plasma processing apparatus 101 becomes the positive offset potential Vpp, and Vpp voltage data D2 indicating the positive offset potential Vpp is output as upper electrode measurement data to the subsequent learning model creation apparatus 70.

[0103] (Learning model creation apparatus 70) FIG. 3 is an explanatory diagram schematically showing the configuration of the learning model creation apparatus 70 in the plasma processing system 500 of Embodiment 1.

[0104] As shown in FIG. 2, the learning model creation apparatus 70, which is a data analysis apparatus, receives the Vdc data D1 and the Vpp voltage data D2 from the plasma processing apparatus 101. Further, it receives the manufacturing content data D3 from the outside. The manufacturing content data D3 is data indicating the manufacturing content for the actual use substrate when the Vdc data D1 and the Vpp voltage data D2 are obtained.

[0105] Note that since the position information corresponding to the plurality of measurement potential data is known in advance, the position information data D12 in the Vdc data D1 may be supplied to the learning model creation apparatus 70 from the outside. On the other hand, the manufacturing content data D3 may be obtained from the plasma processing apparatus 101.

[0106] As shown in FIG. 3, the learning model creation apparatus 70 includes, as main components, a Vdc in-plane distribution creation unit 71, a DC bias in-plane distribution creation unit 72, a learning model creation unit 73, and a learning model recording unit 74.

[0107] The Vdc in-plane distribution creation unit 71, which is an in-plane potential distribution data creation unit, creates Vdc in-plane distribution data D10 indicating the in-plane distribution of the negative offset potential Vdc in the actual use substrate based on the Vdc data D1. This Vdc in-plane distribution data D10 becomes the substrate-corresponding in-plane potential distribution data indicating the in-plane distribution of the negative potential in the actual use substrate.

[0108] As described above, the Vdc sense data D11 of the Vdc data D1 is obtained from the Vdc sensor assembly 190 of the plasma processing apparatus 101. Further, the position information data D12 of the Vdc data D1 is obtained from the plasma processing apparatus 101 or externally. The position information data D12 is data indicating the position of the formation surface on the actual use substrate for each of the plurality of measured potentials indicated by the Vdc sense data D11.

[0109] Therefore, based on the Vdc data D1, the Vdc in-plane distribution creation unit 71 recognizes a plurality of measured potentials for which position information is recognized as a plurality of position-recognized measured potentials. Then, based on the plurality of position-recognized measured potentials, the Vdc in-plane distribution creation unit 71 generates Vdc in-plane distribution data D10 that becomes substrate-corresponding in-plane potential distribution data.

[0110] In this way, the Vdc in-plane distribution creation unit 71 functions as an in-plane potential distribution data creation unit that derives Vdc in-plane distribution data D10 indicating the in-plane distribution of the negative potential on the actual use substrate based on the Vdc data D1 including the Vdc sense data D11.

[0111] The DC bias in-plane distribution creation unit 72, which is a bias potential distribution creation unit, generates BV in-plane distribution data D20 indicating the in-plane distribution of the DC bias BV on the actual use substrate based on the Vdc in-plane distribution data D10 and the Vpp voltage data D2.

[0112] The DC bias BV, which is a bias potential, can be calculated as the difference value between the positive-side offset potential Vpp and the negative-side offset potential Vdc. Since the positive-side offset potential Vpp is a constant value, the BV in-plane distribution data D20 indicating the in-plane distribution of the DC bias BV is data that reflects the Vdc in-plane distribution data D10. This BV in-plane distribution data D20 becomes the analysis target of the learning model creation unit 73.

[0113] In this way, based on the Vpp voltage data D2 which is the upper electrode measurement data and the Vdc in-plane distribution data D10 which is the in-plane potential distribution data corresponding to the substrate, the DC bias in-plane distribution creation unit 72 calculates a plurality of DC biases BV which are the difference values between the upper electrode potential and the plurality of measurement potentials, and obtains BV in-plane distribution data D20 indicating the plurality of DC biases BV as analysis data. Here, the DC bias BV becomes the bias potential, and the BV in-plane distribution data D20 becomes the bias in-plane distribution data.

[0114] The learning model creation unit 73 receives the BV in-plane distribution data D20 obtained from the DC bias in-plane distribution creation unit 72 and the manufacturing content data D3. That is, the learning model creation unit 73 which is a related data creation unit receives the BV in-plane distribution data D20 which is the bias in-plane distribution data as analysis data.

[0115] As described above, the manufacturing content data D3 indicates the manufacturing content set in the plasma processing apparatus 101 when the Vdc data D1 and the Vpp voltage data D2 are obtained from the plasma processing apparatus 101. The manufacturing content data D3 includes gas type data D31, gas flow rate data D32, gas pressure data D33, RF output data D34, and process data D35.

[0116] The gas type data D31 is data indicating the gas type used for the etching process. There is at least one gas type. The gas flow rate data D32 is data indicating the supply flow rate of each of at least one gas satisfying the gas type indicated by the gas type data D31. The gas pressure data D33 is data indicating the etching pressure in the etching chamber 1. The RF output data D34 is data indicating the oscillation frequency of the RF power supply 12, the maximum rated output which is the input power, etc.

[0117] The process data D35 is data indicating the etching process details other than the data shown by the above-described data groups D31 to D34. As data indicated by the process data D35, for example, data indicating the temperature of the lower electrode, the inter-electrode distance between the anode electrode 2 and the lower electrode, the reflected wave output of the RF power supply 12, the achievable vacuum degree in the etching chamber 1 before introducing the etching gas, the exhaust valve opening degree at the gas discharge hole 4, etc. can be considered.

[0118] Thus, the manufacturing content data D3 including the data groups D31 to D35 indicates the manufacturing content including the etching process for the actual use substrate.

[0119] Note that the negative offset potential Vdc and the positive offset potential Vpp are important monitoring parameters in the plasma process as described with reference to FIGS. 20 to 24 in the section of the basic technology, but it is practically impossible to set them arbitrarily.

[0120] That is, the negative offset potential Vdc and the positive offset potential Vpp are affected by the manufacturing content data D3. For example, when the etching pressure indicated by the gas pressure data D33 is decreased, the mean free path of the electrons 11 becomes longer, and as a result, the negative offset potential Vdc tends to increase negatively. In order to set the in-plane distribution of the negative offset potential Vdc on the actual use substrate to an appropriate value, it is necessary to adjust the gas flow rate indicated by the gas flow rate data D32 and the inter-electrode distance between the anode electrode 2 and the lower electrode indicated by the process data D35.

[0121] Also, when the etching gas flow rate indicated by the gas flow rate data D32 is changed, the degree of dissociation of the plasma changes, so the negative offset potential Vdc changes. Therefore, in order to set the in-plane distribution of the negative offset potential Vdc on the actual use substrate to an appropriate value, it is necessary to adjust the etching pressure indicated by the gas pressure data D33 and the inter-electrode distance indicated by the process data D35. Note that the absolute value of the positive offset potential Vpp changes depending on various process parameters indicated by the manufacturing content data D3, but in principle, the in-plane distribution is assumed to be almost uniform.

[0122] The learning model creation unit 73 uses the BV in-plane distribution data D20 as input data and the manufacturing content data D3 as teacher data to perform machine learning, and creates a learned model D5 for estimating the manufacturing content of the actual use substrate corresponding to the BV in-plane distribution data D20.

[0123] That is, the learning model creation unit 73, which is a related data creation unit, analyzes the correlation between the BV in-plane distribution data D20 and the manufacturing content data D3, and creates a learned model D5 as correlation data indicating the correlation between the BV in-plane distribution data D20 and the manufacturing content data D3.

[0124] The data analysis unit is composed of the DC bias in-plane distribution creation unit 72, which is the bias potential distribution creation unit described above, the learning model creation unit 73, which is a related data creation unit, and the learning model recording unit 74 described later.

[0125] The data analysis unit uses the BV in-plane distribution data D20 reflecting the Vdc in-plane distribution data D10 as analysis data, analyzes the correlation with the manufacturing content data D3 indicating the manufacturing content for the embodiment when the Vdc data D1 and the Vpp voltage data D2 are obtained, and obtains a learned model D5 as correlation data indicating the correlation between the analysis data and the manufacturing content data D3.

[0126] Since the BV in-plane distribution data D20 reflecting the Vdc in-plane distribution data D10 and the manufacturing content data D3 are in a multivariate correlation relationship, by providing the BV in-plane distribution data D20 and the manufacturing content data D3 to the learning model creation unit 73, the learning model creation unit 73 can create a learned model D5 that is the result of performing multivariate analysis.

[0127] Note that the learning model creation unit 73 performs multivariate analysis by machine learning using AI technology. However, even if general multivariate analysis processing is performed without using AI technology, it is expected to create correlation data with a similarly high utilization value as the learned model D5.

[0128] Note that the learning model creation unit 73 may create the learned model D5 by changing conditions during etching using a set of Vdc sensor assemblies 190, or may create the learned model D5 using multiple sets of Vdc sensor assemblies 190 under different conditions respectively.

[0129] The learning model recording unit 74 records the learned model D5 created by the learning model creation unit 73. The learned model D5 recorded in the learning model recording unit 74 can be output to the subsequent manufacturing content inference device 80.

[0130] In this way, the learning model creation device 70 executes a method for manufacturing a learned model including the following steps (a) and (b).

[0131] Step (a): The Vdc in-plane distribution creation unit 71 obtains Vdc in-plane distribution data D10 indicating the in-plane distribution of the negative-side offset potential Vdc on the actual use substrate based on the Vdc data D1 including the Vdc sense data D11.

[0132] Step (b): Using the analysis data as input data and the manufacturing content data D3 indicating the manufacturing content of the actual use substrate when the analysis data is obtained as teacher data, the data analysis unit performs machine learning to manufacture a learned model D5 for estimating the manufacturing content of the actual use substrate corresponding to the analysis data. In Embodiment 1, the analysis data is the BV in-plane distribution data D20.

