KR20150039725A - Impedance Matching Method And Impedance Matching System - Google Patents

Abstract

Provided in the present invention are an impedance matching apparatus and an impedance matching method. A variable reactance impedance matching network is arranged between frequency variable RF power changing driving frequency (f) and charge. The impedance matching method of the variable reactance impedance matching network comprises the step of controlling changing amount of capacitance or inductance of variable reactive device of the impedance matching network with a function of difference between target driving frequency (ft) and the driving frequency (f).

Description

TECHNICAL FIELD [0001] The present invention relates to an impedance matching method and an impedance matching method,

The present invention relates to an RF power system, and more particularly, to an RF power system including a frequency variable RF power source.

In the field of plasma processing, such as the manufacture of semiconductors or flat panel displays, an RF power generator provides RF power to a load to discharge capacitively coupled or inductively coupled plasma into the interior of the plasma chamber.

The load is a dynamic load that varies with time including the plasma. Therefore, there is a need for a method for minimizing the reflected wave reflected from the load between the RF power generator and the load to deliver maximum power to the load due to the dynamic load.

For impedance matching between the RF power generator and the load, typically two methods are used. One is to place a separate impedance matching network comprising variable elements between the RF power generator and the load. And the other is to change the frequency of the RF power generator to perform impedance matching.

In the case of a variable element impedance matching network, the impedance matching network utilizes at least two variable reactive elements. The variable reactive element may be a variable capacitor or a variable inductor. The variable reactive element is typically driven by a motor. The variable element has a very wide maximum / minimum ratio of 10 or more so that the variable element impedance matching network can perform impedance matching for a wide range of load impedances. Therefore, even when the state of the plasma is significantly changed, the variable element impedance matching network can perform impedance matching. However, the variable element impedance matching network requires a matching time of several hundred milliseconds to several seconds depending on the driving speed of the motor.

On the other hand, when impedance matching is performed by frequency tuning or frequency tuning, a typical frequency variable range is about 10 percent, and the impedance range of the load capable of impedance matching is very narrow. Thus, if the state of the plasma is severely changed, the frequency tuning matching method can not perform impedance matching. On the other hand, the matching time to reach impedance matching is very short, on the order of a few microseconds to a few milliseconds.

Atomic layer deposition (ALD), one of the plasma processing processes, requires repetition of short process steps. In addition, TSV (Through Silicon Via) process requires repeated deposition and etching processes. In addition, recent deposition or etching processes use multi-step recipes that change process conditions while maintaining RF power. In order to satisfy these new process conditions, the RF power generator and the impedance matching network must perform impedance matching at tens of milliseconds or less. Particularly, when performing the pulse plasma process, the impedance matching should be performed at a few microseconds to several tens of microseconds or less. Accordingly, there is a demand for a new impedance matching method in which the impedance is matched at a range of a few microseconds to several tens of microseconds at a wide range of plasma loads or the reflection frequency is reduced within a predetermined range and the driving frequency is fixed.

SUMMARY OF THE INVENTION The present invention provides an impedance matching system that performs high-speed impedance matching.

The variable reactance impedance matching network according to an embodiment of the present invention is disposed between a frequency variable RF power source for changing the driving frequency f and a load. The impedance matching method of the variable reactance impedance matching network controls the variation of the capacitance or inductance of the variable reactive element of the impedance matching network as a function of the difference between the target driving frequency f _t and the driving frequency f .

In one embodiment of the present invention, the variable reactive element includes a first capacitor and a second capacitor, and the first capacitance of the first capacitor and the second capacitance of the second capacitor The change amount dC1 (dC2) of (C2)

Satisfy the conditions, in which, dω can be a difference between the target drive angular frequency (ω = 2πf _{t t)} and the driving frequency (ω = 2πf).

In one embodiment of the present invention, the variable reactive element includes a first capacitor and a second capacitor, and the first capacitance of the first capacitor and the second capacitance of the second capacitor The change amount dC1 (dC2) of (C2)

Satisfy the conditions, in which, and K1 is a constant, and K2 is a constant, dω can be a difference between the target drive angular frequency (ω = 2πf _{t t)} and the driving frequency (ω = 2πf).

In one embodiment of the present invention, the impedance matching network includes an L-type, an inverted L-type, a T-type, and a pi- ). ≪ / RTI >

In an embodiment of the present invention, if the difference between the target drive frequency f _t and the drive frequency f has a positive value, the change amount of the capacitance or inductance of the variable reactive element has a negative value . If the difference between the target driving frequency f _t and the driving frequency f has a negative value, the variation of the capacitance or inductance of the variable reactive element can be controlled to have a positive value.

In an embodiment of the present invention, the capacitance or the change amount of the inductance of the variable reactive element may additionally depend on a reflection coefficient or impedance function for impedance matching.

In one embodiment of the present invention, the variable reactive element includes a first capacitor and a second capacitor, and the first capacitance of the first capacitor and the second capacitance of the second capacitor The change amount dC1 (dC2) of (C2)

Where A and B are variables dependent on the reflection coefficient or impedance and dω may be the difference between the target drive angular frequency ω _t = 2πf _t and the drive frequency ω = 2πf.

In one embodiment of the present invention,

Wherein g1 is a first weighting function and g2 is a second weighting function and the first weighting function has a large value when the reflection coefficient is large and a small value when the reflection coefficient is small, The second weighting function has a small value when the reflection coefficient is large, and a large value when the reflection coefficient is small.

In one embodiment of the present invention, the frequency variable RF power source can perform impedance matching by changing the driving frequency.

In one embodiment of the present invention, the frequency variable RF power source can increase the driving frequency when the imaginary part of the reflection coefficient has a positive value and decrease the driving frequency when the imaginary part of the reflection coefficient has a negative value .

In one embodiment of the present invention, the frequency variable RF power source may perform impedance matching by scanning a driving frequency.

In one embodiment of the present invention, the capacitance or the change amount of the inductance of the variable reactive element depends on a function of reflection coefficient or impedance for impedance matching: extracting a characteristic vector; Transforming an element vector representing a reactance of the variable reactive element into an analysis vector using a predetermined conversion matrix and displaying a characteristic vector in an analytical coordinate system having the analysis vector as a coordinate axis; Analyzing the characteristic vector in the analysis coordinate system and extracting a displacement vector for impedance matching; Transforming the displacement vector into a transformed element vector using the transform matrix; And extracting a change amount of the capacitance or the inductance using the converted element vector.

According to an embodiment of the present invention, there is provided a method of impedance matching, comprising: measuring an electrical characteristic at an output of a frequency variable RF power source and changing a driving frequency using measured electrical characteristics to perform a first impedance matching; And arranging an impedance matching network including a variable reactive element between the frequency variable RF power source and the load to change a capacitance or an inductance of the variable reactive element. The change amount of the capacitance or inductance of the variable reactive element is given as a function of the difference between the target drive frequency f _t and the drive frequency f.

In one embodiment of the present invention, the impedance matching network includes an L-type, an inverted L-type, a T-type, and a pi- ). ≪ / RTI >

In one embodiment of the present invention, the variable reactive element includes a first capacitor and a second capacitor, and the first capacitance of the first capacitor and the second capacitance of the second capacitor The change amount dC1 (dC2) of (C2)

It is given by, where, dω can be a difference between the target drive angular frequency (ω = 2πf _{t t)} and the driving frequency (ω = 2πf).

In one embodiment of the present invention, the method may further include calculating a predicted driving frequency f _p and providing the calculated driving power to the frequency variable RF power source.

In one embodiment of the present invention, the variable reactive element includes a first capacitor and a second capacitor, and the predicted driving angular frequency? _P is

Is given and, ω _p is a prediction driving angular frequency _{(ω p = 2πf p),} ω is angular frequency (ω = 2πf), C1 driving the first large Passage capacitance of the first capacitor to, and C2 is the second And a second capacitance of the capacitor.

In an embodiment of the present invention, the amount of change in capacitance or inductance of the variable reactive element may depend on a function of reflection coefficient or impedance for impedance matching.

