(12) United States Patent
`Howald
`
`USOO6259334B1
`(10) Patent No.:
`US 6,259,334 B1
`(45) Date of Patent:
`Jul. 10, 2001
`
`(54) METHODS FOR CONTROLLING AN RF
`MATCHING NETWORK
`NG N
`(75) Inventor: Arthur M. Howald, Pleasanton, CA
`(US)
`
`(73) Assignee: Lam Research Corporation, Fremont,
`CA (US)
`Subject to any disclaimer, the term of this
`patent is extended or adjusted under 35
`U.S.C. 154(b) by 0 days.
`
`(*) Notice:
`
`(21) Appl. No.: 09/218,542
`1-1.
`(22) Filed:
`Dec. 22, 1998
`(51) Int. Cl. ................................. H01P5/08; HO3H 7/38
`(52) U.S. Cl. ....................... 333173.333/32. 333/99 PL,
`315/111.21
`t
`(58) Field of Search ........................ 333/17.3, is 99 PL;
`315/111.21
`
`(56)
`
`References Cited
`
`U.S. PATENT DOCUMENTS
`
`3/1976 Schatz ............................... 219/10.49
`3,941,966
`2/1993 Collins et al.
`... 333/17.3
`5,187.454
`10/1995 Mundt ............
`... 361/234
`5,463,526
`5,589,844 * 12/1996 Belcher et al.
`... 343/860
`5,609.720
`3/1997 Lenz et al. .....
`... 156/643.1
`5,670,066
`9/1997 Barnes et al. ..
`... 219/121.58
`5,671,116
`9/1997 Husain ................................. 361/234
`5,689,215
`11/1997 Richardson et al. ................ 333/17.3
`5,708.250
`1/1998 Benjamin et al. ....
`... 219/121.58
`5,737,175
`4/1998 Grosshart et al. ................... 361/234
`5,793,162
`8/1998 Barnes et al. ................... 315/111.21
`5,793.192
`8/1998 Kubly et al. ...
`... 323/312
`5,798.904
`8/1998 Guyot ................................... 361/234
`
`500 y
`
`
`
`OTHER PUBLICATIONS
`M. Lieberman and A. Lichtenberg, “Principles of Plasma
`Discharges and Materials Processing", (C) 1994, Wiley-Inter
`science Publ., John Wiley & Sons, Inc.
`* cited b
`cited by examiner
`Primary Examiner Robert Pascal
`Assistant Examiner Kimberly E Glenn
`(74) Attorney, Agent, or Firm Martine & Penilla, LLP
`(57)
`ABSTRACT
`Disclosed are methods and devices for tuning an impedance
`matching network to a tune point where power reflection is
`at a minimum. The impedance matching network is coupled
`between an rf generator and a load to transmitrf power to the
`load. The impedance matching network includes a set of
`variable impedance elements. The method includes measur
`ing a network impedance value of the impedance matching
`network including the load at current values of the variable
`impedance elements. The method further includes comput
`ing directions (i.e., increasing or decreasing) and relative
`rates of change for the variable impedance element values in
`response to the network impedance of the network Such that
`the directions and relative rates of change for the variable
`impedance elements are adapted to change the reflected
`power in the direction of the most rapid decrease in reflected
`power. In addition, the method includes driving the variable
`impedance elements by adjusting the variable impedance
`elements in the computed directions by the computed rela
`tive rates of change Such that the variable impedance ele
`ments are driven to new current values in the direction of the
`most rapid decrease in reflected power. The method also
`includes repeating operations (a) through (c) until a desired
`level of tuning precision is obtained at the current values of
`the variable impedance elements.
`
`20 Claims, 13 Drawing Sheets
`
`502
`-
`
`504
`- 506
`-
`RF
`Generator
`
`RF
`matching
`network
`
`DC
`Motor
`
`Motor
`
`
`
`
`
`Page 1 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`U.S. Patent
`
`Jul. 10, 2001
`
`Sheet 1 of 13
`
`US 6,259,334 B1
`
`
`
`
`
` (?uV JOJd) | ‘5)|–|
`
`~ool
`
`Page 2 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`U.S. Patent
`
`Jul. 10, 2001
`
`Sheet 2 of 13
`
`US 6,259,334 B1
`
`
`
`c
`S
`
`n
`o
`CD
`O
`S w
`O
`CD
`
`s
`
`Page 3 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`U.S. Patent
`
`Jul. 10, 2001
`
`Sheet 3 of 13
`
`US 6,259,334 B1
`
`302
`Start Y-/
`---
`y
`Activate plasma processing system including
`rf generator
`
`304
`
`306
`Determine magnitude and phase of Z by -
`measuring a voltage, a Current, and the angle
`between the voltage and current
`
`31 O
`No z1
`Increase C2 if magnitude > 50 ohms;
`Decrease C2 if magnitude < 50 ohms
`
`
`
`Phase = 0 degrees?