[0133] Here, the data analysis unit includes the DC bias in-plane distribution creation unit 72 and the learning model creation unit 73, and the analysis data is the BV in-plane distribution data D20 that reflects the Vdc in-plane distribution data D10.

[0134] By using the learned model D5 obtained in step (b) in the above-described method for manufacturing a learned model and performing inference with the manufacturing content inference device 80 described later, the optimal manufacturing conditions for the actual use substrate can be accurately obtained so that the desired in-plane distribution of the negative-side offset potential Vdc can be obtained during the manufacturing of the actual use substrate.

[0135] (Modification example) Although the learning model creation device 70 shown in FIG. 3 had a learning model creation unit 73 inside, a configuration in which a component corresponding to the learning model creation unit 73 is provided outside the learning model creation device 70 as an external learning model creation unit can be considered as a modification example of the learning model creation device 70. In this modification example, the learned model D5 created by the external learning model creation unit is received, and the learned model D5 is recorded by the internal learning model recording unit 74.

[0136] Therefore, in the modification example of the learning model creation device 70, the learning model creation unit 73 shown in FIG. 3 is omitted, the BV plane distribution data D20 is output to an external external learning model creation unit, and the learned model D5 is received from the external external learning model creation unit.

[0137] The external learning model creation unit uses the BV plane distribution data D20 as input data and the manufacturing content data D3 as teacher data to perform machine learning, creates a learned model D5 for estimating the manufacturing content of the actual used substrate corresponding to the BV plane distribution data D20, and outputs the learned model D5 to the modification example of the learning model creation device 70. That is, the learned model D5 is data obtained externally by performing machine learning using the BV plane distribution data D20 as input data and the manufacturing content data D3 as teacher data.

[0138] And the learning model recording unit 74 in the modification example of the learning model creation device 70 receives the learned model D5 from the outside and records the learned model D5.

[0139] In the modification example of the learning model creation device 70 described above, since it is not necessary to have the learning model creation unit 73 inside, the device configuration can be simplified.

[0140] (Manufacturing content inference device 80) FIG. 4 is an explanatory diagram schematically showing the configuration of the manufacturing content inference device 80 in the plasma processing system 501 of the first embodiment.

[0141] As shown in the figure, the manufacturing content inference device 80 includes an inference unit 81 and a processing content determination unit 82 as main components.

[0142] As shown in FIG. 2, the manufacturing content inference device 80 receives the learned model D5 created by the learning model creation device 70 and receives the desired in-plane distribution data D4 from the outside. The desired in-plane distribution data D4 includes the in-plane potential distribution data corresponding to the substrate desired during the manufacture of the actual use substrate.

[0143] The inference unit 81 receives the desired in-plane distribution data D4 and the learned model D5. The desired in-plane distribution data D4 includes the desired Vdc in-plane distribution data D10d indicating the in-plane distribution of the negative offset potential Vdc in the actual use substrate desired during manufacture.

[0144] Specifically, the desired in-plane distribution data D4 includes, in addition to the desired Vdc in-plane distribution data D10d, the desired BV in-plane distribution data D20d. The desired BV in-plane distribution data D20d is data indicating the in-plane distribution of the DC bias BV in the actual use substrate desired during manufacture. Note that the desired Vpp voltage data D2d is the sum of the plurality of negative offset potentials Vdc indicated by the desired Vdc in-plane distribution data D10d and the plurality of DC biases BV indicated by the desired BV in-plane distribution data D20d and becomes a constant value.

[0145] Also, as the desired in-plane distribution data D4, a combination of the desired Vdc in-plane distribution data D10d and the desired Vpp voltage data D2d may be used. This is because the desired BV in-plane distribution data D20d can be obtained from the difference between the positive offset potential Vpp indicated by the desired Vpp voltage data D2d and the plurality of negative offset potentials Vdc indicated by the desired Vdc in-plane distribution data D10d.

[0146] The inference unit 81 uses the learned model D5 to estimate, as inference data D7, data indicating the manufacturing content for the actual use substrate corresponding to the desired in-plane distribution data D4.

[0147] The inference data D7 includes optimal gas flow rate data D72, optimal gas pressure data D73, optimal RF output data D74, and optimal process data D75. The optimal gas flow rate data D72 is data indicating the optimal supply flow rate of each of at least one gas that satisfies the gas type indicated by the gas type data D31. The optimal gas pressure data D73 is data indicating the optimal etching pressure in the etching chamber 1. The RF output data D34 is data indicating the optimal RF output such as the oscillation frequency and the maximum rated output of the input power of the RF power supply 12.

[0148] The optimal process data D75 is data indicating the etching process content other than the data indicated by the above-described data groups D72 to D74. Note that since the gas type of the etching gas in the etching process has already been determined by the gas type data D31 in the manufacturing content data D3, the gas type data is not included in the inference data D7.

[0149] Thus, the inference data D7 including the data groups D72 to D75 is data including the etching process content for the actual use substrate and indicating the optimal manufacturing content in order to realize the in-plane distribution of the DC bias BV indicated by the desired in-plane distribution data D4.

[0150] The process content determination unit 82 receives the inference data D7 and determines the etching process content of the actual process performed by the plasma processing apparatus 101 based on the inference data D7.

[0151] Therefore, the plasma processing system 501 of the first embodiment can perform an etching process that satisfies a desired etching shape and etching rate by using the plasma processing apparatus 101 to perform an etching process with the content determined by the process content determination unit 82 on the actual use substrate.

[0152] (Effect) The plasma processing apparatus 101 used in the plasma processing system 501 of Embodiment 1 includes a Vdc sensor assembly 190. A plurality of Vdc sensors s19 included in the Vdc sensor assembly 190 are provided corresponding to a plurality of locations on the formation surface of the actual use substrate to be manufactured in a plan view. For this reason, the in-plane distribution of the negative potential of the actual use substrate can be accurately recognized from the Vdc sense data D11 indicating a plurality of measured potentials obtained from the Vdc sensor assembly 190.

[0153] Since the Vdc data D1 input to the Vdc in-plane distribution creation unit 71 of the learning model creation device 70 is obtained from a plurality of Vdc sensors s19 provided corresponding to the formation surface of the actual use substrate, the Vdc in-plane distribution creation unit 71 can create highly accurate Vdc in-plane distribution data D10.

[0154] The learning model creation device 70 in the plasma processing system 501 of Embodiment 1 includes, as a data analysis unit, a DC bias in-plane distribution creation unit 72, a learning model creation unit 73, and a learning model recording unit 74.

[0155] As described above, the DC bias in-plane distribution creation unit 72, which is a related data creation unit, obtains BV in-plane distribution data D20 as analysis data reflecting the Vdc in-plane distribution data D10 based on the Vdc in-plane distribution data D10 and the Vpp voltage data D2.

[0156] As described above, the learning model creation unit 73, which is a related data creation unit, creates learned model D5 as related data indicating the relevance between the BV in-plane distribution data D20 and the manufacturing content data D3.

[0157] In this way, in the plasma processing system 501 of Embodiment 1, the data analysis unit in the learning model creation device 70 obtains the learned model D5 as related data indicating the relevance between the BV in-plane distribution data D20, which is analysis data reflecting the Vdc in-plane distribution data D10, and the manufacturing content data D3. The learned model D5 is recorded in the learning model recording unit 74.

[0158] Therefore, by estimating using the learned model D5, the plasma processing system 501 according to Embodiment 1 can recognize the optimal manufacturing conditions for the actual-use substrate so that a desired in-plane distribution of the negative potential can be obtained during the manufacture of the actual-use substrate.

[0159] In addition, the learned model creation unit 73 of the learned model creation device 70 executes machine learning using AI technology to create a learned model D5 for estimating the manufacturing details of the actual-use substrate. Therefore, by performing inference using the learned model D5 created by the learned model creation unit 73, the optimal manufacturing conditions for the actual-use substrate can be accurately obtained so that a desired in-plane distribution of the negative offset potential Vdc can be obtained during the manufacture of the actual-use substrate.

[0160] Specifically, the inference unit 81 in the manufacturing details inference device 80 of the plasma processing system 501 according to Embodiment 1 can accurately obtain inference data D7 indicating the optimal manufacturing details for the actual-use substrate so that the in-plane distribution of the negative potential indicated by the desired in-plane distribution data D4 can be obtained during the manufacture of the actual-use substrate by using the learned model D5. The optimal manufacturing conditions for the actual-use substrate are defined by this inference data D7.

[0161] The plasma processing system 501 according to Embodiment 1 employs the algorithms up to the creation of the learned model D5 in the learned model creation device 70 described above and the algorithms up to obtaining the inference data D7 in the manufacturing details inference device 80. Therefore, the plasma processing system 501 can significantly reduce the number of times of the etching condition optimization experiment and can reduce the number of actual-use substrates required for the etching condition optimization to about one to several compared with the conventional etching condition optimization technology. In addition, it is possible to simplify the in-plane distribution measurement and cross-sectional microscope observation required for the etching characteristic evaluation.

[0162] The plasma processing system 501 of Embodiment 1 includes a learning model creation device 70 shown in FIG. 3 and a manufacturing content inference device 80 shown in FIG. 4. Therefore, in the plasma processing system 501, the number of procedure times of in-plane distribution measurement and cross-sectional shape observation that require shape step measurement, optical film thickness measurement, or cross-sectional electron microscope observation, which are essential for cross-sectional shape observation, can be significantly reduced, and can be made unnecessary depending on the situation.

[0163] In particular, the cross-sectional electron microscope observation performed during film thickness measurement or cross-sectional shape observation requires electron microscope observation after cross-sectional precision processing by focused ion beam (FIB) processing after cutting at an arbitrary location, so it requires a great deal of time and man-hours. In addition, multiple locations, more than several points, within a single substrate are required for cross-sectional electron microscope observation.

[0164] In the plasma processing system 501 of Embodiment 1, since the above procedures can be reduced, simplification of the processing content of process optimization for obtaining the inference data D7 can be realized.

