Control method for radio frequency system used in semiconductor processing equipment

Abstract

This application provides a control method for a radio frequency system in a semiconductor manufacturing equipment, comprising: performing impedance matching at a first output power of a power supply to obtain a first matching state of a matching device; adjusting the power supply to a second output power to perform impedance matching and obtain a second matching state of the matching device; comparing the parameter changes of the matching device in the first and second matching states; and adjusting the parameters of the matching device based on the comparison results, according to the parameters corresponding to the first matching state of the matching device, so that after the power supply returns to the first output power, the power fed into the load unit increases when the plasma coupling efficiency decreases. The control method of this application, by matching the matching device to the changed load impedance of the load unit, can pre-set the matching device, thereby making the system more stable.

Description

[Technical Field]

[0001] This application relates to the field of semiconductor manufacturing, and more specifically, to a control method for a radio frequency system for semiconductor process equipment. [Previous Technology]

[0002] Etching machine systems typically use an RF generator to provide RF energy to a special gas (such as argon, helium, nitrogen, etc.) in a high vacuum state within a chamber. This generates a plasma containing a large number of active particles such as electrons, ions, excited-state atoms, molecules, and free radicals. These active particles interact with the wafer placed in the chamber and exposed to the plasma environment, causing various physical and chemical reactions on the surface of the wafer material. This changes the surface properties of the material, thus completing the etching of the wafer.

[0003] In the prior art, during the semiconductor manufacturing process, some byproducts are gradually generated in the process chamber, which causes the plasma state inside the chamber to change, thereby reducing the plasma feed power. When the plasma feed power decreases, its coupling efficiency will also decrease further, making the plasma easy to extinguish and affecting the stability of the system. [Summary of the Invention]

[0004] This application aims to at least solve the problem in the prior art that the reduced plasma feed power leads to a decrease in coupling efficiency, which easily causes quenching and affects system stability. It proposes a control method for the radio frequency system of semiconductor process equipment.

[0005] To achieve the purpose of this application, a control method for a radio frequency system of a semiconductor process equipment is provided. The radio frequency system includes: a power supply, a matching unit, and a coil. The power supply is electrically connected to the input terminal of the matching unit, and the output terminal of the matching unit is electrically connected to the coil for generating plasma in a process chamber. The coil and the plasma generated therefrom form a load unit. The control method includes: performing impedance matching at a first output power of the power supply to obtain a first matching state of the matching unit; adjusting the power supply to a second output power to perform impedance matching and obtain a second matching state of the matching unit; comparing the parameter changes of the matching unit in the first matching state and the second matching state; and adjusting the parameters based on the comparison results, so that after the power supply is restored to the first output power, the power fed into the load unit increases when the coupling efficiency of the plasma decreases.

[0006] In some embodiments, the matching device includes: a real part adjustment element for matching the real part of the load impedance of the load unit; the step of comparing the parameter changes of the matching device in the first matching state and the second matching state includes: comparing the parameter changes of the real part adjustment element of the matching device in the first matching state and the second matching state.

[0007] In some embodiments, the real part adjustment element is a first adjustable capacitor; comparing the parameter changes of the real part adjustment element of the matching device in the first matching state and the second matching state specifically includes: obtaining a first capacitance value of the first adjustable capacitor in the first matching state of the matching device; obtaining a second capacitance value of the first adjustable capacitor in the second matching state of the matching device; and comparing the magnitude of the first capacitance value and the second capacitance value.

[0008] In some embodiments, the step of presetting the parameters of the matching device based on the comparison result so that the power fed into the load unit increases when the coupling efficiency of the plasma decreases under the first output power of the power supply specifically includes: when the second output power is less than the first output power, if the second capacitance value is greater than the first capacitance value, then the first adjustable capacitor is adjusted to a third capacitance value, the third capacitance value being greater than the first capacitance value; if the second capacitance value is less than the first capacitance value, then the first adjustable capacitor is adjusted to a fourth capacitance value, the fourth capacitance value being less than the first capacitance value.

[0009] In some embodiments, the difference between the first capacitance value and the third capacitance value is 0.2% to 2% of the adjustment range of the first variable capacitor.

[0010] In some embodiments, the difference between the first capacitance value and the fourth capacitance value is 0.2% to 2% of the adjustment range of the first variable capacitor.

[0011] In some embodiments, the difference between the first preset value and the second preset value is greater than or equal to 5W and less than or equal to 20W.

[0012] In some embodiments, the load impedance of the load unit is: where R1 is the equivalent resistance of the coil, ZM is the mutual inductance impedance between the coil and the plasma in the process chamber, R2 is the resistance of the plasma in the process chamber, and ZL2 is the equivalent inductance of the plasma in the process chamber.

[0013] According to a second aspect of this application, a radio frequency system is also disclosed, which is controlled by the above-described control method.

[0014] According to a second aspect of this application, a semiconductor manufacturing apparatus is also disclosed, comprising: a manufacturing chamber, a radio frequency system, and a controller, wherein the controller includes at least one processor and at least one memory, the memory storing a computer program, which, when executed by the processor, implements the control method described above.