`
`
`
`312
`
`No
`
`314
`z
`Increase C1 if phase < 0 degrees;
`Decrease C1 if phase > 0 degrees
`
`316
`
`Plasma processing
`
`complete?
`--
`
`318
`
`Yes
`Die Y
`O Done D
`
`' - MMMMMM
`
`FIG. 3
`(Prior Art)
`
`Page 4 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`U.S. Patent
`
`Jul. 10, 2001
`
`Sheet 4 of 13
`
`US 6,259,334 B1
`
`
`
`4OO
`1.
`
`150 17O 190 210 23O 250 27O 290
`C2(pF)
`
`FIG 4A
`
`1 OO
`
`
`
`45O
`1.
`
`Page 5 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`U.S. Patent
`
`Jul. 10, 2001
`
`Sheet 5 of 13
`
`US 6,259,334 B1
`
`
`
`Page 6 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`U.S. Patent
`
`Jul. 10, 2001
`
`Sheet 6 of 13
`
`US 6,259,334 B1
`
`
`
`9.O
`
`peoT
`
`eUuSelº
`
`Page 7 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`U.S. Patent
`
`Jul. 10, 2001
`
`Sheet 7 of 13
`
`US 6,259,334 B1
`
`702
`
`Calibrate motors with respect to variable
`Capacitors C1 and C2
`
`Start plasma processing
`
`704
`
`706
`
`708
`
`Measure Z for Current values of C1 and C2
`
`710
`
`716
`
`Desired tuning precision?
`
`Determine direction and relative magnitudes of changes in
`C1 and C2 that correspond to the steepest decrease in
`reflected power
`
`Drive capacitors C1 and C2 by adjusting capacitors C1 and
`C2 in the determined directions by the determined relative
`amounts
`
`712
`
`714
`
`FIG 7A
`
`
`
`
`
`
`
`
`
`
`
`Page 8 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`U.S. Patent
`
`Jul. 10, 2001
`
`Sheet 8 of 13
`
`US 6,259,334 B1
`
`712
`
`722
`
`Compute Z from Z C1, C2, L1, and L2
`
`Compute Z from Z for a set of neighboring values of
`variable capacitances C1 and C2
`
`Compute reflected power P from Z for each pair of
`neighboring capacitance values (C1, C2)
`
`
`
`Select neighboring pair of capacitance values (C1, C2")
`with the lowest reflected power P' where the selected
`capacitance values defining the adjustment direction of the
`capacitors C1 and C2, where the Selected capacitance
`values define the direction toward which to adjust the C1
`and C2
`
`732
`
`724
`
`726
`
`728
`
`73O
`
`FIG. 7B
`
`Page 9 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`U.S. Patent
`
`Jul. 10, 2001
`
`Sheet 9 of 13
`
`US 6,259,334 B1
`
`712 y
`
`742
`
`Compute Z from Z. C1, C2, L1, and L2
`
`Compute reflected power P from Z.
`
`744
`
`746
`
`
`
`Compute the negative gradient of the reflected power as a
`function of capacitors C1 and C2, where the negative
`gradient defines the direction toward which to adjust the C1
`and C2
`
`748
`
`750
`
`DOne
`
`FIG. 7C
`
`Page 10 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`U.S. Patent
`
`Jul. 10, 2001
`
`Sheet 10 Of 13
`
`US 6,259,334 B1
`
`
`
`s
`
`200 21 O 220 23O 24O 25O 26O 27O 28O 29O 300
`
`C2 (pF)
`
`FIG. 8
`
`Page 11 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`U.S. Patent
`
`Jul. 10, 2001
`
`Sheet 11 of 13
`
`US 6,259,334 B1
`
`902
`
`Calibrate motors and the associated variable
`capacitors C1 and C2
`
`Measure Z at C1 and C2
`
`Compute Z from Z, C1, C2, L1, and L2
`
`Determine a tune line, which is the locus of points
`(C1, C2) corresponding to a known approximate
`relationship at the tune condition
`
`
`
`
`
`Adjust C1 and C2 by different amounts so that the
`ratio of adjustment allows the tune path to moves the
`shortest distance to intersect the locus of points
`
`Follow the locus of points in the direction of
`decreasing reflected power
`
`916
`
`End
`
`904
`
`906
`
`908
`
`910
`
`912
`
`914
`
`FIG. 9
`
`Page 12 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`U.S. Patent
`
`Jul. 10, 2001
`
`Sheet 12 of 13
`
`US 6,259,334 B1
`
`
`
`85O
`8OO
`75O
`7OO
`65O
`C1 6OO
`(pF) 550
`
`5OO
`45O
`4OO
`35O
`2OO
`
`22O
`
`24O
`C2 (pF)
`
`26O
`
`28O
`
`3OO
`
`FIG 10
`
`Page 13 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`U.S. Patent
`
`Jul. 10, 2001
`
`Sheet 13 of 13
`
`US 6,259,334 B1
`
`1102
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Measure Z at Current values of
`variable capacitances C1 and
`C2
`
`1104
`
`Specified range of tuning
`precision?