[0165] In the plasma processing system 501 of Embodiment 1, an AI technology based on machine learning is used for the learning model creation unit 73 of the learning model creation device 70 and the inference unit 81 of the manufacturing content inference device 80. However, as described above, it is not necessary to use machine learning, and other analysis techniques may be used.

[0166] For example, in a data analysis device corresponding to the learning model creation device 70, as a data analysis unit corresponding to the DC bias in-plane distribution creation unit 72 and the learning model creation unit 73, multivariate analysis by the response surface method that requires careful setting of experimental parameters in advance may be used.

[0167] That is, the data analysis unit may record, as relevance data, a database based on past experimental examples showing the relationship between the correct value of the in-plane distribution of the negative offset potential Vdc and the manufacturing content feature amount composed of data indicating the manufacturing content. This relevance data becomes data to replace the learned model D5. Further, the data analysis unit may create data corresponding to the desired in-plane distribution data D4, that is, the correct amounts of the in-plane distributions of the positive offset potential Vpp and the negative offset potential Vdc in the actually used substrate.

[0168] In this case, the inference unit 81 of the manufacturing content inference device 80 can call the data table of the database created by the data analysis unit and predict the output value corresponding to the inference data D7 by mathematically complementing the values in the data table.

[0169] (Processing circuit) FIG. 5 is a block diagram showing the configuration of a processing circuit 90 corresponding to the components 71 to 73 in the learning model creation device 70 and the components 81 and 82 in the manufacturing content inference device 80 in the plasma processing system 501. The functions of the above-described components 71 to 73, 81, and 82 are realized by the processing circuit 90 shown in FIG. 5. That is, the processing circuit 90 functions as a circuit including the components 71 to 73, 81, and 82. Note that the component 71 includes a Vdc in-plane distribution creation unit 71V in the learning model creation device 70B of Embodiment 7 described later.

[0170] When the processing circuit 90 is dedicated hardware, the processing circuit 90 is, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an ASIC (Application Specific Integrated Circuit), an FPGA (Field-Programmable Gate Array), or a circuit combining these. The components 71 to 73, 81, and 82 may be individually realized by a plurality of processing circuits or may be collectively realized by one processing circuit.

[0171] FIG. 6 is a block diagram showing another configuration example of the processing circuits corresponding to the components 71 to 73, 81, and 82. As shown in the figure, the processing circuit 90 includes a processor 91, a memory 92, and a bus 96 serving as a data transfer path between the processor 91 and the memory 92. By the processor 91 executing a program stored in the memory 92, the functions of the components 71 to 73, 81, and 82 are realized. For example, by the processor 91 executing software or firmware described as a program, the functions of each of the components 71 to 73, 81, and 82 are realized. That is, the learning model creation device 70 and the manufacturing content inference device 80 include a memory 92 that stores a program, a processor 91 that executes the program, and a bus 96 for data transfer between the memory 92 and the processor 91.

[0172] Note that the program causes a computer to execute the processing procedures or processing methods of the functions of the components 71 to 73, 81, and 82 respectively.

[0173] As the processor 91, for example, a central processing unit, a processing device, an arithmetic device, a microprocessor, a microcomputer, a DSP (Digital Signal Processor), etc. can be considered. As the memory 92, for example, a non-volatile or volatile semiconductor memory such as a RAM (Random Access Memory), a ROM (Read Only Memory), a flash memory, an EPROM (Erasable Programmable Read Only Memory), an EEPROM (Electrically Erasable Programmable Read Only Memory), etc. can be considered. Further, as the memory 92, any storage medium to be used in the future, such as a magnetic disk, a flexible disk, an optical disk, a compact disk, a mini disk, a DVD, etc., may be used.

[0174] Further, the learning model recording unit 74 in the learning model creation device 70 can be realized by a memory 92 or an external storage device (not shown) connected to a bus 96. The data D1 for Vdc, the Vpp voltage data D2, the manufacturing content data D3, the desired in-plane distribution data D4, and the inference data D7 are classified into data stored in the memory 92 or an external recording device (not shown). Similarly, the process data D6 used by the learning model creation device 70B in Embodiment 7 to be described later is also data stored in the memory 92 or an external recording device (not shown).

[0175] Some of the functions of the above-described components 71 to 73, 81, and 82 may be realized by dedicated hardware, and some of the other functions may be realized by software or firmware. Thus, the processing circuit 90 realizes the above-described functions by hardware, software, firmware, or a combination thereof.

[0176] <Embodiment 2> FIG. 7 is an explanatory diagram schematically showing a cross-sectional structure of a main part of a plasma processing apparatus 102 included in a plasma processing system 502 according to Embodiment 2 of the present disclosure. The plasma processing system 502 according to Embodiment 2 has a configuration in which the plasma processing apparatus 101 is replaced with the plasma processing apparatus 102 in FIG. 2.

[0177] Similar to the plasma processing apparatus 101 according to Embodiment 1, the plasma processing apparatus 102 shown in FIG. 7 is a parallel plate type RIE etching apparatus that generates plasma in an etching chamber 1 and performs etching processing.

[0178] (Plasma Processing Apparatus 102) Hereinafter, with reference to FIG. 7, the same components as those of the plasma processing apparatus 101 shown in FIG. 1 are denoted by the same reference numerals, and the description thereof will be omitted as appropriate, and the description will be centered on the characteristic parts of the plasma processing apparatus 102 according to Embodiment 2.

[0179] The etching chamber 1 has a chamber inner housing 1B as a main component. Inside the chamber inner housing 1B, an anode electrode 2 serving as an upper electrode is provided above, and a cathode electrode 3S serving as a lower electrode is provided below.

[0180] The cathode electrode 3S is classified into a plurality of divided electrodes 3d. Each of the plurality of divided electrodes 3d is electrically connected to an individual coupling capacitor 34 provided under the bottom surface of the chamber inner housing 1B, so that the DC component is separated. The plurality of coupling capacitors 34 are each connected to a common wiring L3 via an isolation diode 35.

[0181] Each isolation diode 35 is provided so that the DC currents separated by the coupling capacitors 34 from the plurality of divided electrodes 3d do not interfere with each other. The anode is connected to the coupling capacitor 34 side, and the cathode is connected to the common wiring L3.

[0182] The common wiring L3 is connected to a variable capacitor 13 for impedance matching. The common wiring L3 with impedance matching performed by the variable capacitor 13 is connected to an RF power supply 12, and power for plasma discharge is supplied from the RF power supply 12 to each of the plurality of divided electrodes 3d of the cathode electrode 3S. Note that the RF power supply 12 is connected to the ground potential 16.

[0183] A Vdc measurement voltmeter 150, which is a lower electrode voltmeter, is connected to both electrodes of each coupling capacitor 34. That is, a plurality of Vdc measurement voltmeters 150 are provided corresponding to the plurality of divided electrodes 3d.

[0184] Therefore, the potential of each of the plurality of divided electrodes 3d can be measured by the corresponding Vdc measurement voltmeter 150 among the plurality of Vdc measurement voltmeters 150, which are a plurality of lower electrode voltmeters.

[0185] In this way, the plurality of divided electrodes 3d in the cathode electrode 3S function as the plurality of Vdc sensors s19 of the Vdc sensor assembly 190 in Embodiment 1. A voltage measurement mechanism including a coupling capacitor 34, an isolation diode 35, and a Vdc measurement voltmeter 150 is provided for each of the plurality of divided electrodes 3d.

[0186] Therefore, the plurality of divided electrodes 3d in the cathode electrode 3S serve as the plurality of Vdc sensors s19 of the Vdc sensor assembly 190 in Embodiment 1.

[0187] The substrate to be etched 10, which is the actual use substrate, is placed on the cathode electrode 3S.

[0188] Since the cathode electrode 3S is disposed directly below the substrate to be etched 10, the in-plane distribution of the negative offset potential Vdc directly below the substrate to be etched 10 can be comprehensively measured by the plurality of Vdc measurement voltmeters 150. Note that each of the plurality of Vdc measurement voltmeters 150 is a normal voltmeter.

[0189] Data indicating the plurality of measured potentials measured by the plurality of Vdc measurement voltmeters 150 becomes the Vdc sense data D11 in Embodiment 2. Further, the position information of the substrate to be etched 10 corresponding to the plurality of measured potentials becomes the position information data D12. As described above, since the position information corresponding to the plurality of measured potential data is already known, the position information data D12 may be supplied from the outside to the learning model creation device 70.

[0190] The combination of the Vdc sense data D11 and the position information data D12 becomes the data D1 for Vdc. The position information data D12 is data that complements the Vdc sense data D11.

[0191] Note that, in the plasma processing apparatus 200 shown in FIG. 17, the exhaust system such as a vacuum pump using the gas discharge hole 4 and the transfer system such as the lifting motor 32 are not shown in FIG. 7. That is, similar to the plasma processing apparatus 200, the plasma processing apparatus 102 of the second embodiment also has the above-described exhaust system and a transfer system including the substrate lifting pins 29.

[0192] FIG. 8 is a perspective view showing the overall configuration of the cathode electrode 3S. As shown in the figure, the cathode electrode 3S is composed of an aggregate of a plurality of divided electrodes 3d divided in a matrix shape. Since the plurality of divided electrodes 3d are arranged in an array, a complex in-plane distribution applied to the etching target substrate 10 directly above can be comprehensively measured with high resolution by a plurality of voltmeters 150 for Vdc measurement provided corresponding to the plurality of divided electrodes 3d.

[0193] Four lifting pin holes 40 provided penetrating near the center in the cathode electrode 3S are each provided to allow the substrate lifting pins 29 of the transfer system of the etching target substrate 10 to pass through.

[0194] In this way, the plasma processing apparatus 102, which is a main component of the plasma processing system 502 of the second embodiment, is configured.