[0015] The control method for the radio frequency system of semiconductor process equipment of this application, by matching the load impedance of the changed load unit with the matching device, can determine the relationship between the plasma resistance and the load impedance by the change of the matching device, so as to set the matching device in advance. When the feed power decreases and the plasma coupling efficiency decreases, the matching device can be better matched with the load impedance because it has been set in advance, thereby increasing the feed power. After the feed power increases, plasma quenching can be effectively prevented, thereby making the system more stable.

Implementation Method

[0017] In order to enable those skilled in the art to better understand the technical solution of this application, the control method of the radio frequency system for semiconductor process equipment provided in this application will be described in detail below with reference to the accompanying drawings.

[0018] An etching machine system typically uses an RF generator to provide RF energy to a special gas (such as argon, helium, nitrogen, etc.) in a high vacuum state within a chamber. This generates a plasma containing a large number of active particles such as electrons, ions, excited-state atoms, molecules, and free radicals. These active particles interact with the wafer placed in the chamber and exposed to the plasma environment, causing various physical and chemical reactions on the surface of the wafer material. This changes the surface properties of the material, thus completing the etching of the wafer.

[0019] As shown in Figure 1, the radio frequency (RF) system of a conventional etching machine is connected to an RF power supply 1 via a 50Ω coaxial cable 3. The output impedance of the RF power supply 1 is typically a standard value, for example, 50Ω. To couple the power output from the RF power supply 1 into the process chamber 4, the matching network of the matching network 2 dynamically adjusts its parameters according to the changes in the equivalent impedance of the process chamber 4 to ensure that its input impedance is also equal to 50Ω. Input impedance refers to the impedance presented when viewed from the input port of the matching network 2 towards its internal circuitry (including the matching network) and its connected load (process chamber 4) between the RF power supply 1 and the matching network 2.

[0020] As shown in Figure 2, the existing matching unit 2 consists of four modules: an input sensor 5, a matching circuit unit 6, a control unit 7, and an execution unit 8. The input sensor 5 is connected between the output terminal of the RF power supply 1 and the input terminal of the matching circuit unit 6. The input sensor 5 is used to feed back the measured voltage and current information to the control unit 7. The control unit 7 is used to calculate the information fed back by the input sensor 5 and give an adjustment signal, which is then sent to the execution unit 8. The execution unit 8 is used to control the impedance of the matching circuit unit 6 under the control of the adjustment signal, so as to match the output impedance of the RF power supply 1 with the input impedance of the load (such as the process chamber 4).

[0021] The execution unit 8 of the matching circuit 2 is generally a stepper motor. The stepper motor drives the capacitor adjustment mechanism of the adjustable capacitors C10 and C20 in the matching circuit unit 6. There is a one-to-one correspondence between the rotation position of the stepper motor and the capacitance value of the adjustable capacitor. Therefore, by adjusting the rotation position and direction of the stepper motor, the capacitance value of the variable capacitor can be precisely adjusted, so that the matching circuit 2 can dynamically adjust the input impedance, thereby achieving impedance matching.

[0022] As shown in Figure 3, the matching process of the matching device 2 is as follows: The matching device 2 completes impedance matching by following steps: Step S01: Turn on the RF power supply 1 and output RF power; Step S02: Determine the magnitude and phase of the RF impedance ZM based on the voltage, current, and phase information fed back by the input sensor 5; Step S03: Determine whether the magnitude of the RF impedance ZM is equal to 50Ω. If yes, proceed to step S05; otherwise, proceed to step S04; Step S04: Increase or decrease the capacitance of the adjustable capacitor C20 based on the comparison between the magnitude and 50Ω. Specifically, if the magnitude is greater than 50Ω, increase the capacitance of the adjustable capacitor C20; if the magnitude is less than 50Ω, decrease the capacitance of the adjustable capacitor C20; Step S05: Determine whether the phase of the RF impedance ZM is equal to 0 degrees. If yes, proceed to step S07; otherwise, proceed to step S06. Step S06: Based on the phase-to-0 degree comparison, increase or decrease the capacitance of the adjustable capacitor C10; specifically, if the phase is greater than 0 degrees, decrease the capacitance of the adjustable capacitor C10; if the phase is less than 0 degrees, increase the capacitance of the adjustable capacitor C10. Step S07: Determine whether the plasma process is complete. If yes, the process ends; if no, return to step S02.