In one embodiment of the present invention, the capacitance or the change amount of the inductance of the variable reactive element depends on a function of reflection coefficient or impedance for impedance matching: extracting a characteristic vector; Transforming an element vector representing a reactance of the variable reactive element into an analysis vector using a predetermined conversion matrix and displaying a characteristic vector in an analytical coordinate system having the analysis vector as a coordinate axis; Analyzing the characteristic vector in the analysis coordinate system and extracting a displacement vector for impedance matching; Transforming the displacement vector into a transformed element vector using the transform matrix; And extracting a change amount of the capacitance or the inductance using the converted element vector.

In one embodiment of the present invention, the variable reactive element includes a first capacitor and a second capacitor, and the first capacitance of the first capacitor and the second capacitance of the second capacitor The change amount dC1 (dC2) of (C2)

Where A and B are variables dependent on the reflection coefficient or impedance and dω may be the difference between the target drive angular frequency ω _t = 2πf _t and the drive frequency ω = 2πf.

In one embodiment of the present invention,

Wherein g1 is a first weighting function and g2 is a second weighting function and the first weighting function has a large value when the reflection coefficient is large and a small value when the reflection coefficient is small, The second weighting function has a small value when the reflection coefficient is large, and a large value when the reflection coefficient is small.

The impedance matching network according to an embodiment of the present invention is disposed between a frequency variable RF power source for changing a driving frequency f and a load. The impedance matching method of the variable reactance impedance matching network changes the capacitance or inductance of the variable reactive element of the impedance matching network to induce the variable frequency RF power source to operate at the target driving frequency.

In an embodiment of the present invention, the amount of change in capacitance or inductance of the variable reactive element of the impedance matching network may be controlled as a function of the difference between the target drive frequency f _t and the drive frequency f .

In one embodiment of the present invention, the impedance matching network may further include calculating a predicted driving frequency f _p and providing the predicted driving frequency f _p to the frequency variable RF power source.

An RF power system in accordance with an embodiment of the present invention includes an impedance matching network that delivers the output of a frequency variable RF power source and a frequency variable RF power source to a load. The method of impedance matching of an RF power system includes measuring a first electrical characteristic of an output terminal of a frequency variable RF power supply; Checking the impedance matching state using the first electrical characteristic in the frequency variable RF power source; Changing a driving frequency of the frequency variable RF power source; Measuring a second electrical characteristic in the impedance matching network; Inspecting the impedance matching state using the second electrical characteristic in the impedance matching network and checking whether the driving frequency is a target driving frequency; Calculating a first variation of the inductance or capacitance of the variable reactive element for impedance matching when impedance matching is not performed in the impedance matching network; Calculating a second change amount of an inductance or a capacitance turn of the variable reactive element for changing the drive frequency when the drive frequency in the impedance matching network does not match the target drive frequency; And calculating a total change amount due to the first change amount and the second change amount, and controlling the variable reactive element using the total change amount.

In one embodiment of the present invention, the first amount of change may be given as a function of the difference between the driving frequency and the target driving frequency.

In one embodiment of the present invention, the second variation amount may be a function depending on the impedance or reflection coefficient.

In one embodiment of the present invention, the driving vector or the total variation amount for controlling the variable reactive element includes a product of the first variation amount and the first weighting function, the second variation amount and the second weighting function, The first weighting function has a large value when the reflection coefficient is large, a small value when the reflection coefficient is small, a small value when the reflection coefficient is large, and a small value when the reflection coefficient is small. .

In one embodiment of the present invention, the step of changing the driving frequency of the frequency variable RF power supply includes: calculating a first reflection coefficient using the first electrical characteristic; And increasing the frequency when the imaginary part of the first reflection coefficient has a positive value and decreasing the frequency when the imaginary part of the first reflection coefficient has a negative value.

In one embodiment of the present invention, the step of calculating the first variation includes: extracting a characteristic vector; Transforming an element vector representing a reactance of the variable reactive element into an analysis vector using a predetermined conversion matrix and displaying a characteristic vector in an analytical coordinate system having the analysis vector as a coordinate axis; Analyzing the characteristic vector in the analysis coordinate system and extracting a displacement vector for impedance matching; Transforming the displacement vector into a transformed element vector using the transform matrix; And extracting a change amount of the capacitance or the inductance using the converted element vector.

An RF power system in accordance with an embodiment of the present invention includes an impedance matching network that delivers the output of a frequency variable RF power source and a frequency variable RF power source to a load. A method of impedance matching of an RF power system, comprising: measuring a first electrical characteristic of an output terminal of a frequency variable RF power source; Checking the impedance matching state using the first electrical characteristic in the frequency variable RF power source; Changing a driving frequency of the frequency variable RF power source; Checking whether the driving frequency is a target driving frequency; Calculating a first variation of the inductance or capacitance of the variable reactive element for impedance matching when impedance matching is not performed; Calculating a second variation amount of the inductance or capacitance of the variable reactive element for changing the driving frequency when the driving frequency does not match the target driving frequency; And calculating a total change amount due to the first change amount and the second change amount, and controlling the variable reactive element using the total change amount.

In one embodiment of the present invention, the first amount of change may be given as a function of the difference between the driving frequency and the target driving frequency.

In one embodiment of the present invention, the second variation amount may be a function depending on the impedance or reflection coefficient.

In one embodiment of the present invention, the drive vector for controlling the variable reactive element includes a product of the first variation and the first weighting function, the second variation and the second weighting function, and the first weighting function Function has a large value when the reflection coefficient is large and a small value when the reflection coefficient is small and the second weighting function has a small value when the reflection coefficient is large and a large value when the reflection coefficient is small.

In one embodiment of the present invention, the step of changing the driving frequency of the frequency variable RF power supply includes: calculating a first reflection coefficient using the first electrical characteristic; And increasing the frequency when the imaginary part of the first reflection coefficient has a positive value and decreasing the frequency when the imaginary part of the first reflection coefficient has a negative value.

In one embodiment of the present invention, the step of calculating the first variation includes: extracting a characteristic vector; Transforming an element vector representing a reactance of the variable reactive element into an analysis vector using a predetermined conversion matrix and displaying a characteristic vector in an analytical coordinate system having the analysis vector as a coordinate axis; Analyzing the characteristic vector in the analysis coordinate system and extracting a displacement vector for impedance matching; Transforming the displacement vector into a transformed element vector using the transform matrix; And extracting a change amount of the capacitance or the inductance using the converted element vector.

In one embodiment of the present invention, the impedance matching network may further include calculating a predicted driving frequency f _p and providing the predicted driving frequency f _p to the frequency variable RF power source.

The impedance matching network according to an embodiment of the present invention is disposed between a frequency variable RF power source outputting RF power having a predetermined frequency variable range and a load. The impedance matching network is characterized in that the impedance matching network comprises at least two variable reactive elements and the variation in inductance or capacitance of the variable reactive element is given as a function of the difference between the driving frequency and the target driving frequency.

The matching sensor unit may further include a matching sensor unit disposed between the impedance matching network and the frequency variable RF power source for measuring an electrical characteristic in the load direction in the impedance matching network, Characteristics, and the matching sensor unit can measure the driving frequency transmitted to the load.

The power sensor unit may further include a power source sensor unit disposed between the impedance matching network and the frequency variable RF power source for measuring an electrical characteristic in the load direction in the impedance matching network, Characteristics, and the impedance matching network may be provided with a driving frequency from the frequency variable RF power source.

In one embodiment of the present invention, the impedance matching network may be provided with electrical characteristics in the direction from the output terminal of the variable frequency RF power source to the load from the frequency variable RF power source.

In one embodiment of the present invention, the impedance matching network may provide an expected driving frequency for impedance matching to the frequency variable RF power source.

The RF power system according to an exemplary embodiment of the present invention includes a frequency variable RF power source having a predetermined frequency variable range and changing a driving frequency for impedance matching; And an impedance matching network that receives an output from the frequency variable RF power source and transfers the output to a load. The impedance matching network changes the capacitance or inductance of the variable reactive element of the impedance matching network so that the frequency variable RF power source operates to operate at the target driving frequency.