[0195] In the plasma processing apparatus 102, the cathode electrode 3S functioning as the Vdc sensor assembly 190 has a plurality of divided electrodes 3d each functioning as a main part of the Vdc sensor s19, and the etching target substrate 10, which is an actual use substrate, is placed on the cathode electrode 3S.

[0196] In the plasma processing apparatus 102 in the plasma processing system 502 of the second embodiment, the potentials of the plurality of divided electrodes 3d of the cathode electrode 3S provided directly below the etching target substrate 10, which is an actual use substrate, are measured by a plurality of voltmeters 150 for Vdc measurement.

[0197] Therefore, from the plurality of measured potentials measured by the plurality of voltage meters 150 for Vdc measurement, the in-plane distribution of the negative offset potential Vdc on the substrate 10 to be etched can be accurately detected.

[0198] As described above, the plasma processing apparatus 102 used in the plasma processing system 502 of the second embodiment includes the cathode electrode 3S as the Vdc sensor assembly 190. The plurality of divided electrodes 3d of the cathode electrode 3S are provided corresponding to a plurality of locations on the formation surface of the substrate 10 to be manufactured in plan view. Therefore, from the plurality of measured potentials obtained from the plurality of voltage meters 150 for Vdc measurement provided corresponding to the plurality of divided electrodes 3d, the in-plane distribution of the negative offset potential Vdc of the substrate 10 to be etched can be accurately recognized.

[0199] As shown in FIG. 2, Vdc data D1 including Vdc sense data D11 indicating a plurality of measured potential data obtained from the plurality of voltage meters 150 for Vdc measurement in the plasma processing apparatus 102 and position information data D12 indicating the position information of the plurality of measured potential data on the substrate 10 to be etched is output to the subsequent learning model creation apparatus 70.

[0200] As shown in FIG. 2, Vpp voltage data D2 indicating the positive offset potential Vpp obtained from the Vpp measurement voltage meter 18 in the plasma processing apparatus 102 is output to the subsequent learning model creation apparatus 70.

[0201] (Learning model creation apparatus 70) The configuration of the learning model creation apparatus 70 in the plasma processing system 502 of the second embodiment is the same as the configuration shown in FIG. 3. And the learning model creation apparatus 70 receives the Vdc data D1, the Vpp voltage data D2, and the manufacturing content data D3, and creates the learned model D5, similar to the first embodiment.

[0202] Therefore, the plasma processing system 502 of the second embodiment has the same effects as the plasma processing system 501 of the first embodiment with respect to the learning model creation apparatus 70.

[0203] (Manufacturing content inference device 80) The configuration of the manufacturing content inference device 80 in the plasma processing system 502 of the second embodiment is the same as the configuration shown in FIG. 4. And, similar to the first embodiment, the manufacturing content inference device 80 receives the desired in-plane distribution data D4 and the learned model D5 and creates the inference data D7.

[0204] Therefore, the plasma processing system 502 of the second embodiment has the same effects as the plasma processing system 501 of the first embodiment with respect to the manufacturing content inference device 80.

[0205] <Embodiment 3> FIG. 9 is a perspective view showing the overall configuration of the sensor-embedded substrate 10S in the plasma processing device 103 included in the plasma processing system 503 of the third embodiment of the present disclosure. FIG. 10 is an explanatory view schematically showing the cross-sectional structure of the sensor-embedded substrate 10S and its surroundings.

[0206] The plasma processing system 503 of the third embodiment has a configuration in which the plasma processing device 101 is replaced with the plasma processing device 103 in FIG. 2.

[0207] (Plasma processing device 103) Hereinafter, with reference to FIGS. 9 and 10, the same components as those of the plasma processing device 101 shown in FIG. 1 are denoted by the same reference numerals, and the description thereof is appropriately omitted, and the description will be centered on the characteristic parts of the plasma processing device 103 of the third embodiment.

[0208] The plasma processing device 103 is characterized by having the sensor-embedded substrate 10S shown in FIGS. 9 and 10. The sensor-embedded substrate 10S, which is an actual use substrate, incorporates a plurality of Vdc sensors s19. This sensor-embedded substrate 10S is placed on the cathode electrode 3, which is a lower electrode.

[0209] The plasma processing apparatus 103 has a structure in which the substrate to be etched 10 in the plasma processing apparatus 200 of the basic technology shown in FIG. 17 is replaced by a combination of a substrate 10S with built-in sensors, lead-out TAB wirings 38 and 38Y. Therefore, the plasma processing apparatus 103 has a cathode electrode 3 similar to that of the plasma processing apparatus 200. However, the voltmeter 15 for Vdc measurement is omitted.

[0210] A plurality of Vdc sensors s19 are built into the sensor-integrated substrate 10S in a manner arranged in an array, and are configured to be able to comprehensively measure the in-plane distribution of the complex negative offset potential Vdc applied to the sensor-integrated substrate 10S itself with high resolution.

[0211] The plurality of Vdc sensors s19 are electrically connected to the lead-out TAB (Tape Automated Bonding) wirings 38X and 38Y. Here, assuming that the plurality of Vdc sensors s19 are arranged in a matrix defined in the X direction and the Y direction, by using the lead-out TAB wiring 38X as a plurality of scanning lines that scan along the X direction and using the lead-out TAB wiring 38Y as a plurality of output lines that draw out the measured potential along the Y direction, the measured potential of each of the plurality of Vdc sensors s19 can be sequentially taken out from the lead-out TAB wiring 38Y.

[0212] The lead-out TAB wirings 38X and 38Y are respectively provided from the side surface of the sensor-integrated substrate 10S so as to be connected to the plurality of Vdc sensors s19 within the sensor-integrated substrate 10S. In this way, the Vdc sense data D11 indicating the plurality of measured potentials measured by the plurality of Vdc sensors s19 built into the sensor-integrated substrate 10S can be taken out by the lead-out TAB wirings 38 and 38Y.

[0213] As shown in FIG. 10, the sensor-integrated substrate 10S is placed on the cathode electrode 3. Since the sensor-integrated substrate 10S is the actual substrate to be manufactured, various manufacturing processes are executed on the sensor-integrated substrate 10S.

[0214] In the example shown in FIG. 10, an etched material 21 is already formed on the surface of the substrate 10S with built-in sensors, and a resist 20 patterned on the etched material 21 is provided. That is, FIG. 10 shows a state in which an etching process is performed on the etched material 21 using the resist 20 as a mask.

[0215] In this state, an etching gas is introduced into the etching chamber 1 of the plasma processing apparatus 103, and an RF discharge is started between the anode electrode 2 and the cathode electrode 3 to execute the etching process. By the etching process, the etching reaction with respect to the etched material 21 proceeds. Simultaneously with the etching of the etched material 21, a plurality of measured potentials measured by a plurality of Vdc sensors s19 built in the substrate 10S with built-in sensors are taken out to the outside via the extraction TAB wirings 38 and 38Y. Data indicating the plurality of taken-out measured potentials becomes Vdc sense data D11.

[0216] As described above, in the plasma processing apparatus 103 of Embodiment 3, the substrate 10S with built-in sensors incorporates a plurality of Vdc sensors s19. That is, the substrate 10S with built-in sensors functions as the Vdc sensor assembly 190 of Embodiment 1 and the actual use substrate. At this time, with respect to the plasma processing apparatus 103, there is no need to perform special modification to the configuration of a conventional parallel plate type RIE etching apparatus, and the negative offset potential Vdc can be measured with high in-plane resolution.

[0217] As described above, the plasma processing apparatus 103, which is a main component of the plasma processing system 503 of Embodiment 3, is configured.

[0218] In the plasma processing apparatus 103 of Embodiment 3, the substrate 10S with built-in sensors serving as the actual use substrate incorporates a plurality of Vdc sensors s19, and this substrate 10S with built-in sensors is placed on the cathode electrode 3S.

[0219] In the plasma processing apparatus 103 in the plasma processing system 503 of Embodiment 3, the in-plane distribution of the negative offset potential Vdc of the sensor-embedded substrate 10S, which is the actual use substrate, can be accurately detected from a plurality of measured potentials obtained from the sensor-embedded substrate 10S incorporating a plurality of Vdc sensors s19.

[0220] As described above, the plasma processing apparatus 103 used in the plasma processing system 503 of Embodiment 3 includes the sensor-embedded substrate 10S as a Vdc sensor assembly 190. The plurality of Vdc sensors s19 included in the sensor-embedded substrate 10S are provided corresponding to a plurality of locations on the formation surface of the sensor-embedded substrate 10S to be manufactured in a plan view. Therefore, the in-plane distribution of the negative offset potential Vdc of the sensor-embedded substrate 10S can be accurately recognized from the plurality of measured potentials obtained from the plurality of Vdc sensors s19.

[0221] As shown in FIG. 2, Vdc data D1 including Vdc sense data D11 indicating a plurality of measured potential data obtained from a plurality of Vdc sensors s19 in the plasma processing apparatus 103 and position information data D12 indicating the position information of the etched substrate 10 for the plurality of measured potential data is output to the subsequent learning model creation device 70.

[0222] Furthermore, similar to Embodiment 1 and Embodiment 2, Vpp voltage data D2 indicating the positive offset potential Vpp obtained from the Vpp measurement voltmeter 18 in the plasma processing apparatus 103 is output to the subsequent learning model creation device 70.

[0223] (Learning model creation device 70) The configuration of the learning model creation device 70 in the plasma processing system 503 of Embodiment 3 is the same as the configuration shown in FIG. 3. The learning model creation device 70 receives the Vdc data D1, the Vpp voltage data D2, and the manufacturing content data D3, and creates a learned model D5, similar to Embodiment 1.

[0224] Therefore, the plasma processing system 503 of Embodiment 3 has the same effect as the plasma processing system 501 of Embodiment 1 with respect to the learning model creation device 70.

[0225] (Manufacturing content inference device 80) The configuration of the manufacturing content inference device 80 in the plasma processing system 503 of Embodiment 3 is the same as the configuration shown in FIG. 4. And, similar to Embodiment 1, the manufacturing content inference device 80 receives the desired in-plane distribution data D4 and the learned model D5 and creates the inference data D7.