[0023] In the prior art, when the impedance magnitude measured by the input sensor 5 of the matching network 2 is 50Ω and the phase is 0 degrees, the positions of the two adjustable capacitors are fixed. However, in practical applications, the phase detected by the input sensor 5 is almost impossible to be exactly equal to 0 degrees. Due to different process conditions and preset conditions, when the adjustable capacitors are finally fixed, the phase detected by the input sensor 5 may be greater than 0 degrees or less than 0 degrees. When the phase is less than 0 degrees, the back-end circuit exhibits capacitive behavior. For capacitors, they typically have the characteristic of clamping voltage, and under external disturbances, their voltage can remain relatively stable, but the current is difficult to keep stable. Therefore, when there is an external disturbance, the current in the entire network will change abruptly, which in turn leads to abrupt changes in the coil current. The change in coil current will cause changes in the magnetic field inside the process chamber, such as the etching chamber, and the change in the magnetic field will affect the plasma state maintained by the magnetic field, ultimately leading to changes in the plasma. Changes in the plasma will cause changes in the impedance of the matching network. The aforementioned changes mainly involve changes in the electromagnetic field, which usually occur on a time scale of microseconds. However, the capacitor used for impedance adjustment inside the matching unit 2 is driven by a motor, which is a mechanical motion with a response time typically on the order of milliseconds. Therefore, the response speed of this mechanical motion is difficult to keep up with the rapid changes in the downstream plasma, easily leading to mismatch. This mismatch can severely affect the stability of the system, causing inhomogeneities in the process or a decline in equipment performance.

[0024] In addition, in the prior art, during the semiconductor manufacturing process, some byproducts are gradually generated in the process chamber, which changes the plasma state inside the chamber and reduces the plasma feed power. When the plasma feed power decreases, its coupling efficiency also decreases further, making the plasma more prone to quenching. Once the plasma quenches, the process gas is no longer conductive, and the resistance increases, resulting in a sudden change in the impedance at the load end. In this case, due to the mechanical motion characteristics of the motor in the matching network, it is difficult to keep up with the rapid change in the downstream impedance in time, and the matching network is prone to sudden mismatch, thus affecting the stability of the system.

[0025] To address the above-mentioned problems, this application discloses a control method for a radio frequency (RF) system in a semiconductor process apparatus. As shown in FIG4, the semiconductor process apparatus includes: a process chamber 10, an RF system 20, and a controller. The controller is used to control the RF system 20, which includes: a power supply (not shown in the figure), a matching unit 21, and a coil 22. The output terminal of the power supply is electrically connected to the input terminal of the matching unit 21, and the output terminal of the matching unit 21 is electrically connected to the coil 22. The power supply loads RF power into the coil 22 through the matching unit 21 to excite the process gas to form plasma.

[0026] In this embodiment, the semiconductor process equipment is an inductively coupled plasma (ICP) etching equipment. The control method of this application will be described in detail below using the radio frequency (RF) system of the ICP etching equipment as an example: As shown in Figures 4 and 5, Figure 5 is an equivalent circuit model of the RF system 20 of this application. The RF system 20 includes: a matching unit 21 and a coil 22. The input terminal of the matching unit 21 is electrically connected to the output terminal of a power supply (not shown in the figure), and the output terminal of the matching unit 21 is electrically connected to the coil 22. The power supply loads RF power into the coil 22 through the matching unit 21. The electric field generated by the coil 22 is used to excite the process gas to form plasma. The coil 22 and the plasma forming load unit are also included.

[0027] In this embodiment, the matching device 21 is an L-type network matching device 21, where L is the inductance in the matching network, C1 is the first adjustable capacitor used to match the real impedance of the load unit, and C2 is the second adjustable capacitor used to match the imaginary impedance of the load unit.

[0028] In the load unit, R1 is the equivalent ohmic resistance of coil 22, M is the mutual inductance between coil 22 and the plasma in process chamber 10, and R2 is the resistance of the plasma in process chamber 10. L1 is the inductance of coil 22, and L2 is the inductance of the plasma in process chamber 10. The impedance Zload of the load unit is: ZM is the mutual inductance impedance between coil 22 and the plasma in process chamber 10, and ZL2 is the equivalent inductive reactance of the plasma in process chamber 10.

[0029] Among them, depending on the size of the chamber, most plasma inductance is about 0.25μH. During the plasma ignition process, the process chamber 10 will go through the process of no plasma to plasma formation, so the value of R2 can vary from 1e6Ω to several Ω.

[0030] As shown in Figure 6, the curves showing the relationship between the real and imaginary parts of the impedance of Zload and the plasma resistance R2 in the process chamber 10 are calculated according to the above formula. The vertical axis represents the real and imaginary parts of the impedance of Zload, in ohms; the horizontal axis represents the plasma resistance R2; the dotted line corresponding to Z_load_r (ohms) is the curve showing the change of the real part of the impedance with the plasma resistance R2, while the solid line corresponding to Z_load_i (ohms) is the curve showing the change of the imaginary part of the impedance with the plasma resistance R2.

[0031] It should be noted that the resistance value of the plasma resistor R2 is different under different process gas ratios, different gas pressures, and different power conditions. However, under the same process gas ratios and gas pressures, the higher the power, the lower the resistance value of the plasma resistor R2 will be. This is because, during the ignition matching process, as the power fed in increases, the number of charged particles excited by the process gas will also increase accordingly, and the conductivity of the process gas will be enhanced. Therefore, the resistance value of R2 will undergo a process of changing from large to small, which is the process of the horizontal axis changing from right to left in Figure 6.