In one embodiment of the present invention, the impedance matching network may further include a matching sensor unit disposed between the impedance matching network and the frequency variable RF power source to measure an electrical characteristic in the load direction in the impedance matching network, The matching sensor unit may calculate the impedance or reflection coefficient using the electrical characteristics, and the matching sensor unit may measure the driving frequency transmitted to the load.

In one embodiment of the present invention, the frequency variable RF power source further includes a power source sensor unit disposed between the impedance matching network and the frequency variable RF power source to measure an electrical characteristic in the load direction in the impedance matching network, The power source sensor unit may calculate an impedance or a reflection coefficient using the electrical characteristics, and the impedance matching network may receive a driving frequency from the frequency variable RF power source.

In one embodiment of the present invention, the impedance matching network may be provided with the impedance or reflection coefficient in the direction from the output of the frequency variable RF power source toward the load from the frequency variable RF power source.

In one embodiment of the present invention, the impedance matching network may provide an expected driving frequency for impedance matching to the frequency variable RF power source.

An electrical apparatus according to an embodiment of the present invention includes two or more variable reactive elements that transmit the output of a frequency variable RF power source to a load. Wherein the variable inductance or the variation of the capacitance of the variable reactive element includes a function of a difference between a driving frequency of the frequency variable RF power supply and a target driving frequency and the target frequency is within the frequency variable range of the frequency variable RF power generator exist.

An RF power system in accordance with an embodiment of the present invention includes an impedance matching network that delivers the output of a frequency variable RF power source and a frequency variable RF power source to a load. The method of impedance matching of an RF power system includes measuring a first electrical characteristic of an output terminal of a frequency variable RF power supply; Checking the impedance matching state using the first electrical characteristic in the frequency variable RF power source; Changing a driving frequency of the frequency variable RF power source; Measuring a second electrical characteristic in the impedance matching network; Inspecting an impedance matching state using the second electrical characteristic in the impedance matching network; Calculating a first variation amount of the inductance or capacitance of the variable reactive element for impedance matching; And controlling the variable reactive element using the first change amount.

The impedance matching system according to an embodiment of the present invention provides high-speed impedance matching and fixes the driving frequency to a constant value to improve plasma process stability and plasma process reproducibility.

1 is a conceptual diagram illustrating an RF power system according to an embodiment of the present invention.
2 is a block diagram illustrating the RF power system of FIG. 1 in greater detail.
3 is a flow chart illustrating a method of controlling an RF power system according to an embodiment of the present invention.
4 is a flowchart illustrating an impedance matching method according to an embodiment of the present invention.
5 is a flowchart illustrating a frequency variable impedance matching method according to an embodiment of the present invention.
6 is a smith chart illustrating frequency variable impedance matching according to an embodiment of the present invention.
7 is a flowchart illustrating an impedance matching method according to another embodiment of the present invention.
8 is a flowchart illustrating an impedance matching method according to another embodiment of the present invention.
Figures 9 and 10 illustrate an RF power system in accordance with yet another embodiment of the present invention.
11 is a Smith chart showing an impedance matching area in a standard L-type impedance matching network.
FIG. 12 is a graph showing impedance matching conditions using only a variable frequency RF power source when impedance matching is possible using only a frequency variable RF power source.
FIGS. 13 to 17 are graphs showing simulation results showing the locus of impedance matching according to an embodiment of the present invention. FIG.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following drawings, like reference numerals refer to like elements, and the size of each element in the drawings may be exaggerated for clarity and convenience of explanation.

The impedance matching network according to an embodiment of the present invention may be disposed between a frequency variable RF power source and a load to induce impedance matching at a target driving frequency. Therefore, the frequency variable RF power source can change the driving frequency to perform high-speed impedance matching or reduce the reflected power to within a predetermined range. Even if the frequency variable RF power source performs impedance matching, if the driving frequency is different from the target driving frequency, the impedance matching network induces the frequency variable RF power source to perform impedance matching on the target driving frequency . That is, when the frequency variable RF power source satisfies the condition of impedance matching, and the driving frequency and the target driving frequency are different, the impedance matching network may continue to operate to intentionally deviate from the impedance matching condition. Accordingly, the frequency variable RF power source changes the driving frequency under the changed conditions to perform the impedance matching. As a result, when the driving frequency reaches the target driving frequency and impedance matching is performed, the reactance of the variable reactive element of the impedance matching network is fixed to a constant value. This process can be repeated with the load changed over time.

Meanwhile, when the frequency variable RF power source can not independently perform impedance matching, the impedance matching network may perform impedance matching by itself, and induce the frequency variable RF power source to perform impedance matching with the target driving frequency have.

Specifically, first, the frequency variable RF power source minimizes the reflected power. When the reflection coefficient in the direction of the load is large at the input terminal of the impedance matching network, the impedance matching network drives the variable reactive element to perform impedance matching. Accordingly, the frequency variable RF power source changes the driving frequency with respect to the changed impedance according to the variable active element to perform impedance matching.

In addition, when the reflection coefficient in the direction of the load at the input terminal of the impedance matching network is small (when almost impedance matching has been reached), the impedance matching network is controlled so as to be largely dependent on the function of the difference between the driving frequency and the target frequency. The device can be controlled. Therefore, the frequency variable RF power source can perform impedance matching at a high speed.

1 is a conceptual diagram illustrating an RF power system according to an embodiment of the present invention.

2 is a block diagram illustrating the RF power system of FIG. 1 in greater detail.

3 is a flow chart illustrating a method of controlling an RF power system according to an embodiment of the present invention.

4 is a flowchart illustrating an impedance matching method according to an embodiment of the present invention.

5 is a flowchart illustrating a frequency variable impedance matching method according to an embodiment of the present invention.

1 to 5, a frequency variable RF power source 110 transmits RF power to a load through an impedance matching network 130. The load 140 may typically be a dynamic load such as a plasma load. The frequency variable RF power supply 110 is connected to the impedance matching network 130 through a transmission line 120 having a characteristic impedance Z ₀ and the impedance matching network 130 is disposed adjacent to the load, And may transfer power to the load (140).

The frequency variable RF power supply 110 may include a driving frequency control loop and a power control loop. The frequency variable RF power source 110 may receive the set power and transmit the set power to the load 140 through the power control loop. In addition, the frequency variable RF power source 110 can calculate impedance or reflection coefficient by measuring electrical characteristics of the output terminals N1 and N2. The frequency variable RF power source 110 may change the driving frequency f so that the reflection coefficient becomes zero for impedance matching.

The frequency variable RF power source 110 measures the electrical characteristics at high speeds at the output terminals N1 and N2 to determine the impedance matching state and changes the driving frequency to perform impedance matching regardless of the impedance matching network 130 . The frequency variable RF power source 110 can calculate the impedance Z _T of the output terminals N1 and N2. The frequency variable RF power source may include a power source sensor unit 116, an amplification unit 114, and a control unit 112 for measuring electrical characteristics at output terminals N1 and N2. The measurement signal or the processed signal of the power source sensor unit 116 may be provided to the control unit. The controller 112 may perform a forward power control algorithm and a frequency variable impedance matching algorithm.

Also, the impedance matching network 130 can independently calculate the impedance Zin by measuring the electrical characteristics of the input terminals N3 and N4 of the impedance matching network. The impedance Zin of the input terminals N3 and N4 of the impedance matching network 130 and the impedance Z _T of the output terminals N1 and N2 of the frequency variable RF power source may have a predetermined relationship.

The impedance matching network 130 includes two variable active elements 132 and 134. The reactance of the two variable reactive elements 132 and 134 may be an inductive reactance or a capacitive reactance. Variable capacitors 132 and 134 are mainly variable capacitors. The illustrated impedance matching network may be changed to a standard L type or another type. For impedance matching, the impedance matching network may further include a fixed inductor or a fixed capacitor. Each of the variable reactive elements 132 and 134 can control a reactance through a switch by connecting a plurality of fixed capacitors in parallel.