[0226] Therefore, the plasma processing system 503 of Embodiment 3 has the same effect as the plasma processing system 501 of Embodiment 1 with respect to the manufacturing content inference device 80.

[0227] <Embodiment 4> FIG. 11 is a perspective view showing the overall configuration of the Vdc sensor substrate 19 in the plasma processing apparatus 104 included in the plasma processing system 504 of Embodiment 4 of the present disclosure. FIG. 12 is an explanatory view schematically showing the cross-sectional structure of the Vdc sensor substrate 19 and its periphery.

[0228] In Embodiment 4, the plasma processing system 504 has a configuration in which the plasma processing apparatus 101 is replaced with the plasma processing apparatus 104 in FIG. 2.

[0229] (Plasma processing apparatus 104) Hereinafter, with reference to FIGS. 11 and 12, the same components as those of the plasma processing apparatus 101 shown in FIG. 1 are denoted by the same reference numerals, and the description thereof will be appropriately omitted, and the description will be centered on the characteristic parts of the plasma processing apparatus 104.

[0230] The plasma processing apparatus 104 is characterized in that a new Vdc sensor substrate 19 is provided. The Vdc sensor substrate 19 incorporates a plurality of Vdc sensors s19. This Vdc sensor substrate 19 is placed on the cathode electrode 3 which is a lower electrode. The etched substrate 10 which is an actual use substrate is placed on the Vdc sensor substrate 19.

[0231] The plasma processing apparatus 104 has a structure in which a Vdc sensor substrate 19 is provided between the cathode electrode 3 and the substrate to be etched 10 in the plasma processing apparatus 200 of the basic technology shown in FIG. 17. Therefore, the plasma processing apparatus 104 has the same cathode electrode 3 and substrate to be etched 10 as the plasma processing apparatus 200. However, the voltmeter 15 for Vdc measurement is omitted.

[0232] A plurality of Vdc sensors s19 are built into the Vdc sensor substrate 19 in a manner of being arranged in an array except for a plurality of lift pin holes 40, and are configured to be able to cover and measure the in-plane distribution of the complex negative offset potential Vdc on the directly above substrate to be etched 10 with high resolution.

[0233] Each of the four lift pin holes 40 provided penetrating through the center of the Vdc sensor substrate 19 is provided to allow the substrate lift pins 29 of the conveyance system of the substrate to be etched 10 to pass through.

[0234] The plurality of Vdc sensors s19 are electrically connected to the TAB wirings 38X and 38Y for extraction. Here, assuming that the plurality of Vdc sensors s19 are arranged in a matrix defined in the X direction and the Y direction, by using the TAB wiring 38X for extraction as a plurality of scanning lines that scan along the X direction and using the TAB wiring 38Y for extraction as a plurality of output lines that extract the measured potential along the Y direction, the measured potential of each of the plurality of Vdc sensors s19 can be extracted from the TAB wiring 38Y for extraction.

[0235] The TAB wirings 38X and 38Y for extraction are provided so as to be connected to the plurality of Vdc sensors s19 in the Vdc sensor substrate 19 from the side surfaces of the Vdc sensor substrate 19, respectively. In this way, the Vdc sense data D11 indicating the plurality of measured potentials measured by the plurality of Vdc sensors s19 built into the Vdc sensor substrate 19 can be extracted by the TAB wirings 38 and 38Y for extraction.

[0236] As shown in Fig. 12, a Vdc sensor substrate 19 is placed on the cathode electrode 3, and an etched substrate 10 is placed on the Vdc sensor substrate 19. Since the etched substrate 10 is an actual use substrate, various manufacturing processes are performed on the etched substrate 10.

[0237] In the example shown in Fig. 12, an etched material 21 has already been formed on the surface of the etched substrate 10, and a patterned resist 20 is provided on the etched material 21. That is, Fig. 12 shows a state in which an etching process is performed on the etched material 21 using the resist 20 as a mask.

[0238] In the plasma processing apparatus 104, an etching gas is introduced into the etching chamber 1 in the state shown in Fig. 12, and an RF discharge is started between the anode electrode 2 and the cathode electrode 3 to execute the etching process. The etching reaction with respect to the etched material 21 proceeds by the etching process. At the same time as the etched material 21 is etched, a plurality of measured potentials measured by a plurality of Vdc sensors s19 built in the Vdc sensor substrate 19 are taken out to the outside through the extraction TAB wirings 38 and 38Y. Data indicating the plurality of taken-out measured potentials becomes Vdc sense data D11.

[0239] Thus, in the plasma processing apparatus 104 of the fourth embodiment, the Vdc sensor substrate 19 incorporates a plurality of Vdc sensors s19. That is, the Vdc sensor substrate 19 functions as the Vdc sensor assembly 190 of the first embodiment.

[0240] At this time, by replacing only a plurality of substrate lifting pins 29 used in the transfer system of the etched substrate 10 from a conventional parallel plate type RIE etching apparatus, a plasma processing apparatus 104 capable of measuring the negative offset potential Vdc with high in-plane resolution can be realized. The reason for replacing the plurality of substrate lifting pins 29 is that the arrangement height of the etched substrate 10 becomes higher because the Vdc sensor substrate 19 is provided between the cathode electrode 3 and the etched substrate 10.

[0241] In this way, the plasma processing apparatus 104, which is a main component of the plasma processing system 504 of the fourth embodiment, is configured.

[0242] In the plasma processing apparatus 104, the Vdc sensor substrate 19 incorporates a plurality of Vdc sensors s19, and the Vdc sensor substrate 19 is placed on this Vdc sensor substrate 19.

[0243] By adopting this structure, the plasma processing apparatus 104 of the fourth embodiment can collect high in-plane resolution Vdc sense data D11 from a plurality of Vdc sensors s19 by replacing only the substrate lifting pins 29 in a conventional parallel plate type RIE etching apparatus, as described above.

[0244] The plasma processing apparatus 104 in the plasma processing system 504 of the fourth embodiment can accurately detect the in-plane distribution of the negative offset potential Vdc in the etched substrate 10, which is the actual use substrate, from a plurality of measured potentials obtained from the plurality of Vdc sensors s19 incorporated in the Vdc sensor substrate 19.

[0245] As described above, the plasma processing apparatus 104 used in the plasma processing system 504 of the fourth embodiment includes a Vdc sensor substrate 19 as a Vdc sensor assembly 190. A plurality of Vdc sensors s19 included in the Vdc sensor substrate 19 are provided corresponding to a plurality of locations on the formation surface of the substrate 10 to be etched, which is the object of manufacturing, in a plan view. Therefore, the in-plane distribution of the negative offset potential Vdc on the substrate 10 to be etched can be accurately recognized based on a plurality of measured potentials obtained from the plurality of Vdc sensors s19.

[0246] As shown in FIG. 2, Vdc data D1 including Vdc sense data D11 indicating a plurality of measured potentials obtained from a plurality of Vdc sensors s19 in the plasma processing apparatus 104 and position information data D12 indicating the position information of the substrate 10 to be etched for the plurality of measured potential data is output to the subsequent learning model creation device 70.

[0247] Furthermore, similar to the first to third embodiments, Vpp voltage data D2 indicating the positive offset potential Vpp obtained from the Vpp measuring voltmeter 18 in the plasma processing apparatus 104 is output to the subsequent learning model creation device 70.

[0248] (Learning Model Creation Device 70) The configuration of the learning model creation device 70 in the plasma processing system 504 of the fourth embodiment is the same as the configuration shown in FIG. 3. And the learning model creation device 70 receives the Vdc data D1, the Vpp voltage data D2, and the manufacturing content data D3, and creates a learned model D5, similar to the first embodiment.

[0249] Therefore, the plasma processing system 504 of the fourth embodiment has the same effect as the plasma processing system 501 of the first embodiment with respect to the learning model creation device 70.

[0250] (Manufacturing Content Inference Device 80) The configuration of the manufacturing content inference device 80 in the plasma processing system 504 of Embodiment 4 is the same as the configuration shown in FIG. 4. Similar to Embodiment 1, the manufacturing content inference device 80 receives the desired in-plane distribution data D4 and the learned model D5, and creates the inference data D7.

[0251] Therefore, the plasma processing system 504 of Embodiment 4 has the same effect as the plasma processing system 501 of Embodiment 1 with respect to the manufacturing content inference device 80.

[0252] <Embodiment 5> Embodiment 5 is characterized by the arrangement structure of the plurality of Vdc sensors s19 on the Vdc sensor substrate 19 shown in Embodiment 4. That is, in the plasma processing apparatus 104 of Embodiment 4, the plasma processing system of Embodiment 5 is the same as the plasma processing system 504 of Embodiment 4 except that the Vdc sensor substrate 19 shown in FIGS. 11 and 12 is replaced with the Vdc sensor substrate 19B shown in FIG. 13, and the plasma processing apparatus of Embodiment 5 is the same as the plasma processing apparatus 104 of Embodiment 4.

[0253] (Vdc sensor substrate 19B) FIG. 13 is an explanatory diagram schematically showing the arrangement structure of the Vdc sensors on the Vdc sensor substrate 19B of Embodiment 5. As shown in FIG. 13, the Vdc sensor substrate 19B incorporates Vdc sensors s42 to s46 as potential measurement sensors corresponding to the plurality of Vdc sensors s19 of the Vdc sensor substrate 19. There is at least one Vdc sensor s42 to s46 respectively. Hereinafter, for convenience of explanation, without specifically showing the number of each of the Vdc sensors s42 to s46, they will be simply referred to as Vdc sensor s42, Vdc sensor s43, Vdc sensor s44, Vdc sensor s45, and Vdc sensor s46 for explanation.

[0254] Similar to the plurality of Vdc sensors s19 of the Vdc sensor substrate 19 of Embodiment 4, the Vdc sensors s42 to s46 of the Vdc sensor substrate 19B of Embodiment 5 are electrically connected to the TAB wirings 38X and 38Y for extraction.