[0032] As shown in Figure 6, Z_load_r undergoes a process of first increasing and then decreasing as the resistance of R2 changes from large to small, i.e., the Z_load_r curve in Figure 6 has a peak point. When the entire etching system experiences disturbances such as unstable gas pressure, the plasma impedance inside the process chamber 10 changes abruptly, and the matching device 21 cannot adjust in time, the feed power will decrease. After the matching stabilizes, if the resistance of R2 of the plasma is to the left of the peak point shown in Figure 6, then as the feed power decreases, the resistance of R2 increases, and Z_load_r will also increase. Conversely, if the resistance of R2 of the plasma is to the right of the peak point shown in Figure 6, then as the feed power decreases, the resistance of R2 increases, and Z_load_r will decrease.

[0033] The following section describes in detail the relationship between the real part of the impedance of Zload and the size of C1 of the matching device 21 during matching, in conjunction with the Smith chart and the matching path of the impedance Zload of the load unit on the Smith chart.

[0034] As shown in the Smith chart of the matcher 21 in Figure 7, it should be noted in advance that the polar coordinate system is used in Figure 7. The right end of the horizontal axis in Figure 7 is infinity, the left end of the horizontal axis is 0, and the midpoint of the horizontal axis (▲ position) is the matching position.

[0035] As shown in Figure 7, when the coordinate point moves along the arc of the same constant resistance circle (for example, along the arc of the first constant resistance circle 41), the real part of the impedance of Zload remains unchanged, while the imaginary part of the impedance of Zload changes. When the coordinate point moves between different constant resistance circles (for example, from the first constant resistance circle 41 to the second constant resistance circle 42), the real part of the impedance changes accordingly. Furthermore, the point where the constant resistance circle intersects the horizontal axis represents the magnitude of the real part of the impedance corresponding to that circle. The closer the intersection point of the constant resistance circle and the horizontal axis is to the right end of the horizontal axis, the smaller the diameter of the constant resistance circle, but the larger the real part of the impedance represented by the constant resistance circle. For example, in Figure 7, the diameter of the first constant resistance circle 41 is smaller than the diameter of the second constant resistance circle 42, but the real part of the impedance represented by the first constant resistance circle 41 is larger than the real part of the impedance represented by the second constant resistance circle 42.

[0036] As shown in Figure 7 , when the coordinate points move along the circular arc of the same isoconductance circle (e.g., along the circular arc of the first isoconductance circle 51 ), the imaginary part of the impedance is basically unchanged, while the real part of the impedance changes;

[0037] When matching the matcher 21, when adjusting the size of C1, that is, when adjusting the impedance real part of Zload, the coordinate points in Figure 7 will move along the arc of the isoconductance circle where it is located, and when adjusting the size of C2 When adjusting the impedance imaginary part of Zload, the coordinate points in Figure 7 will move along the arc of the constant impedance circle where it is located.

[0038] Based on the above rules, Z1 and Z2 represent the two possible impedance values ​​​​of Zload in Figure 5 , respectively, where the impedance value of the × position Z1 is at the coordinate point of the Smith circle graph, and the impedance value of the + position Z2 is at the coordinate point of the Smith circle graph. R is the real part value of the impedance of Z1, R' is the real part value of the impedance of Z2, and R>R', and the imaginary part values ​​​​of the impedances of Z1 and Z2 are both jX.

[0039] Combined with FIG. and at the same time, the coordinate points of Z1 and Z2 are located on the first isoconductance circle 51 , and therefore, the impedance imaginary part values ​​are the same for both.

[0040] When the impedance value of the load unit's impedance Zload is Z1, when the matcher 21 matches the impedance imaginary part of Z1, by adjusting C2, the imaginary part of the front end impedance of the matcher 21 changes along the first certain resistance circle 41 in Figure 7 from the × position on the first equal conductance circle 51 to the ● position on the second equal conductance circle 52 . Thereafter, when matcher 21 matches the impedance real part of Z1, by regulating C1, the real part of the front end impedance of matcher 21 changes from ● position along the second isoconductance circle 52 to the ▲ position in Figure 7 , where, in the process of moving from the ● position to the ▲ position, the distance passed on the second conductance circle 52 represents the magnitude of the change in C1, the longer the distance passed, the smaller the amount of change in C1.

[0041] When the impedance value of the load unit Zload is Z2, when the matching device 21 matches the imaginary part of the impedance of Z2, by adjusting C2, the imaginary part of the impedance of the front end of the matching device 21 changes from the + position on the first equal conductance circle 51 to the ■ position on the second equal conductance circle 52 in FIG. 7; then, when the matching device 21 matches the real part of the impedance of Z2, by adjusting C1, the impedance of the front end of the matching device 21 changes from the ■ position along the second equal conductance circle 52 to the ▲ position in FIG. 7. The distance traveled on the second equal conductance circle 52 during the process of moving from the ■ position to the ▲ position represents the magnitude of the change of C1. The longer the distance traveled, the greater the change of C1, and vice versa.