The impedance matching network 130 may include an impedance matching control algorithm and a frequency restoration algorithm. The impedance matching control algorithm may change the reactance of the variable reactance elements 132 and 134 for impedance matching. In addition, the frequency reconstruction algorithm may change the reactance of the variable reactance elements 132 and 134 so that impedance matching is performed at a specific target drive frequency. The impedance matching network 130 may independently calculate the impedance Zin by measuring the electrical characteristics of the input terminals N3 and N4.

Hereinafter, the operation principle of the RF power system according to an embodiment of the present invention will be described.

The RF power system 100 includes a frequency tunable RF power source 110 and an impedance matching network 130 that transfers the output of the frequency tunable RF power source to the load 140. The impedance matching method of the RF power system includes a step S112 of measuring first electrical characteristics I and V of output terminals N1 and N2 of the variable frequency RF power supply 110, (S116) of checking the impedance matching state using the first electrical characteristic (I, V) (S116), changing a driving frequency (f) of the frequency variable RF power supply (S118) (S222) of measuring a second electrical characteristic (I'.V ') in the impedance matching network 130, checking the impedance matching state using the second electrical characteristic in the impedance matching network 130, determining whether the driving frequency is a target driving frequency The first variation amount dC '₁;dC' of the inductance or capacitance of the variable elements 132 and 134 for impedance matching when the impedance matching is not performed in the impedance matching network, ₂₎ for calculating Series (S232), the second amount of change (dC ''_1; dC 'of the impedance when the in the matching network driving frequency does not match the target drive frequency the inductance of the variable reactive element for driving frequencies change, or greater Passage turn' ₂ Calculating a total change amount (dC ₁ ; dC ₂ ) due to the first change amount and the second change amount, and controlling the variable reactive element using the total change amount (step S236 and S238.

The impedance matching network 130 is a standard L type and may include a first variable capacitor 132 and a second variable capacitor 134. The first variable capacitor 132 is connected in series to the load, and the second variable capacitor 132 can be connected in parallel to the load. The first variable capacitor and the second variable capacitor may be a vacuum variable capacitor. The first variable capacitor may include a first drive motor 132a and the second variable capacitor may include a second drive motor 134a. The first driving motor 132a and the second driving motor 134a may be connected to the motor driving unit 137. [ The impedance controller 138 controls the motor driver 137 via the drive vectors V1 and V2. Also, the impedance controller 138 can directly receive the second electrical characteristics I, V or receive the calculated electrical characteristics S11, Z. The matching sensor unit 136 may measure the driving frequency and provide the measured impedance to the impedance controller 138. The impedance controller 138 may execute at least one of an impedance matching control algorithm and a driving frequency restoration algorithm.

The impedance matching control algorithm in the impedance matching network 130 determines the variation amount dC1 dC2 of the variable capacitance C1 and C2 of the first variable capacitor 132 and the second variable capacitor 134 through the impedance analysis, Can be obtained as follows.

Here, dC ₁ is the variation of the capacitance (C ₁ ) of the first variable capacitor, and dC ₂ is the variation of the capacitance (C ₂ ) of the second variable capacitor. A and B are parameters determined for impedance matching at a predetermined driving frequency (f). A and B can be given as a function of the impedance or a reflection coefficient. A and B can be obtained by a conventional method.

The impedance matching network 130 performs impedance matching by varying the capacitance of the variable reactance elements 132 and 134. Assuming that the impedance matching network 130 performs impedance matching or the frequency variable RF power source changes the driving frequency to perform impedance matching, the amount of change dC ₁ , dC ₂ is zero, . However, the impedance matching network 130 receives power from the frequency variable RF power source 110. Therefore, when the impedance matching is performed or when the reflection coefficient is zero, the driving frequency f of the frequency variable RF power source is set to the impedance Z _L of the load and the conditions C ₁ and C ₂ of the impedance matching network You can depend on it.

The reactance components of the variable reactive elements 132 and 134 are expressed by the product of the frequency and the inductance or expressed as the product of the frequency and the capacitance. Therefore, it is assumed that the impedance matching is performed in a state where the load impedance Z _L is fixed. In this case, when the driving frequency is to be changed to the target driving frequency while maintaining the same reactance, the variation amount dω of the driving angular frequency, the variation amount dC of the capacitance or the variation amount dω of the driving angular frequency, The change amount (dL) satisfies the following conditions.

That is, when the driving angular frequency? Or the driving frequency f increases, the capacitance or inductance decreases. When the target driving frequency f _t is set and the driving frequency f is different from the target driving frequency f _t , the variation dω of the driving angular frequency depends on the variation dC of the capacitance.

If the difference between the target driving frequency f _t and the driving frequency f has a positive value, the capacitance or the change amount of the inductance of the variable reactive elements 132 and 134 can be controlled to have a negative value. If the difference between the target driving frequency f _t and the driving frequency f has a negative value, the capacitance or the change amount of the inductance of the variable reactive elements 132 and 134 can be controlled to have a positive value.

Therefore, when the impedance matching condition is satisfied, the variation amount (dC ₁ ; dC ₂ ) of the variable capacitance of the reactive element can be given as follows in order to match the driving frequency to the target driving frequency.

C ₁ is the capacitance of the first capacitor in its current state, and C ₂ is the capacitance of the second capacitor in its present state. d? is a change amount of each frequency, and when the value of d? is large, the change amount dC _{1 of} the first capacitance and the change amount dC ₂ of the first capacitance change greatly. The ratio (dC ₁ / dC ₂ ) of the variation amount (dC ₁ ) of the _first capacitance to the variation amount (dC ₂ ) of the _second capacitance has a constant directivity (C ₁ / C ₂ ). The change amount dC ₁ of the _first capacitance and the change amount dC ₂ of the second capacitance are changed so that the ratio dC ₁ / dC ₂ is maintained.

Specifically, dω = ω _t -ω (t = 0) can be given. Also, dC ₁ = C ₁ (t = 0 ⁺ ) -C ₁ (t = 0) can be given. ? _t is the target drive angular frequency, and? (t = 0) is the current drive angular frequency. C ₁ (t = 0) is the current capacitance and C ₁ (t = 0 ⁺ ) is the future capacitance.

When the first capacitance (C ₁ ) and the second capacitance (C ₂ ) are changed using Equation (3), the impedance matching network and the variable frequency RF power supply may not satisfy the impedance matching condition have. Accordingly, the frequency variable RF power source can change the driving frequency to satisfy the impedance matching condition.

Also, if the changed drive frequency is different from the target drive frequency, the change amount dC _{1 of} the first capacitance and the change amount dC ₂ of the first capacitance change again through the equation (3). As a result, until the driving frequency reaches the target driving frequency, the first and second capacities of the impedance matching network are continuously changed.

Accordingly, the impedance matching network 130 can guide the driving frequency of the frequency variable RF power source to reach the target driving frequency irrespective of the driving frequency changing algorithm of the frequency variable RF power source 110. When the frequency variable RF power source 110 can perform the impedance matching by changing the driving frequency, the impedance matching network 130 can operate only the frequency recovery algorithm without operating the impedance matching algorithm.

On the other hand, assuming that the current impedance matching is achieved, the frequency variable RF generator 110 can satisfy the impedance matching condition at the predicted driving angular frequency? _P. The predicted driving angular frequency can be given as follows.

The frequency variable RF power source 110 may receive the predicted driving angular frequency and change the driving frequency to be changed at a high speed to the predicted driving angular frequency. However, in the impedance matching network, the variation of the first capacitance of the first variable capacitor and the variation of the second capacitance of the second variable capacitor may have a time delay due to the driving speed of the motor. In this case, it is preferable that the driving speed of the motor is changed to the maximum speed in this case. However, in order to adjust the speed, the predicted drive angular frequency may be set different from the target drive frequency.

According to a modified embodiment of the present invention, since dC ₁ and dC ₂ have the same sign, the predicted driving angular frequency may depend only on the sign of dC ₁ or dC ₂ .