[0255] As shown in FIG. 13, the plurality of potential measurement sensors provided on the Vdc sensor substrate 19B are classified into Vdc sensors s42 to s46. The Vdc sensor s42 is provided in the central region of the Vdc sensor substrate 19 (hereinafter abbreviated as the "substrate central region"), the Vdc sensor s43 is provided in the peripheral region of the Vdc sensor substrate 19 (hereinafter abbreviated as the "substrate peripheral region"), and the Vdc sensor s44 is provided in the peripheral region of the lifting pin hole 40 through which the substrate lifting pin 29 penetrates (hereinafter abbreviated as the "peripheral region of the pin hole").

[0256] In the substrate 10 to be etched, the above-described substrate central region, substrate peripheral region, and peripheral region of the pin hole are each a negative potential focus region where a large variation in the negative offset potential Vdc is predicted.

[0257] Further, the Vdc sensor s45 is provided in the resist pattern sparse region R45 during the etching process, and the Vdc sensor s46 is provided in the resist pattern dense region R46 during the etching process. The resist pattern sparse region R45 is a region where the formation density of the resist pattern to be formed during the etching process is lower than the normal level, and the resist pattern dense region R46 is a region where the formation density of the resist pattern to be formed during the etching process is higher than the normal level.

[0258] When the resist pattern formation density of the resist pattern sparse region R45 is M45, the resist pattern formation density of the resist pattern dense region R46 is M46, and the resist pattern formation density of the normal level is M0, the density relationship of {M46 > M0 > M45} is established. Note that the reference density M0 may be divided for determining the resist pattern sparse region R45 and for determining the resist pattern dense region R46.

[0259] In this way, the resist pattern sparse region R45 and the resist pattern dense region R46 are recognized from the resist pattern to be formed above the substrate 10 to be etched, which is the actual use substrate.

[0260] In the substrate 10 to be etched, the resist pattern sparse region R45 and the resist pattern dense region R46 are negative potential attention regions where significant fluctuations in the negative offset potential Vdc are predicted respectively.

[0261] Thus, the Vdc sensors s42 to s46, which are a plurality of potential measurement sensors provided on the Vdc sensor substrate 19, have a sensor arrangement characteristic of being arranged corresponding to the negative potential attention regions where significant fluctuations in the negative offset potential Vdc of the substrate 10 to be etched are predicted, based on the shape of the substrate 10 to be etched and the coarseness density of the resist pattern to be formed on the substrate 10 to be etched.

[0262] For example, in the Vdc sensor substrate 19B, the Vdc sensors s42 to s46 are arranged in regions that overlap in plan view with respect to the central region of the substrate, the peripheral region of the substrate, and the region around the pin hole in the substrate 10 to be etched, as well as the resist pattern sparse region R45 and the resist pattern dense region R46.

[0263] Therefore, since the Vdc sensor substrate 19B used in the plasma processing apparatus of Embodiment 5 has the above-described sensor arrangement characteristic, the number of arranged potential measurement sensors corresponding to the Vdc sensors s42 to s46 is minimized, and the in-plane distribution of the negative offset potential Vdc corresponding to the substrate 10 to be etched can be accurately recognized.

[0264] That is, by using the Vdc sensor substrate 19B of Embodiment 5, the number of potential measurement sensors required for machine learning and the number of data indicating the measured potential can be reduced by using a conventional parallel plate type RIE etching apparatus as it is, and the cost required for a plurality of potential measurement sensors can be reduced.

[0265] As a result, it is possible to significantly reduce the number of times of the etching condition optimization experiment for obtaining the highly accurate learned model D5 by reducing the load on the processing circuit necessary for the machine learning process in the learning model creation device 70. Further, conventionally, a huge number of test substrates were required for data analysis, but by adopting the Vdc sensor substrate 19B of the fifth embodiment, the number of etched substrates 10 necessary for the etching condition optimization can be reduced from one to several.

[0266] Note that in FIG. 13, a Vdc sensor substrate 19B obtained by improving the Vdc sensor substrate 19 of the fourth embodiment is shown. Similarly, the cathode electrode 3S of the second embodiment may be improved to have the above-described sensor arrangement characteristics, or the sensor-embedded substrate 10S of the third embodiment may be improved to have the above-described sensor arrangement characteristics.

[0267] For example, when improving the cathode electrode 3S of the second embodiment, a measure such as selectively providing a voltage measurement mechanism along the above-described sensor arrangement characteristics among a plurality of divided electrodes 3d can be considered.

[0268] <Embodiment 6> Embodiment 6 is characterized by the circuit configuration of a plurality of Vdc sensors s19 in the Vdc sensor substrate 19C. Therefore, in the plasma processing apparatus 104 of the fourth embodiment, the plasma processing system of the sixth embodiment is the same as the plasma processing system 504 of the fourth embodiment, and the plasma processing apparatus of the sixth embodiment is the same as the plasma processing apparatus 104 of the fourth embodiment, except that the circuit configuration shown by the Vdc sensor substrate 19C is adopted as the circuit configuration of a plurality of Vdc sensors s19 in the Vdc sensor substrate 19.

[0269] (Vdc sensor substrate 19C) FIG. 14 is a circuit diagram showing the circuit configuration of the Vdc sensor s19 in the Vdc sensor substrate 19C of Embodiment 6. An XY orthogonal coordinate system is shown in FIG. 14. As shown in FIG. 14, each of the plurality of Vdc sensors s19 includes a Vdc detection capacitor 47 and a switching transistor 50 as main components. The Vdc detection capacitor 47 serves as a charge storage capacitor, and the switching transistor 50 serves as a charge extraction transistor.

[0270] The plurality of Vdc sensors s19 are arranged in a matrix. Here, it is assumed that the plurality of Vdc sensors s19 are arranged in NX (≥2) × NY (≥2), with the X direction being the row direction and the Y direction being the column direction. NX is the first number, and NY is the second number. Thus, (NX × NY) Vdc sensors s19 are provided as the plurality of Vdc sensors s19.

[0271] NX gate wirings L1 extend in the Y direction, and NY source wirings L2 extend in the X direction. The NX gate wirings L1 serve as the scanning lines of the first number, and the NY source wirings L2 serve as the output lines of the second number.

[0272] Therefore, the Vdc sensor substrate 19C has NX gate wirings L1 each provided along the column direction and NY source wirings L2 each provided along the row direction.

[0273] Then, the aggregate of the NX gate wirings L1 is drawn out to the outside as the TAB wiring 38Y for extraction shown in FIGS. 11 and 12, and the aggregate of the NY source wirings L2 is drawn out to the outside as the TAB wiring 38X for extraction shown in FIGS. 11 and 12.

[0274] In each Vdc sensor s19, one electrode of the Vdc detection capacitor 47, which is a charge storage capacitor, is connected to the drain, which is one electrode of the switching transistor 50. Note that the other electrode of the Vdc detection capacitor 47 is set to, for example, a ground potential.

[0275] The gate, which serves as the control electrode of the switching transistor 50 that is a charge extraction transistor, is connected to the corresponding gate wiring L1 among the NX gate wirings L1. The source, which is the other electrode of the switching transistor 50, is connected to the corresponding source wiring L2 among the NY L2s.

[0276] Since the electrons 11 by the plasma dissociated in the etching chamber 1 descend toward the cathode electrode 3, they are accumulated in the Vdc detection capacitor 47 of the Vdc sensor substrate 19C provided directly below the substrate 10 to be etched. That is, the negative charge corresponding to the negative offset potential Vdc at the corresponding location on the substrate 10 to be etched is accumulated in the Vdc detection capacitor 47.

[0277] The plurality of Vdc sensors s19 in the Vdc sensor substrate 19C of the sixth embodiment each include the Vdc detection capacitor 47 and the switching transistor 50 as main components. Therefore, the negative charge corresponding to the negative offset potential Vdc can be accumulated in the Vdc detection capacitor 47.

[0278] When one of the NX gate wirings L1, which are the scanning lines of the first number, is selected as the selection scanning line, among the plurality of Vdc sensors s19, the NY Vdc sensors s19 having the switching transistor 50 whose gate is connected to the selection scanning line are in a selected state as NY selection potential measurement sensors. A selection signal for turning on the switching transistor 50 is applied to the selection scanning line.

[0279] Then, in each of the NY Vdc sensors s19 in the selected state, the negative charge accumulated in the Vdc detection capacitor 47 is output to the outside as NY measurement potentials from the NY source wirings L2 through the on-state switching transistor 50.

[0280] Therefore, by sequentially selecting the NX gate wirings L1 as selected scanning lines, it is possible to output the measured potentials of all (NX×NY) Vdc sensors s19 to the outside. In this way, the plasma processing apparatus having the Vdc sensor substrate 19C of the sixth embodiment can detect the in-plane distribution of the negative offset potential Vdc of the substrate 10 to be etched with high accuracy using multiple measured potentials.

[0281] By employing the circuit configuration described above, the multiple Vdc sensors s19 of the Vdc sensor substrate 19C of embodiment 6 can detect the in-plane distribution of the negative offset potential Vdc on the substrate 10 to be etched with a high resolution of several hundred μm.

[0282] In the sixth embodiment, the Vdc sensor board 19 of the fourth embodiment is improved to provide a Vdc sensor board 19C having a circuit configuration of a plurality of Vdc sensors s19 as shown in Fig. 14. As another aspect, the sensor-embedded board 10S of the third embodiment can be improved to provide a modified example of the sensor-embedded board 10S having a circuit configuration of a plurality of Vdc sensors s19 as shown in Fig. 14.

[0283] Similarly, a circuit configuration for Vdc sensors s42 to s46 may be realized by improving the Vdc sensor substrate 19B of embodiment 5. However, since the Vdc sensors s42 to s46 are not arranged in a matrix, some ingenuity is required in the arrangement and number of gate wiring L1 and source wiring L2.