[0042] As shown in Figure 7, when the matching device 21 is matching, the distance from the + position to the ■ position is equivalent to the distance from the X position to the ● position. These two distances correspond to the magnitude of the change in C2 during the matching process. This indicates that when the matching device 21 is matching Z1 and Z2, the position of the capacitor C2 is basically the same, that is, the change in the real part of the load is not sensitive to C2.

[0043] As for the change of C1, as shown in Figure 7, when the matching device 21 matches the real part of the impedance of Z1, it adjusts C1 to make the real part of the front end impedance change from the ● position to the ▲ position. When the matching device 21 matches the real part of the impedance of Z2, it adjusts C1 to make the real part of the front end impedance change from the ■ position to the ▲ position. It can be seen that the distance of the change in matching Z1 is smaller than the distance of the change in matching Z2. In other words, compared with Z2, the change of C1 is smaller when the matching device 21 matches the real part of the impedance of Z1.

[0044] For the matching circuit 21 of the L-type matching network, when matching the real part of the impedance of Z1, the real part of the front-end impedance is changed from the ● position to the ▲ position by increasing the value of C1; similarly, when matching the real part of the impedance of Z2, the real part of the front-end impedance is changed from the ■ position to the ▲ position by increasing the value of C1. Therefore, the larger the change in C1, the larger the adjusted C1 value, and the smaller the change in C1, the smaller the adjusted C1 value.

[0045] It can be seen from this that for the matching device 21 of the L-type matching network, the real impedance of Z1 is greater than the real impedance of Z2, and the capacitance value of C1 required by the matching device 21 when matching Z1 is less than the capacitance value of C1 required by the matching device 21 when matching Z2.

[0046] It can be assumed that if the real part of the impedance of Zload decreases (for example, Z1 becomes Z2), then C1 in the matching circuit 21 is equivalent to adjusting the capacitance value from matching the real part of the impedance of Z1 to matching the real part of the impedance of Z2. Since the capacitance value of C1 required by the matching circuit 21 when matching Z1 is less than the capacitance value of C1 required when matching Z2, the capacitance value of C1 needs to be increased when the real part of the impedance of Zload decreases. Conversely, if the real part of the impedance of Zload increases (for example, Z2 becomes Z1), the capacitance value of C1 needs to be decreased.

[0047] In other words, substituting the above conclusions into Figure 6, the real part of the impedance Zload is Z_load_r. The larger Z_load_r is, the smaller C1 is. Similarly, the smaller Z_load_r is, the larger C1 is, and conversely, the smaller C1 is, the larger Z_load_r is, and the larger C1 is, the smaller Z_load_r is.

[0048] Based on the above principle, as shown in Figure 8, the control method disclosed in this application includes: Step S10: Under the first output power of the power supply, impedance matching is completed and the first matching state of the matching device 21 is obtained.

[0049] Step S20: Adjust the power supply to the second output power, complete impedance matching, and obtain the second matching state of the matching device 21.

[0050] Step S30: Compare the parameter changes of the matcher 21 in the first matching state and the second matching state; Step S40: Based on the comparison result, adjust the parameters of the matcher 21 based on the parameters corresponding to the first matching state so that after the power supply is restored to the first output power, the power fed into the load unit increases when the plasma coupling efficiency decreases.

[0051] The stability of plasma is closely related to the characteristic that its coupling efficiency changes with its feed power. During semiconductor manufacturing, some byproducts are gradually generated in the process chamber, which leads to changes in the plasma state inside the chamber, resulting in a decrease in the plasma feed power. When the feed power decreases, but the plasma coupling efficiency (i.e., the proportion of feed power absorbed by the plasma) increases, the plasma is in a negative feedback state, and the matching system is in a stable state. Conversely, when the plasma feed power decreases and the plasma coupling efficiency further decreases, the plasma is in a positive feedback state, and the matching system is prone to mismatch or even quenching. Therefore, if the state of the plasma can be determined in advance and the entire matching system is pre-set to a negative feedback state, the stability of the matching system can be greatly improved.

[0052] The specific method for predicting the state of the plasma in advance is as follows: by changing the output power of the power supply, a controllable load change is actively introduced, that is, the feed power of the load unit changes accordingly with the output power of the power supply, thereby causing a change in the resistance value of the plasma R2, and thus causing a change in the Zload of the load unit. After detecting the change in Zload, the matching device 21 will automatically adjust its parameters (the size of the first adjustable capacitor C1) to match the changed Zload. During this matching process, by monitoring the direction of change of the parameters (the size of the first adjustable capacitor C1), the change of the real part of the impedance in Zload can be determined, thereby determining whether the resistance value of the plasma R2 is located to the left or right of the peak point in Figure 6. Specifically, taking the matching network of the matching unit 21 as an L-shaped network as an example, if the output power of the power supply decreases, the capacitance of the first adjustable capacitor C1 increases, and the real part of the impedance in Zload decreases. That is, as the resistance R2 of the plasma increases, the real part of the impedance in Zload decreases. At this time, the resistance R2 of the plasma is located to the right of the peak point in Figure 6. If the output power of the power supply decreases, the capacitance of the first adjustable capacitor C1 decreases, and the real part of the impedance in Zload increases. That is, as the resistance R2 of the plasma increases, the real part of the impedance in Zload increases. At this time, the resistance R2 of the plasma is located to the left of the peak point in Figure 6.