Under the condition that the impedance matching is achieved, the predicted driving frequency fp calculated above can be provided to the frequency variable RF generator 110. [ Accordingly, the frequency variable RF generator can reach the target driving frequency without going through a separate impedance matching algorithm. Thus, while meeting the impedance matching condition, the driving frequency can easily reach the target driving frequency quickly. Thus, RF power can be delivered to the load at the target driving frequency while satisfying the impedance matching condition.

According to a modified embodiment of the present invention, even if the impedance matching is not achieved, the calculated predicted driving frequency may be provided to the frequency variable RF generator 110.

By changing the driving frequency, the impedance region of the load capable of achieving impedance matching is limited. Accordingly, when the impedance matching can not be achieved by changing the driving frequency, the impedance matching network 130 may include an algorithm for changing the variable reactive elements 132 and 134 to find the impedance matching condition. Specifically, the conditions of the plasma load can vary greatly with time. For example, low power may be required in the first time period and high power may be required in the second time period. In this case, the impedance matching can be performed by changing the driving frequency in the first time interval, but the impedance matching can not be achieved by changing the driving frequency in the second time interval. Therefore, there is a demand for a variable reactive element capable of performing impedance matching under a wide range of load conditions.

The impedance matching network may have an algorithm for finding impedance matching conditions and an algorithm for driving the driving frequency to reach the target driving frequency. Algorithm for deriving so as to reach an impedance matching algorithm to find the condition and the driving frequency to a target drive frequency is represented by the amount of change (dC ₁₎ and the second amount of change (dC ₂₎ of the variable increases Passage turn of the first variable capacitance increases Passage.

A and B are arbitrary values. A and B do not satisfy the impedance matching condition, the variation amount (dC ₁ ) of the first capacitance and the variation amount (dC ₂ ) of the _second capacitance are set so as to satisfy the impedance matching The first and second terms can compete. The first term is a term established for impedance matching and the second term is a term for deriving the driving frequency to the target driving frequency. If both the first and second terms operate, a haunting issue may occur.

Therefore, for efficient operation, Equation (4) can be modified as follows.

Here, g ₁ is a first weighting function, and g ₂ is a second weighting function. The first weighting function g ₁ has a large value when the reflection coefficient is large, and a small value when the reflection coefficient is small. On the other hand, the second weighting function g ₂ has a small value when the reflection coefficient is large and a large value when the reflection coefficient is small.

Referring to Equation (6), if the reflection coefficient is largely deviated from the impedance matching condition, the first term of the right side mainly operates and impedance matching is performed. The first variation of the variable capacitance increases Passage (dC ₁₎ and the second variable greater amount of change in the turn Passage (dC ₂₎ can be changed while maintaining the direction (A / B). Specifically, the first variable capacitance and the second capacitance can be changed while the ratio of A and B is kept constant.

Referring to Equation (6), when the impedance matching condition is almost reached and the reflection coefficient is small, the second term of the right side is mainly operated, and the impedance matching network determines that the driving frequency of the frequency variable RF power supply 110 is . The first variation of the variable capacitance increases Passage (dC ₁₎ and the second variable greater amount of change in the turn Passage (dC ₂₎ can be changed while maintaining the direction (C1 / C2).

According to an embodiment of the present invention, when the frequency variable RF power source 110 changes the driving frequency to perform impedance matching, the impedance matching network 130 operates mainly the driving frequency changing algorithm. Accordingly, the frequency variable RF power source 110 performs impedance matching at a high speed, and the impedance matching network 130 maintains the impedance matching, while the frequency variable RF power source 110 changes the driving frequency to the target driving frequency . Therefore, when the load is a plasma load, the stability of the hardware can be secured by quick impedance matching. Further, the driving frequency converges to the target driving frequency, and the process reproducibility can be ensured. Particularly, in a pulse plasma process or a plasma process having a multi-stage recipe, process stability and process reproducibility are secured.

The impedance matching method according to an embodiment of the present invention can be applied to a continuous wave (CW) plasma or a pulse plasma.

According to an embodiment of the present invention, when impedance matching can not be performed using only the frequency variable RF power source 110, the impedance matching network 130 mainly operates an impedance matching algorithm. The impedance matching network 130 changes the reactance of the variable reactive element to perform impedance matching. Accordingly, when impedance matching is performed, the impedance matching network 130 mainly operates a driving frequency change algorithm. Accordingly, the impedance matching network 130 can induce the driving frequency of the variable frequency RF power supply to be changed to the target driving frequency while satisfying the impedance matching condition.

According to a modified embodiment of the present invention, the frequency variable RF power source may provide a driving frequency to the impedance matching network. In addition, the frequency variable RF power source may transmit the measured electrical characteristics to the impedance matching network. Accordingly, the impedance matching network can eliminate the separate driving frequency measuring step and the electrical characteristic measuring step.

[Frequency variable impedance matching]

Hereinafter, a frequency variable impedance matching method according to an embodiment of the present invention will be described.

The frequency variable RF power source 110 can calculate the output impedance Z _T , the reflection coefficient S11, or the reflected wave power by measuring the electrical characteristic at the output terminal N1.N2. The frequency variable RF power source 110 may vary the driving frequency in the range of the minimum frequency and the maximum frequency. Typically, the variable range may be 5 percent of the reference frequency or center frequency. Coarse frequency scans can be performed at regular intervals to find the drive frequency with the lowest reflection coefficient in the range of the minimum frequency and the maximum frequency. Thus, a section with the lowest reflection coefficient can be found. A fine frequency scan may again be performed in the detected section. Accordingly, the driving frequency with the lowest reflection coefficient can be selected.

6 is a smith chart illustrating frequency variable impedance matching according to an embodiment of the present invention.

Referring to FIGS. 5 and 6, if the load is fixed and the drive frequency is changed, the reflection coefficient on the Smith chart may follow a constant conductance circle. Therefore, if the imaginary part of the reflection coefficient has a positive value, the driving frequency is increased. On the other hand, if the imaginary part of the reflection coefficient has a negative value, the driving frequency is decreased. Accordingly, the driving frequency can be stopped at a point where the reflection coefficient is minimum or at a point where the reflection coefficient is zero. The frequency variable RF power source can perform frequency variable impedance matching.

The variable frequency RF power source can perform impedance matching by measuring the electrical characteristics at the output of the frequency variable RF power source and changing the driving frequency using the measured electrical characteristics (S118). First, the electrical characteristics of the output terminal can be measured by the frequency variable RF power source (S112). Then, using the driving frequency and the measured electrical characteristic can be calculated reflection coefficient (S11) or impedance (Z _T) (S114). The matching state is checked using a reflection coefficient or a voltage standing wave ratio (VSWR) (S118a). If the reflection coefficient is more than the allowable value, the driving frequency may be changed for impedance matching (S118c). Further, when the driving frequency reaches the maximum value or the minimum value, the driving frequency can be maintained at the maximum value or the minimum value (S118b). As the drive frequency is changed, the impedance (Z _T ) in the direction immediately before the load can be changed.

Hereinafter, a variable reactance impedance matching method according to an embodiment of the present invention will be described.

According to the present invention, the variable reactive element may be one of a variable capacitor providing a variable capacitance or a variable inductor providing a variable inductance.

[Various types of matching systems]

The impedance matching network can be classified into various types according to the manner in which the variable reactive element or the passive elements are connected to the transmission line. If the impedance matching network has first and second variable capacitors, the L-type, the inverted L-type, and the second variable capacitors may be connected to the transmission line 120 according to a manner in which the first and second variable capacitors are connected to the transmission line 120. [ Type, an inverted L-type, a T-type, and a pi-type.

The impedance matching of the impedance matching network includes a step of extracting a characteristic vector (S232a), an element vector representing a reactance of the variable reactive element is converted into an analysis vector using a predetermined conversion matrix, and an analysis coordinate system (S232b) of extracting a displacement vector for impedance matching by analyzing the characteristic vector in the analytical coordinate system (S232c), converting the displacement vector into a conversion element vector using the conversion matrix , And extracting a change amount of capacitance or inductance using the converted element vector (step 232e).

[Selection of characteristic vector]

 According to an embodiment of the present invention, the characteristic vector may be a physical quantity having a standardized size, which is defined based on an electrical characteristic measured at an input terminal of the impedance matching network or at an output terminal of the variable frequency RF power source.