[0284] The above-mentioned NX gate wirings L1 correspond to one of the lead-out TAB wirings 38X and 38Y shown in the third and fourth embodiments, and the above-mentioned NY source wirings L2 correspond to the other of the lead-out TAB wirings 38X and 38Y shown in the third and fourth embodiments.

[0285] Therefore, even when wiring cannot be drawn out from below a plurality of Vdc sensors s19, such as the sensor-embedded substrate 10S of Embodiment 3, the Vdc sensor substrate 19 of Embodiment 4, and the Vdc sensor substrate 19B of Embodiment 5, the in-plane distribution of the negative-side offset potential Vdc in the substrate 10 to be etched can be detected via the TAB wirings 38 and 38Y for drawing out.

[0286] <Embodiment 7> (First) Regarding Embodiment 7 and Embodiment 8 shown below, a system including the plasma processing systems 501 to 504 of Embodiments 1 to 4 may be referred to as the "plasma processing system 500", and a device including the plasma processing apparatuses 101 to 104 of Embodiments 1 to 4 may be referred to as the "plasma processing apparatus 100" for explanation. Note that the plasma processing apparatus 104 includes Embodiment 5 in which the Vdc sensor substrate 19 is replaced with the Vdc sensor substrate 19B, and Embodiment 6 in which the circuit configuration of the Vdc sensor substrate 19C is adopted for the Vdc sensor substrate 19.

[0287] That is, Embodiment 7 has the same configuration as the plasma processing system 500 and the plasma processing apparatus 100 shown in Embodiments 1 to 6, except for the characteristic parts described below.

[0288] FIG. 15 is an explanatory diagram schematically showing the configuration of the learning model creation device 70B of Embodiment 7 included in the plasma processing system 500. The plasma processing system 500 of Embodiment 7 has a configuration in which the learning model creation device 70 shown in FIG. 3 is replaced with the learning model creation device 70B shown in FIG. 15.

[0289] (Learning Model Creation Device 70B) Hereinafter, with reference to FIG. 15, the same configurations as those of the learning model creation device 70 shown in FIG. 3 are denoted by the same reference numerals and the description thereof is appropriately omitted, and the description will be centered on the characteristic parts of the learning model creation device 70B which is a data analysis device.

[0290] As shown in FIG. 2, the learning model creation device 70B, which is a data analysis device, receives Vdc data D1, Vpp voltage data D2, and production content data D3 from the plasma processing device 100. The learning model creation device 70B further receives process data D6 from the outside. The process data D6 is data necessary for pass / fail determination using the Mahalanobis-Taguchi method (hereinafter abbreviated as the "M-T method") among the data included in the production content data D3.

[0291] As shown in FIG. 15, the learning model creation device 70B includes a Vdc in-plane distribution creation unit 71V, a DC bias in-plane distribution creation unit 72, a learning model creation unit 73, and a learning model recording unit 74 as main components.

[0292] Similar to the Vdc in-plane distribution creation unit 71 of the learning model creation device 70 shown in FIG. 3, the Vdc in-plane distribution creation unit 71V recognizes a plurality of measured potentials with recognized position information as a plurality of position-recognized measured potentials based on the Vdc data D1.

[0293] Furthermore, the Vdc in-plane distribution creation unit 71V performs pass / fail determination using the M-T method based on the recognized plurality of position-recognized measured potentials, the Vpp voltage data D2, and the process data D35, validates the position-recognized measured potentials that are determined to be good as good determination measured potentials among the plurality of position-recognized measured potentials, and invalidates the position-recognized measured potentials that are determined to be bad.

[0294] Therefore, the position-recognized measured potentials determined to be bad by the M-T method are excluded from the processing targets of the Vdc in-plane distribution creation unit 71V. For example, measured potentials obtained from a malfunctioning Vdc sensor s19 are assumed as the position-recognized measured potentials determined to be bad.

[0295] In this way, the Vdc in-plane distribution creation unit 71V narrows down from the plurality of position determination measured potentials to a plurality of good determination measured potentials. Thereafter, the Vdc in-plane distribution creation unit 71V generates Vdc in-plane distribution data D10 based on the plurality of good determination measured potentials. This Vdc in-plane distribution data D10 becomes the in-plane potential distribution data corresponding to the substrate.

[0296] In this way, the in-plane distribution creation unit 71V for Vdc functions as an in-plane potential distribution data creation unit that derives Vdc in-plane distribution data D10 indicating the in-plane distribution of the negative offset potential Vdc in the substrate to be etched 10 by using only a plurality of good determination measurement potentials based on the Vdc data D1 including the Vdc sense data D11.

[0297] The processing by the DC bias in-plane distribution creation unit 72, the learning model creation unit 73, and the learning model recording unit 74 is performed in the same manner as the DC bias in-plane distribution creation unit 72, the learning model creation unit 73, and the learning model recording unit 74 of the learning model creation apparatus 70 of the first embodiment shown in FIG. 3, respectively.

[0298] In this way, in the learning model creation apparatus 70B of the seventh embodiment, the Vdc in-plane distribution creation unit 71V, which is an in-plane potential distribution data creation unit, derives Vdc in-plane distribution data D10, which is substrate-corresponding in-plane potential distribution data, based on a plurality of good determination measurement potentials narrowed down by a pass / fail determination using the M-T method from the plurality of position recognition measurement potentials indicated by the Vdc data D1.

[0299] Therefore, the learning model creation apparatus 70B can improve the inference accuracy based on the learned model D5, which is the finally created relevance data, to the extent that the accuracy of the Vdc in-plane distribution data D10 can be increased.

[0300] By having the Vdc in-plane distribution creation unit 71V, the learning model creation apparatus 70B can exclude information from the failed Vdc sensor s19 and create BV in-plane distribution data D20, which is more accurate analysis data. In addition, it becomes easier to identify the failed Vdc sensor s19, contributing to cost reduction in the maintenance management of the plasma processing apparatus 100.

[0301] In the learning model creation device 70B shown in FIG. 15 as well, similar to the learning model creation device 70 shown in FIG. 3, a modification example can be considered. The modification example of the learning model creation device 70B has a configuration in which a component corresponding to the learning model creation unit 73 is provided outside the learning model creation device 70B as an external learning model creation unit.

[0302] (Manufacturing content inference device 80) The configuration of the manufacturing content inference device 80 in the plasma processing system 500 of Embodiment 7 is the same as the configuration shown in FIG. 4. And, similar to Embodiment 1, the manufacturing content inference device 80 receives the desired in-plane distribution data D4 and the learned model D5, and creates the inference data D7.

[0303] Therefore, regarding the manufacturing content inference device 80, the plasma processing system 500 of Embodiment 7 can be expected to exhibit an inference accuracy equal to or higher than that of the plasma processing system 501 of Embodiment 1 to the extent that the accuracy of the learned model D5 is improved.

[0304] <Embodiment 8> Embodiment 8 defines a method for determining the number of settings of a plurality of Vdc sensors s19 used in the plasma processing device 100.

[0305] FIG. 16 is an explanatory diagram schematically showing the planar structure of the wafer 60 used as the substrate to be etched 10.

[0306] As shown in the figure, the wafer 60 serving as the actual use substrate is circular in plan view, and the wafer 60 is divided into a plurality of dies (not shown). The plurality of dies are chips that each function as various devices.

[0307] In FIG. 16, the diameter of the wafer 60 that is circular in plan view is defined as R60, and Sin45° = Cos45° ≒ 0.707 is used. Therefore, with respect to the wafer diameter R60, the length L60 of each side of the largest square that can be secured within the wafer 60 is obtained by {L60 = R60 × 0.707…(1)}. Also, the area SX of the largest square within the wafer 60 is {SX = (R × 0.707) 2…(2)}. Assuming that a plurality of dies are cut out from the largest square within the wafer 60, at least two Vdc sensors s19 mounted on each die are required depending on the density of the pattern.

[0308] Therefore, in Embodiment 8, the minimum required number N19 of Vdc sensors when providing a plurality of Vdc sensors s19 is determined by the following formula (3).

[0309] N19 = {(R × 0.707) 2 / DX} × 2…(3) In formula (3), DX represents the largest area among the areas of the plurality of dies.

[0310] Thus, in the plasma processing system 500 of Embodiment 8, when the largest square area within the planar structure of the wafer 60 is SX and the largest area of each of the plurality of dies is DX, the number NV of the plurality of Vdc sensors s19 is set to satisfy the conditional expression {NV >= N19 = (SX / DX) × 2…(4)}.

[0311] In the method for determining the set number NV of Embodiment 6, when the wafer 60 with a wafer diameter R60 of 6 inches has a die size of a 2 cm × 2 cm square, the minimum required number N19 of Vdc sensors s19 is 58.

[0312] To process the 58 measured potentials in combination with the manufacturing content data D3, the feature amount of the data is 58 × (the number of items of the process data). Here, as the manufacturing content data D3, the Vpp measurement voltage indicated by the Vpp voltage data D2, the etching gas type indicated by the gas type data D31, the etching gas flow rate indicated by the gas flow rate data D32, the etching gas pressure indicated by the RF output data D34, the electrode distance between the anode electrode 2 and the cathode electrode 3 indicated by the process data D35, and the RF input power indicated by the RF output data D34 are adopted. Here, it is assumed that there is one type of etching gas.

[0313] In this case, the number of required items as the manufacturing content data D3 is six, and the feature amount of the processing data is at least 348. To predict the optimal manufacturing content for obtaining the desired in-plane distribution data D4 from data of this scale, a machine learning model using the regression analysis method is almost essential.

[0314] Note that as a specific method of the regression analysis method, any regression analysis method used in machine learning, such as the multiple regression analysis method, Ridge regression analysis method, Lasso regression analysis method, Logistic regression analysis method, etc., may be selected.

[0315] As described above, the plasma processing apparatus 100 in the plasma processing system 500 of Embodiment 8 sets the set number NV of the plurality of Vdc sensors s19 so as to satisfy the above-described conditional expression (4), thereby minimizing the number of Vdc sensors s19 and enabling the in-plane distribution of the negative potential corresponding to the actually used substrate to be recognized at a practical level.