[0053] The specific way to pre-set the entire matching system to a negative feedback state is as follows: Taking the matching network of the matcher 21 as an L-shaped network as an example, if the resistance R2 of the plasma is located to the right of the peak point in Figure 6 (manifested as an increase in the capacitance of the first adjustable capacitor C1), the parameter (the first adjustable capacitor C1) can be proactively set to be slightly larger than the value at the matching point based on the first matching state of the matcher 21. When the feed power decreases due to external disturbances, the resistance R2 of the plasma increases, and the real part of the impedance in Zload decreases, the plasma coupling efficiency increases due to the pre-increased parameter, and the entire matching system is in a negative feedback state, which can effectively prevent plasma extinguishing and make the system more stable. If the resistance R2 of the plasma is located to the left of the peak point in Figure 6 (manifested as a decrease in the capacitance of the first adjustable capacitor C1), the parameter (the first adjustable capacitor C1) can be proactively set to be slightly smaller than the value at the matching point based on the first matching state of the matcher 21. When the feed power decreases due to external disturbances, causing the resistance R2 of the plasma to increase and the real part of the impedance in Zload to increase, the coupling efficiency of the plasma increases due to the parameter being reduced in advance. The entire matching system is in a negative feedback state, which can effectively prevent plasma extinguishing and make the system more stable.

[0054] As can be seen from the above, by judging the state of the plasma in advance and setting the parameters of the matching device 21 accordingly, a controllable impedance margin can be prepared in advance for the real part of the impedance in Zload to cope with the dynamic adjustment when the power disturbance occurs. Thus, when the feed power of the plasma decreases due to external disturbance, the matching device 21 can better match the Zload after the impedance changes, thereby increasing the coupling efficiency of the plasma. This can realize that the entire matching system is in a negative feedback state, effectively preventing plasma quenching, and thus making the system more stable.

[0055] Step S30 further includes the following steps: Step S31: In the first matching state of the matching device 21, obtain the first capacitance value of the first adjustable capacitor C1; Step S32: In the first matching state of the matching device 21, obtain the second capacitance value of the first adjustable capacitor C1; Step S33: Compare the magnitudes of the first capacitance value and the second capacitance value.

[0056] Based on the aforementioned conclusion, by comparing the first capacitance value and the second capacitance value of the first adjustable capacitor C1, the change of the real part of the impedance of Zload can be reflected. Thus, based on the change of the real part of the impedance of Zload, it can be determined whether the resistance value of resistor R2 is located to the left or right of the peak point in Figure 6, and then the trend of the change of the real part of the impedance of Zload when the input power decreases can be predicted.

[0057] Taking the matching network of the matching unit 21 as an L-type network as an example, when the second output power is less than the first output power, the feed power decreases and the resistance of R2 increases. When the matching is stable, if the second capacitance value is greater than the first capacitance value, that is, the capacitance value of the first adjustable capacitor C1 increases, it means that as the resistance of R2 increases, the real part of the impedance of Zload, that is, Z_load_r, decreases. Therefore, the resistance of the plasma R2 is to the right of the peak point in Figure 6. Conversely, if the second capacitance value is less than the first capacitance value, that is, the capacitance value of the first adjustable capacitor C1 decreases, it means that as the resistance of R2 increases, the real part of the impedance of Zload, that is, Z_load_r, increases. Therefore, the resistance of the plasma R2 is to the left of the peak point in Figure 6.

[0058] Therefore, based on the above principle, step S40 further includes the following steps: Step S41: If the second capacitance value is greater than the first capacitance value, the capacitance value of the first adjustable capacitor C1 is adjusted to the third capacitance value, which is greater than the first capacitance value.

[0059] When the second capacitance value is greater than the first capacitance value, it means that as the resistance of R2 increases, Z_load_r decreases. The resistance of the plasma R2 is to the right of the peak point in Figure 6. The capacitance value of the first adjustable capacitor C1 can be set to a third capacitance value that is slightly larger than the first capacitance value. When the input power decreases, since the capacitance value of the first adjustable capacitor C1 is preset to the third capacitance value, the Z_load_r after the input power decreases will be more matched with the capacitance value of the first adjustable capacitor C1, thereby increasing the input power and avoiding the occurrence of slow response and mismatch of the actuator, making the matching more stable.

[0060] Step S42: If the second capacitance value is less than the first capacitance value, adjust the capacitance value of the first adjustable capacitor C1 to the fourth capacitance value, which is less than the first capacitance value.

[0061] When the second capacitance value is less than the first capacitance value, it means that as the resistance of R2 increases, Z_load_r increases. The resistance of the plasma R2 is to the left of the peak point in Figure 6. The capacitance value of the first adjustable capacitor C1 can be set to a slightly smaller fourth capacitance value. When the input power decreases, since the capacitance value of the first adjustable capacitor C1 is preset to the fourth capacitance value, the Z_load_r after the input power decreases will be more matched with the capacitance value of the first adjustable capacitor C1, thereby increasing the input power and avoiding the occurrence of slow response and mismatch of the actuator, making the matching more stable.