According to an embodiment of the present invention, the characteristic vector may be defined from a reflection coefficient S11 of the transmission line. The reflection coefficient S11 =? Of the transmission line is defined by the characteristic impedance Z0 of the transmission line and the impedance Z in the direction of the load.

The magnitude of the reflection coefficient (i.e., S = | S11 |) may be a value between 0 and 1. The impedance Z of the transmission line represents the impedance of the system including the impedance matching network and the load. The phase of the reflection coefficient varies depending on the position of the transmission line. Therefore, the reflection coefficient can be mutually converted according to the position of the transmission line.

The impedance matching network may include at least two variable reactive elements. In this case, in order to explicitly determine the reactance of each variable reactive element, the characteristic vector must be a physical quantity including at least two components. For example, the characteristic vector Q may be defined as a two-dimensional vector having a real part (Re {S11}) and an imaginary part (Im {S11}) of the reflection coefficient as follows.

[Selection of analytical coordinate system]

The analytical coordinate system is selected to represent a predetermined phase space quantitatively related to the electrical characteristics (C1, C2) of the impedance matching network and the electrical characteristics (Z, S11) of the transmission line. To this end, the coordinate of the analytical coordinate system is selected from physical quantities related to the electrical characteristics of the matching impedance matching network, and the electrical characteristic of the transmission line is represented by a point in the analytical coordinate system thus selected.

According to an embodiment of the present invention, the coordinates of the analytical coordinate system (hereinafter referred to as analytical coordinates) may be expressed as a function of the electrical characteristics (for example, reactance) of the variable reactive elements constituting the impedance matching network, The characteristic vector representing the measured electrical characteristic of the transmission line can be represented by a point on the analytical coordinate system.

The analysis coordinate system is preferably selected to injectively map a quantitative relationship between the electrical characteristics of the matching system and the electrical characteristics of the transmission line. The term "injective mapping" refers to a relationship in which one matching point corresponds to one analytical coordinate.

The analytical coordinates may be physical quantities obtained by transforming measurable electrical characteristics (e.g., reactance) of the impedance matching network using a predetermined transformation matrix T. [ For example, the analytical coordinates G may be obtained through an inner product of a predetermined transformation matrix T and a predetermined device vector X as expressed as follows.

The element vector X may be selected according to the type of the matching system while having as a component the physical quantities related to the electrical characteristics of the variable reactive elements constituting the impedance matching network. In addition, the transformation matrix T may be selected according to the type of the impedance matching network and the physical quantity of the element vector (X). As a result, the analysis coordinates G are also selected according to the type of the matching system.

More specifically, the impedance matching network may comprise two variable reactive elements. In this case, the analytical coordinates G1 and G2 can be obtained through the dot product of the physical quantities X1 and X2 related to the electrical characteristics of each of the variable active elements and the predetermined quadratic matrix T . According to the invention, the elements of the transformation matrix T (i.e., a11, a12, a21, a22) may be selected from values between -1 and 1.

Meanwhile, the transformation matrix T may be prepared through various methods. For example, the transformation matrix may comprise at least one of an analysis through a theoretical approach and a simulation analysis using a computer, an impedance measurement of the impedance matching network according to the variable reactive element value, and an analysis of empirical data on the matching process Can be obtained through analytical methods. This analysis is made based on the type of the matching system and the physical quantity of the element vector (X). In addition, the shape and rank of the transformation matrix T are determined by the number of variable reactive elements constituting the matching system. That is, when the matching system includes a greater number of variable reactive elements, the shape and order of the transformation matrix may increase.

As described above, the impedance matching network may comprise two variable capacitors. In this case, the transformation matrix T and the element vector X can be expressed as a function of the capacitance of the variable capacitors.

In the case of the L-type or pie-type impedance matching network, the element vector X can be given as follows.

Where? Is the angular frequency and Ci is the capacitance of each variable capacitor.

In the case of an inverse-L or T type matching network, the element vector X can be given as follows.

[Determination of Displacement Vector]

The displacement vector expresses the magnitude or position of the characteristic vector corresponding to the measured state of the transmission line in the analytical coordinate system (hereinafter, referred to as the measured characteristic vector) and expresses the magnitude of the coordinate movement required to move the impedance matching line to the impedance matching line .

In the case of the L-type or pie-type impedance matching network, the displacement vector dG1 (dG2) can be selected as follows.

In the case of an inverse-L or T type matching network, the displacement vector dG1 (dG2) can be selected as follows.

According to an embodiment of the present invention, the displacement vector dG is a physical quantity obtained by converting reflection coefficient or impedance for impedance matching analysis in the analytical coordinate system. Therefore, in order to control the impedance matching network, it is necessary to convert the displacement vector into a magnitude of an electrical characteristic of the elements constituting the impedance matching network (that is, a reactance of the variable active element) or a physical quantity related thereto .

To this end, a process of inverse transformation of the displacement vector dG into a reduced device vector dX 'having a variable physical quantity dimension of the variable active device is required. Further, a process of converting the converted element vector dX 'into a drive vector V for controlling the driving of the variable reactive elements is required.

Considering that the analytical coordinates G1 and G2 are obtained through the transformation matrix T, in the case of the impedance matching network (variable capacitor) of the L type or the pie type, the transformed element vector dX ' Can be obtained through an inner product of the inverse matrix (T ^-1 ) of the transformation matrix and the displacement vector (dG).

(T- ¹ ) of the transformation matrix and the displacement vector dG in the case of an inverse L-type or T-type matching network ((variable capacitor) Can be obtained.

According to an embodiment of the present invention, the reactance of the variable reactive element can be changed by controlling a switching element that switches a predetermined driving motor, a linear motion of the linear motion means, or a fixed capacitor connected in parallel.

For example, when the variable reactive element is a variable capacitor, the amount of change (dC ₁ ; dC ₂ ) of the capacitance of the impedance matching network of L-type or P-type can be expressed as follows.

For example, when the variable reactive element is a variable capacitor, the amount of change (dC ₁ ; dC ₂ ) of the capacitance of the impedance matching network of L-type or P-type can be expressed as follows.

[Combination of Frequency Restoration Algorithm and Variable Reactance Impedance Matching Algorithm]

Considering the frequency variable impedance matching, when the variable reactive element is a variable capacitor, the amount of change (dC ₁ ; dC ₂ ) of the capacitance of the impedance matching network of L-type or P-type can be expressed as follows.

In the above equation, the first term of the right side is the variation of the capacitance for reactance impedance matching, and the second term of the right side is the variation of the capacitance to induce the driving frequency to the target driving frequency. g ₁ is a first weighting function, and g ₂ is a second weighting function. The first weighting function g1 has a large value when the reflection coefficient is large and a small value when the reflection coefficient is small. On the other hand, the second weighting function g2 has a small value when the reflection coefficient is large, and a large value when the reflection coefficient is small.

If the reflection coefficient is large, the first term of the right side is mainly operated, and the impedance matching is performed with a slow speed. On the other hand, when the reflection coefficient is small, the second term of the right side mainly operates and the driving frequency can be guided to the target driving frequency at a relatively high speed.

According to a modified embodiment of the present invention, in the first term of the right side, in order to increase the calculation speed, the driving angular frequency, C ₁ , and C ₂ can be treated as constants. In the second term of the right side, in order to increase the calculation speed, the driving angular frequency, C ₁ , and C ₂ , can be treated as a constant.

Considering the frequency variable impedance matching, when the variable reactive element is a variable capacitor, the amount of change (dC ₁ ; dC ₂ ) of the capacitance of the inverted L or T type impedance matching network can be expressed as follows.

In the above equation, the first term of the right side is the variation of the capacitance for reactance impedance matching, and the second term of the right side is the variation of the capacitance to induce the driving frequency to the target driving frequency. g ₁ is a first weighting function, and g ₂ is a second weighting function. The first weighting function g1 has a large value when the reflection coefficient is large and a small value when the reflection coefficient is small. On the other hand, the second weighting function g2 has a small value when the reflection coefficient is large, and a large value when the reflection coefficient is small.