[0316] Further, the plasma processing system 500 of Embodiment 8 can easily determine that it is preferable to perform data processing by a machine learning method using the regression analysis method from the beginning among various data analysis methods used by the plasma processing system 500 by obtaining the minimum required number N19 of the Vdc sensors s19 based on Expression (3), and can reduce the labor and time required for examining the data analysis processing method.

[0317] <Scope of application of plasma processing apparatus 100> The plasma processing apparatus 100 supplies a manufacturing gas from a plurality of gas supply holes provided in the anode electrode 2 that also serves as a shower head SH for introducing an etching gas. In the above-described Embodiments 1 to 8, the plasma processing apparatus 100 can be used as a shape processing apparatus such as an RIE etching apparatus by supplying an etching gas as the manufacturing gas.

[0318] Further, as another aspect, the plasma processing apparatus 100 supplies a film-forming gas as a manufacturing gas from the gas supply holes described above, uses the etching chamber 1 as a film-forming chamber, and uses the substrate to be etched 10 as a substrate for thin-film deposition 22, and generates cations 7 and activated radicals 8 of the film-forming gas, whereby the plasma processing apparatus 100 can be used as a thin-film manufacturing apparatus.

[0319] Thus, the plasma processing apparatus 100 used in the plasma processing system 500 can be used as a shape processing apparatus such as an etching apparatus by using an etching gas as a manufacturing gas, and can be used as a thin-film manufacturing apparatus by using a film-forming gas as a manufacturing gas.

[0320] <Others> Although the present disclosure has been described in detail, the above description is illustrative in all aspects and the present disclosure is not limited thereto. It is understood that innumerable modifications not illustrated can be assumed without departing from the scope of the present disclosure.

Description of Reference Numerals

[0321] 1 Etching chamber, 2 Anode electrode, 3, 3S Cathode electrode, 3d Divided electrode, 10 Substrate to be etched, 10S Substrate with built-in sensor, 18 Voltage meter for Vpp measurement, 19, 19B, 19C Vdc sensor substrate, 20 Resist, 21 Substance to be etched, 70, 70V Learning model creation device, 71, 71V Vdc in-plane distribution creation unit, 72 DC bias in-plane distribution creation unit, 73 Learning model creation unit, 74 Learning model recording unit, 80 Manufacturing content inference device, 81 Inference unit, 82 Processing content determination unit, 100 - 104 Plasma processing apparatus, 150 Voltage meter for Vdc measurement, 190 Vdc sensor assembly, 500 - 504 Plasma processing system, s19 Vdc sensor.

Claims

1. A plasma processing system having a plasma processing apparatus for generating plasma in a chamber and a data analysis apparatus, wherein the plasma processing apparatus includes: a lower electrode provided in the chamber and connected to an AC power supply; an upper electrode provided in the chamber opposite to the lower electrode and connected to a reference potential; an actual use substrate disposed on the lower electrode side among the upper electrode and the lower electrode; a plurality of potential measurement sensors provided corresponding to a plurality of locations on the formation surface of the actual use substrate, each for measuring the negative potential at the corresponding location on the actual use substrate; the actual use substrate being a substrate to be manufactured; measurement potential data indicating a plurality of measured potentials measured using the plurality of potential measurement sensors is provided to the data analysis apparatus; the data analysis apparatus includes: an in-plane potential distribution data creation unit that obtains substrate-corresponding in-plane potential distribution data indicating the in-plane distribution of the negative potential in the actual use substrate based on the measurement potential data; a data analysis unit that obtains relevance data indicating the relevance between the analysis data reflecting the substrate-corresponding in-plane potential distribution data and the manufacturing content data indicating the manufacturing content for the actual use substrate when the measurement potential data is obtained. A plasma processing system.

2. The plasma processing system according to claim 1, wherein the plasma processing apparatus further includes an upper electrode voltmeter for measuring the potential of the upper electrode, and upper electrode measurement data indicating the upper electrode potential measured by the upper electrode voltmeter is provided to the data analysis apparatus; the data analysis unit in the data analysis apparatus includes: a bias potential distribution creation unit that calculates a plurality of bias potentials that are difference values between the upper electrode potential and the plurality of measured potentials based on the upper electrode measurement data and the substrate-corresponding in-plane potential distribution data, and obtains bias in-plane distribution data indicating the plurality of bias potentials as the analysis data; a correlation data creation unit that analyzes the relevance between the bias in-plane distribution data and the manufacturing content data, and obtains the relevance data indicating the relevance between the bias in-plane distribution data and the manufacturing content data. A plasma processing system.

3. The plasma processing system according to claim 1, wherein the lower electrode has a plurality of divided electrodes, and the plasma processing apparatus includes: ​ It further includes a plurality of lower electrode voltage meters provided corresponding to the plurality of divided electrodes, and the plurality of lower electrode voltage meters measure the voltage of the corresponding divided electrode among the plurality of divided electrodes. The plurality of potential measurement sensors include the plurality of divided electrodes. The actual use substrate is placed on the lower electrode. Plasma processing system.

4. The plasma processing system according to claim 1, The actual use substrate includes a sensor-embedded substrate with the plurality of potential measurement sensors built therein. The actual use substrate is placed on the lower electrode. Plasma processing system.

5. The plasma processing system according to claim 1, The plasma processing apparatus, further includes a sensor substrate with the plurality of potential measurement sensors built therein on the lower electrode. The actual use substrate is placed on the sensor substrate. Plasma processing system.

6. The plasma processing system according to any one of claims 1 to 5, The plurality of potential measurement sensors, are arranged corresponding to a negative potential focus region where it is predicted that the fluctuation of the negative potential of the actual use substrate is large based on the shape of the actual use substrate or the roughness density of a resist pattern to be formed above the actual use substrate. Plasma processing system.

7. The plasma processing system according to any one of claims 1, 2, 4, and 5, The plurality of potential measurement sensors are arranged in a matrix. A first number of scanning lines provided along one of the row direction and the column direction of the plurality of potential measurement sensors, A second number of output lines provided along the other of the row direction and the column direction of the plurality of potential measurement sensors, Each of the plurality of potential measurement sensors, a charge storage capacitor, and includes a charge extraction transistor having one electrode connected to the charge storage capacitor, a control electrode connected to the corresponding scanning line among the first number of scanning lines, and the other electrode connected to the corresponding output line among the second number of output lines. Plasma processing system.

8. The plasma processing system according to any one of claims 1 to 5, The actual use substrate is a circular wafer in plan view, and the wafer is divided into a plurality of dies. Let the maximum square area in the planar structure within the wafer be SX. When the maximum area among the areas of the plurality of dies is DX, The number NV of the plurality of potential measurement sensors, A plasma processing system configured to satisfy the conditional expression {NV ≥ (SX / DX) × 2}. Plasma processing system.

9. The plasma processing system according to any one of Claims 1 to 5, wherein the plasma processing apparatus has a gas supply hole for supplying a manufacturing gas used for manufacturing the actual use substrate into the chamber, and the manufacturing gas includes an etching gas or a film-forming gas. Plasma processing system.

10. The plasma processing system according to any one of Claims 1 to 5, wherein in the data analysis device, the in-plane potential distribution data creation unit determines the presence or absence of abnormalities in the plurality of measured potentials by performing a pass / fail determination using the Mahalanobis-Taguchi method, excludes the measured potentials for which abnormalities are determined from among the plurality of measured potentials, and derives the in-plane potential distribution data corresponding to the substrate. Plasma processing system.

11. The plasma processing system according to any one of Claims 1 to 5, wherein the data analysis unit includes a learning model creation unit, the learning model creation unit performs machine learning using the analysis data as input data and the manufacturing content data as teacher data, and creates a learned model for estimating the manufacturing content of the actual use substrate corresponding to the analysis data, and the relevance data is the learned model. Plasma processing system.

12. The plasma processing system according to Claim 1, wherein the data analysis unit includes a learning model recording unit, the data analysis device receives a learned model from the outside, the learned model is obtained by performing machine learning using the analysis data as input data and the manufacturing content data as teacher data, and is data for estimating the manufacturing content of the actual use substrate corresponding to the analysis data, the learning model recording unit records the learned model, and the relevance data is the learned model. Plasma processing system.

13. The plasma processing system according to Claim 11, further comprising a manufacturing content inference device, wherein the manufacturing content inference device receives desired in-plane distribution data including the in-plane potential distribution data desired during manufacturing of the actual use substrate, and includes an inference unit that uses the learned model to estimate, as inference data, data indicating the manufacturing content for the actual use substrate corresponding to the desired in-plane distribution data. Plasma processing system.

14. A method for manufacturing a trained model using a plasma processing apparatus that generates plasma in a chamber and a learning model creation apparatus, wherein the plasma processing apparatus includes a lower electrode provided in the chamber and connected to an AC power source, an upper electrode provided in the chamber opposite to the lower electrode and connected to a reference potential, an actual use substrate disposed on the lower electrode side among the upper electrode and the lower electrode, a plurality of potential measurement sensors provided corresponding to a plurality of locations on the formation surface of the actual use substrate, each for measuring the negative potential at the corresponding location on the actual use substrate, wherein the actual use substrate is a substrate to be manufactured, measurement potential data indicating a plurality of measured potentials measured using the plurality of potential measurement sensors is obtained, and the learning model creation apparatus (a) obtains substrate-corresponding in-plane potential distribution data indicating the in-plane distribution of the negative potential in the actual use substrate based on the measurement potential data, (b) performs machine learning using the analysis data reflecting the substrate-corresponding in-plane potential distribution data as input data and the manufacturing content data indicating the manufacturing content of the actual use substrate when the analysis data is obtained as teacher data, and creates a trained model for estimating the manufacturing content of the actual use substrate corresponding to the analysis data. A method for manufacturing a trained model.

Images