[0062] It should be noted that in this embodiment, the difference between the first output power and the second output power is greater than or equal to 5W and less than or equal to 20W. That is to say, the output power of the power supply cannot be reduced too much during the adjustment process in order to ensure the stability of the system.

[0063] It should also be noted that if the difference between the third and fourth capacitance values ​​and the first capacitance value is too large, the reflected power will also be greater, thereby affecting the feed efficiency. Therefore, in order to ensure the feed efficiency, the difference between the first capacitance value and the third capacitance value is 0.2% to 2% of the adjustment range of the first variable capacitor. The difference between the first capacitance value and the fourth capacitance value is 0.2% to 2% of the adjustment range of the first variable capacitor.

[0064] It should be noted that in the above embodiment, the matching device 21 is an L-shaped matching network, but this is not limiting. That is to say, the matching device 21 can also use other matching networks, as long as it has a real part adjustment element for matching the real part of the load impedance of the load unit. The magnitude of the real part of the impedance of the load unit can be changed by adjusting the real part adjustment element.

[0065] In another embodiment shown in FIG9, a semiconductor process apparatus is also disclosed, the structure of which is the same as that in the above embodiment, the only difference being that, in this embodiment, the matching device 21 is a T-type network. In the T-type network matching device 21, L is still the inductor in the matching network, C1 is still the first adjustable capacitor used to match the real part of the impedance of the load unit, and C2 is still the second adjustable capacitor used to match the imaginary part of the impedance of the load unit.

[0066] However, unlike the matching device 21 of the L-type matching network, for the matching device 21 of the T-type matching network, when matching the real part of the impedance of Z1, the real part of the impedance at the front end of the matching device 21 is changed from the ● position to the ▲ position by decreasing the value of C1; similarly, when matching the real part of the impedance of Z2, the real part of the impedance at the front end of the matching device 21 is changed from the ■ position to the ▲ position by decreasing the value of C1. Therefore, the larger the change in C1, the smaller the adjusted C1 value, and the smaller the change in C1, the larger the adjusted C1 value.

[0067] It can be seen from this that for the matching device 21 of the T-type matching network, the real part of the impedance of Z1 is greater than the real part of the impedance of Z2, and the capacitance value of C1 required when matching device 21 matches Z1 is greater than the capacitance value of C1 required when matching device 21 matches Z2.

[0068] It can be assumed that if the real part of the impedance of Zload decreases (for example, Z1 becomes Z2), then C1 in the matching circuit 21 is equivalent to adjusting from the capacitance value matching the real part of Z1 to the capacitance value matching the real part of Z2. Since the capacitance value of C1 required by the matching circuit 21 when matching Z1 is greater than the capacitance value of C1 required when matching Z2, the capacitance value of C1 also needs to be reduced when the real part of the impedance of Zload decreases. Conversely, if the real part of the impedance of Zload increases (for example, Z2 becomes Z1), the capacitance value of C1 needs to be increased.

[0069] In other words, substituting the above conclusion into Figure 6, the real part of the impedance of Zload is Z_load_r. When Z_load_r is larger, C1 is larger. Similarly, the smaller Z_load_r is, the smaller C1 is. Conversely, the larger C1 is, the larger Z_load_r is, and the smaller C1 is, the smaller Z_load_r is. Based on the matching device 21 of the T-type network in another embodiment shown in Figure 9, this application also discloses a control method, which is basically the same as the control method in the embodiment shown in Figure 5. The only difference is that the result of determining whether the resistance value of resistor R2 is located to the left or right of the peak point in Figure 6 by comparing the first capacitance value and the second capacitance value of the first adjustable capacitor C1 is different.

[0070] Specifically, when using a matcher 21 with a T-type matching network, if the second output power is less than the first output power, the input power decreases and the resistance of R2 increases. When the matching is stable, if the second capacitance value is greater than the first capacitance value, that is, the capacitance value of the first adjustable capacitor C1 increases, it means that as the resistance of R2 increases, the real part of the impedance of Zload, that is, Z_load_r, increases. Therefore, the resistance of the plasma R2 is to the left of the peak point in Figure 6. The capacitance value of the first adjustable capacitor C1 can be set to a third capacitance value that is slightly larger than the first capacitance value. When the input power decreases, since the capacitance value of the first adjustable capacitor C1 is preset to the third capacitance value, the Z_load_r after the input power decreases will be more matched with the capacitance value of the first adjustable capacitor C1, thereby increasing the input power and avoiding the occurrence of slow response of the actuator and easy mismatch, making the matching more stable. Conversely, if the second capacitance value is smaller than the first capacitance value, that is, the capacitance value of the first adjustable capacitor C1 becomes smaller, it means that as the resistance value of R2 increases, the real part of the impedance of Zload, that is, Z_load_r, decreases. Therefore, the resistance value of the plasma R2 is to the right of the peak point in Figure 6. The capacitance value of the first adjustable capacitor C1 can be set to a slightly smaller fourth capacitance value. When the input power decreases, since the pre-set capacitance value of the first adjustable capacitor C1 is the fourth capacitance value, it will make Z_load_r after the input power decreases more matched with the capacitance value of the first adjustable capacitor C1, thereby increasing the input power and avoiding the occurrence of slow response and mismatch of the actuator, making the matching more stable.