If the reflection coefficient is large, the first term of the right side is mainly operated, and the impedance matching is performed with a slow speed. On the other hand, when the reflection coefficient is small, the second term of the right side mainly operates and the driving frequency can be guided to the target driving frequency at a relatively high speed.

According to a modified embodiment of the present invention, in the first term of the right side, in order to increase the calculation speed, the driving angular frequency can be treated as a constant. In the second term of the right side, in order to increase the calculation speed, the driving angular frequency, C ₁ , and C ₂ , can be treated as a constant.

[Driving vector]

The drive vector V may have values for the numerical control of the drive motor as its components and the magnitude and the physical dimension of the drive motor may be determined by the method of numerical control and the type of drive motor, And can be variously modified accordingly. For example, the drive vector V may be given as a scalar product of a change amount (dC ₁ ; dC ₂ ) of the capacitance and a predetermined numerical control factor (M).

V1 and V2 denote control parameters input for driving the driving motors connected to the first and second variable capacitors, respectively. Also, the numerical control factor M may be a reference magnitude of the numerical control (e.g., a standard speed of the motor), and the change amount dC ₁ ; dC ₂ ) Is selected to have the same dimension as the drive vector (V).

According to a modified embodiment of the present invention, V1 and V2 may be control parameters for the linear motion of the linear motion means or control parameters of the switching elements for switching parallel connected fixed capacitors.

7 is a flowchart illustrating an impedance matching method according to another embodiment of the present invention.

Referring to Fig. 7, the variable reactance impedance matching network is disposed between the frequency variable RF power source for changing the driving frequency f and the frequency variable RF power source for changing the driving frequency f disposed between the load and the load. The impedance matching method of the variable reactance impedance matching network changes the capacitance or inductance of the variable reactive element of the impedance matching network to induce the variable frequency RF power source to operate at the target driving frequency at step S200. The amount of change in capacitance or inductance of the variable reactive element of the impedance matching network is controlled as a function of the difference between the target drive frequency f _t and the drive frequency f.

The impedance matching method includes: S110 performing a first impedance matching by measuring an electrical characteristic at an output of the frequency variable RF power source and changing a driving frequency using the measured electrical characteristic; And an impedance matching network including a variable reactive element between the frequency variable RF power source and the load to change the capacitance or inductance of the variable reactive element (S200). The change amount of the capacitance or inductance of the variable reactive element is given as a function of the difference between the target drive frequency f _t and the drive frequency f.

The step S110 of performing the first impedance matching may include the step S112 of measuring a first electrical characteristic of the output terminal of the frequency variable RF power source; (S114, S116) checking the impedance matching state using the first electrical characteristic in the frequency variable RF power source; And changing a driving frequency of the frequency variable RF power source (S118).

The step (S200) of changing the capacitance or inductance of the variable reactive element includes the steps of measuring an electrical characteristic of the impedance matching network (S222); Checking whether the driving frequency is a target driving frequency (S228); Calculating a change amount of the inductance or the capacitance of the variable reactive element for changing the driving frequency when the driving frequency does not match the target driving frequency (S234); And controlling the variable reactive element by using the change amount (S236, S238).

8 is a flowchart illustrating an impedance matching method according to another embodiment of the present invention.

Referring to FIG. 8, the impedance matching network and the frequency variable RF power source operate in an integrated manner. Accordingly, only the electrical characteristic of the output terminal of the frequency variable RF power source can be measured (S322). Accordingly, the impedance Zin or the reflection coefficient S11 in the direction in which the load is viewed from the variable reactive element can be calculated through calculation (S324). If the impedance matching condition is satisfied, it is checked whether the driving frequency is the target driving frequency (S326, S328). If the impedance matching condition is not satisfied, the driving frequency is varied (S329, S330). If the impedance matching condition is not satisfied, the inductance of the variable reactive element for impedance matching or the first variation amount of the variable capacitance turn is extracted (S332). If the drive frequency is not the target drive frequency, the second change amount of the variable inductance or the variable capacitance for the drive frequency change is extracted (S334). Subsequently, the first change amount is multiplied by the first weighting function, the second change amount is multiplied by the second weighting function, and the drive vector is calculated (S336). The drive vector drives a variable inductance or a variable capacitance (S338).

9 illustrates an RF power system in accordance with another embodiment of the present invention.

Referring to FIG. 9, the variable reactive element may include a fixed capacitor connected in parallel. The fixed capacitor may include a switch to change the capacitance. Thus, since the variable capacitance does not use the driving motor, the impedance matching speed and the target driving frequency restoration speed can be increased.

10 illustrates an RF power system in accordance with another embodiment of the present invention.

Referring to FIG. 10, the variable reactive element may include a fixed capacitor connected in parallel and a variable capacitor. The fixed capacitor may include a switch to change the capacitance. Accordingly, the variable capacitance moves at a high speed near the target value by using the fixed capacitor and the switch, and the variable capacitor using the driving motor can perform fine impedance matching or frequency restoration.

11 is a Smith chart showing an impedance matching area in a standard L-type impedance matching network.

Referring to FIGS. 1 and 11, the first fixed inductor L1 is 100 nH, the second fixed inductor L2 is 500 nH, the variable range of the first variable capacitor 132 is 150 pF to 1000 pF, , And the variable range of the second variable capacitor 134 is 85 pF to 500 pF. An area capable of impedance matching through the first variable capacitor and the second variable capacitor is displayed.

FIG. 12 is a graph showing impedance matching conditions using only a variable frequency RF power source when impedance matching is possible using only a frequency variable RF power source.

Referring to FIG. 12, the conditions used in the simulation are as follows. The first fixed inductor is 100 nH, the second fixed inductor is 500 nH, the variable range of the first variable capacitor is 150 pF to 1000 pF, and the variable range of the second variable capacitor is 85 pF to 500 pF. For frequency variable impedance matching, the first variable capacitor was fixed at 724.5 pF and the second variable capacitor was fixed at 150 pF. The time interval of the control loop is 100 microseconds (μsec). The frequency tunable algorithm increases the driving frequency when the frequency variable RF power source has a positive value of the imaginary part of the reflection coefficient and decreases the driving frequency when the imaginary part of the reflection coefficient has a negative value.

According to the simulation results, frequency variable impedance matching was performed at about 2 msec. In this case, the y-axis of the driving frequency is given by 50 + (f-fo) / (2? F). Therefore, 2? F is a frequency variable region.

The drive frequency increased from the initial value (50 percent) to 70 percent after about 2 msec. The algorithm for changing the driving frequency can perform the impedance matching at a speed as high as about 2 msec. However, the driving frequency increased from 50 percent to 70 percent. In a situation where the impedance of the plasma load changes with time, the driving frequency can not be kept constant, and plasma process stability and reproducibility can be deteriorated.

On the other hand, if the time interval of the control loop is 100 microseconds (μsec) or less, the algorithm for changing the driving frequency can perform the impedance matching in less than 2 msec.

13 is a graph showing impedance matching conditions using only a variable reactive element at a fixed driving frequency when impedance matching can be performed only by a frequency variable RF power source.

Referring to FIG. 13, the conditions used in the simulation are as follows. The first fixed inductor is 100 nH, the second fixed inductor is 500 nH, the variable range of the first variable capacitor is 150 pF to 1000 pF, and the variable range of the second variable capacitor is 85 pF to 500 pF. The initial condition of the first variable capacitor is 724.5 pF and the initial condition of the second variable capacitor is 150 pF. Equation 16 is used for impedance matching.

For matching of the variable reactive elements, the driving frequency was fixed at 50 percent. The time interval of the control loop is 100 microseconds (μsec).

Due to the time required to drive the motor of the variable capacitor, the reactance change algorithm of the variable reactive element requires about 200 msec (milliseconds) for impedance matching. Therefore, the time required for the reflected power to decrease to 50 percent is required to be approximately 40 msec.

In the situation where the impedance of the plasma load varies with time, the time required for the reflected power to decrease to 50 percent is about 40 msec. Therefore, plasma process stability and reproducibility can be deteriorated.