[0071] The semiconductor manufacturing equipment of this application includes a controller comprising at least one processor and at least one memory, wherein the memory stores a computer program, and the computer program is executed by the processor to implement the above-described control method.

[0072] As shown in FIG4, the semiconductor process equipment further includes an inlet assembly, a lower electrode assembly 30, and a controller. The controller includes at least one processor and at least one memory, the memory storing a computer program that, when executed by the processor, implements the method of any of the above embodiments.

[0073] For example, the controller can be a host computer or a slave computer. The controller can open the valve of the air intake assembly to introduce the corresponding process gas into the process chamber 10; the controller can also control the opening and closing degree of the valve of the air intake assembly to control the flow rate of the process gas. The controller can also control the air extraction assembly to extract air from the process chamber 10, thereby controlling the pressure inside the process chamber 10 and discharging reaction byproducts.

[0074] The lower electrode assembly 30 includes a wafer carrier 31, a lower RF power supply, and a lower matching unit 32. The controller is also used to control the lower RF power supply to provide lower electrode power to the lower electrode of the wafer carrier 31 through the lower matching unit 32, so that the lower electrode of the wafer carrier 31 provides RF bias voltage to attract plasma above the object to be etched and bombard the object to be etched.

[0075] It is understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of this application, and this application is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and substance of this application, and these modifications and improvements are also considered to be within the scope of protection of this application. [Simplified Explanation of the Diagram]

[0016] Figure 1 is a schematic diagram of a process etching machine radio frequency system in the prior art; Figure 2 is a schematic diagram of a matching device in the prior art; Figure 3 is a matching flowchart of a matching device in the prior art; Figure 4 is a schematic diagram of a semiconductor process equipment according to an embodiment of this application; Figure 5 is an equivalent circuit model of a radio frequency system according to an embodiment of this application; Figure 6 is a graph showing the variation of the real and imaginary parts of the load impedance Zload with the plasma resistance R2 according to an embodiment of this application; Figure 7 is a Smith chart of a matching device according to an embodiment of this application; Figure 8 is a flowchart of the control method according to an embodiment of this application; Figure 9 is an equivalent circuit model of a radio frequency system according to another embodiment of this application.

Claims

1. A control method for a radio frequency (RF) system in a semiconductor manufacturing equipment, the RF system comprising: The system comprises a power supply, a matching device, and a coil. The power supply is electrically connected to the input terminal of the matching device, and the output terminal of the matching device is electrically connected to the coil. The matching device is used to excite process gas in a process chamber to form plasma. The coil and the plasma form a load unit. The matching device includes a first variable capacitor for matching the real part of the load impedance of the load unit. The control method includes: completing impedance matching at a first output power of the power supply to obtain a first matching state of the matching device; adjusting the power supply to a second output power to complete impedance matching and obtain a second matching state of the matching device; comparing the parameter changes of the matching device in the first and second matching states, including: obtaining a first capacitance value of the first variable capacitor in the first matching state of the matching device; obtaining a second capacitance value of the first variable capacitor in the second matching state of the matching device; comparing the magnitudes of the first and second capacitance values; and adjusting the parameters based on the comparison results, so that after the power supply returns to the first output power, the power fed into the load unit increases when the coupling efficiency of the plasma decreases. If the second output power is less than the first output power, and the second capacitance value is greater than the first capacitance value, then the first variable capacitor is adjusted to a third capacitance value, which is greater than the first capacitance value; if the second capacitance value is less than the first capacitance value, then the first variable capacitor is adjusted to a fourth capacitance value, which is less than the first capacitance value.

2. The control method as described in claim 1, wherein, The difference between the first capacitance value and the third capacitance value is 0.2% to 2% of the adjustment range of the first variable capacitor.

3. The control method as described in claim 1, wherein, The difference between the first capacitance value and the fourth capacitance value is 0.2% to 2% of the adjustment range of the first variable capacitor.

4. The control method as described in claim 1, wherein, The difference between the first output power and the second output power is greater than or equal to 5W and less than or equal to 20W.

5. The control method as described in claim 2, wherein, The load impedance of the load unit is: where R1 is the equivalent resistance of the coil, ZM is the mutual inductance impedance between the coil and the plasma in the process chamber, R2 is the resistance of the plasma in the process chamber, and ZL2 is the equivalent inductance of the plasma in the process chamber.

6. A radio frequency system controlled by any one of claims 1 to 5.

7. A semiconductor manufacturing apparatus, comprising: A process chamber, a radio frequency system, and a controller, the controller including at least one processor and at least one memory storing a computer program that, when executed by the processor, implements the control method as described in any one of claims 1 to 5.

Images