14 is a graph showing conditions of impedance matching employing an algorithm for restoring a driving frequency to a target driving frequency by simultaneously using a frequency variable RF power source and a variable reactive element when impedance matching can be performed only by a frequency variable RF power source.

Referring to FIG. 14, the conditions used in the simulation are as follows. The first fixed inductor is 100 nH, the second fixed inductor is 500 nH, the variable range of the first variable capacitor is 150 pF to 1000 pF, and the variable range of the second variable capacitor is 85 pF to 500 pF. The initial condition of the first variable capacitor is 724.5 pF and the initial condition of the second variable capacitor is 150 pF. The initial condition of the driving frequency is 50 percent. The time interval of the control loop is 100 microseconds (μsec). The frequency tunable algorithm increases the driving frequency when the frequency variable RF power source has a positive value of the imaginary part of the reflection coefficient and decreases the driving frequency when the imaginary part of the reflection coefficient has a negative value. Equation 18 is used for impedance matching and frequency recovery.

Impedance matching was achieved at about 2 msec by changing the driving frequency. Then, the driving frequency converges to the target driving frequency (50 percent) at about 100 msec while satisfying the impedance matching condition. Therefore, the impedance matching time is as fast as about 2 msec, and the time for the driving frequency to reach the target driving frequency is about 100 msec, which is shorter than the impedance matching time using the variable active element.

The time the reflected power drops below 50 percent is about 0.6 msec. Thus, the process instability due to the reflected power is improved. That is, a quick impedance matching is performed within about 2 msec, and the driving frequency restoration is performed within about 100 msec.

In the case of a multi-step plasma process, each step can typically have a plasma state of a few seconds or less. Therefore, if impedance matching is performed within 2 msec and frequency recovery is performed within 100 msec, the process stability can be significantly improved.

In the case of a plasma process using a pulsed plasma, the pulse width is typically within a few msec. Thus, according to one embodiment of the present invention, impedance matching is performed within the width of the pulse. Then, after passing through a plurality of pulse sequences, the driving frequency restoration can be performed. Therefore, process stability and process reproducibility are improved.

15 is a graph showing impedance matching conditions using only a variable frequency RF power source when impedance matching is impossible using only a frequency variable RF power source.

Referring to FIG. 15, the conditions used in the simulation are as follows. The first fixed inductor is 100 nH, the second fixed inductor is 500 nH, the variable range of the first variable capacitor is 150 pF to 1000 pF, and the variable range of the second variable capacitor is 85 pF to 500 pF. For frequency variable impedance matching, the first variable capacitor was fixed at 500 pF and the second variable capacitor was fixed at 150 pF. The time interval of the control loop is 100 microseconds (μsec). The frequency tunable algorithm increases the driving frequency when the frequency variable RF power source has a positive value of the imaginary part of the reflection coefficient and decreases the driving frequency when the imaginary part of the reflection coefficient has a negative value.

According to the simulation results, the impedance matching can not be performed within the frequency variable range. Therefore, when the range of the plasma load is wide, the frequency variable RF power source can not be used.

16 is a graph showing impedance matching conditions using only a variable active element at a fixed driving frequency when impedance matching is impossible using only a frequency variable RF power source.

 Referring to FIG. 16, the conditions used in the simulation are as follows. The first fixed inductor is 100 nH, the second fixed inductor is 500 nH, the variable range of the first variable capacitor is 150 pF to 1000 pF, and the variable range of the second variable capacitor is 85 pF to 500 pF. The initial condition of the first variable capacitor is 500 pF and the initial condition of the second variable capacitor is 150 pF. Equation 16 is used for impedance matching.

The time to achieve 50% or less of reflected power is about 100 mec. Also, about 400 msec is required for impedance matching. Thus, in the case of a multi-step plasma process, each step can typically have a plasma state of a few seconds or less. Therefore, it is difficult to secure process stability and process reproducibility.

17 is a graph showing conditions of impedance matching employing an algorithm for restoring a driving frequency to a target driving frequency by simultaneously using a frequency variable RF power supply and a variable reactive element when impedance matching is impossible using only a frequency variable RF power source.

Referring to FIG. 17, the conditions used in the simulation are as follows. The first fixed inductor is 100 nH, the second fixed inductor is 500 nH, the variable range of the first variable capacitor is 150 pF to 1000 pF, and the variable range of the second variable capacitor is 85 pF to 500 pF. The initial condition of the first variable capacitor is 500 pF and the initial condition of the second variable capacitor is 150 pF. The initial condition of the driving frequency is 50 percent. The time interval of the control loop is 100 microseconds (μsec). The frequency tunable algorithm increases the driving frequency when the frequency variable RF power source has a positive value of the imaginary part of the reflection coefficient and decreases the driving frequency when the imaginary part of the reflection coefficient has a negative value. Equation 18 is used for impedance matching and frequency recovery.

Approximately 300 msec is required for impedance matching. However, the time to reduce the reflected power to less than 50 percent is within 3 msec through the frequency tunable algorithm. Thus, in a state in which the reflected power is minimized using the frequency variable algorithm, the impedance matching is performed using the reactive element. At the same time, the driving frequency converges to the target driving frequency at the same time.

Therefore, even when the range of the plasma load is very wide, impedance matching and driving frequency restoration can be performed while maintaining the reflected power within a predetermined range. Therefore, impedance matching is performed at a faster rate than that of performing impedance matching only with a reactive element, and the time for achieving less than 50 percent of reflected power can be significantly reduced.

While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims. Accordingly, the true scope of the present invention should be determined by the technical idea of the appended claims.

110: Frequency variable RF power source
120: transmission line
130: Impedance Matching Network
140: Load

Claims (4)

A variable frequency RF power source having a predetermined frequency variable range and changing a driving frequency for impedance matching; And
And an impedance matching network that receives an output from the frequency variable RF power source and transmits the output to a load,
Wherein the impedance matching network changes the capacitance or inductance of the variable reactive element of the impedance matching network to guide the frequency variable RF power source to operate at the target driving frequency,
Wherein the impedance matching network further comprises a matching sensor unit disposed between the impedance matching network and the frequency variable RF power source for measuring an electrical characteristic in the load direction in the impedance matching network,
Wherein the matching sensor unit calculates an impedance or reflection coefficient using the electrical characteristic,
And the matching sensor unit measures a driving frequency transmitted to the load. A variable frequency RF power source having a predetermined frequency variable range and changing a driving frequency for impedance matching; And
And an impedance matching network that receives an output from the frequency variable RF power source and transmits the output to a load,
Wherein the impedance matching network changes the capacitance or inductance of the variable reactive element of the impedance matching network to guide the frequency variable RF power source to operate at the target driving frequency,
Wherein the frequency variable RF power source further comprises a power source sensor unit disposed between the impedance matching network and the frequency variable RF power source for measuring an electrical characteristic in the load direction in the impedance matching network,
Wherein the power source sensor unit calculates an impedance or a reflection coefficient using the electrical characteristic,
Wherein the impedance matching network is provided with a driving frequency from the frequency variable RF power source. A variable frequency RF power source having a predetermined frequency variable range and changing a driving frequency for impedance matching; And
And an impedance matching network that receives an output from the frequency variable RF power source and transmits the output to a load,
Wherein the impedance matching network changes the capacitance or inductance of the variable reactive element of the impedance matching network to guide the frequency variable RF power source to operate at the target driving frequency,
Wherein the impedance matching network is provided with the impedance or reflection coefficient in the direction from the output of the frequency variable RF power source to the load from the frequency variable RF power source. A variable frequency RF power source having a predetermined frequency variable range and changing a driving frequency for impedance matching; And
And an impedance matching network that receives an output from the frequency variable RF power source and transmits the output to a load,
Wherein the impedance matching network changes the capacitance or inductance of the variable reactive element of the impedance matching network to guide the frequency variable RF power source to operate at the target driving frequency,
Wherein the impedance matching network provides a predicted driving frequency for impedance matching to the frequency variable RF power source.

Images