`
`Yes
`
`Compute directions and relative
`magnitudes by which to change
`capacitors C1 and C2 to achieve
`the steepest decrease in
`reflected power
`
`Adjust variable capacitors C1
`and C2 in the computed
`directions and by the computed
`relative amounts corresponding
`to the steepest decrease in
`reflected power
`
`1108
`
`
`
`1112
`
`Adjust variable capacitors
`C1 and C2 in acCordance
`with impedance matching
`network Control method of
`Figure 3
`
`1110
`
`1114
`
`FIG. 11
`
`Page 14 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`US 6,259,334 B1
`
`1
`METHODS FOR CONTROLLING AN RF
`MATCHING NETWORK
`
`BACKGROUND OF THE INVENTION
`
`1. Field of the Invention
`The present invention relates generally to the manufacture
`of Semiconductor devices. More specifically, the present
`invention relates to improved methods and apparatus for
`tuning rf matching networks for a plasma processing cham
`ber.
`2. Description of the Related Art
`Semiconductor processing Systems are generally used to
`proceSS Semiconductor wafers for fabrication of integrated
`circuits. For example, plasmaenhanced Semiconductor pro
`ceSSes are commonly used in etching, oxidation, chemical
`Vapor deposition, or the like. The plasma-enhanced Semi
`conductor processes are typically carried out by means of
`plasma processing Systems.
`FIG. 1 illustrates a representative plasma processing SyS
`tem 100 for processing a semiconductor wafer 102. The
`plasma processing System 100 includes a plasma processing
`chamber 104, which is well known in the art. The processing
`chamber 104 includes an electrostatic chuck 112 for Sup
`porting and clamping the wafer 102 in place for plasma
`processing. The plasma processing System 100 also includes
`an rf generator 106 and an rf matching network 110 coupled
`to the rf generator 106 by means of a cable 108. The rf
`matching network 110 is coupled to deliver rf power from
`the rf generator 106 to the electrostatic chuck 112.
`When the rf generator 106 is energized after a source gas
`(not shown) has been introduced into the chamber 104, a
`plasma 114 is created from the source gas. The wafer 102 is
`disposed over the electroStatic chuck 112 to be processed by
`the plasma. A heat transfer gas (e.g., helium) 116 may be
`provided to the wafer 102 under pressure via one or more
`ports 118 through the electrostatic chuck 112. The heat
`transfer gas 116 acts as a heat transfer medium between the
`wafer 102 and electrostatic chuck 112 to facilitate control of
`the wafer temperature during processing.
`In this arrangement however, the rf power Supplied to the
`plasma processing chamber 104 may be reflected back from
`the plasma processing chamber 104, thereby reducing the
`efficiency of the plasma processing system 100. The rf
`power reflection is generally caused by a mismatch in
`impedance of the rf generator 106 and a load formed by the
`plasma 114 and the chuck 112. The rf generator 106 has an
`output impedance Zo, which is typically 502. The cable 108
`has a matching characteristic impedance equal to the output
`impedance of the rf generator 106. The plasma 114 and the
`chuck 112 together form the load characterized by a com
`plex load impedance Z. If Z, is not equal to Zo, which is
`the complex conjugate of Zo, then an impedance mismatch
`exists between the generator and the load.
`The rf matching network 110 is provided between the rf
`generator 106 and the plasma processing chamber 104 to
`minimize reflection of rf power from the plasma processing
`chamber 104. The rf matching network 110 typically
`includes two or more variable impedance elements (e.g.,
`capacitors, inductors). The variable impedance elements
`may be tuned to provide an impedance Z that matches the
`impedance of the rf generator 106.
`FIG. 2 shows a circuit diagram of an exemplary rf
`matching network 110 coupled to the load 202, which is
`equivalent to the combination of the electroStatic chuck 112
`and plasma 114. The rf matching network 110 includes a
`
`15
`
`25
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`2
`variable capacitor C1 coupled to an inductor L1 in Series.
`The rf matching network 110 also includes a variable
`capacitor C2 coupled in Series to an inductor L2. The
`variable capacitors C1 and C2 are coupled to each other at
`a junction A. The electrode and plasma load 202 is coupled
`in Series with the inductor L2 and is coupled to a junction B.
`In this configuration, the variable capacitorS C1 and C2
`may be tuned to provide an impedance Z acroSS the
`junctions A and B, which matches the impedance of the rif
`generator 106. The impedance, Z, represents the total
`impedance of the network 110 in combination with the load
`202. Ideally, when the impedance Z is equal to the output
`impedance of the rf generator 106, the rf power reflected is
`at Zero percent.
`For example, if the impedance of the rf generator 106 is
`502, then the magnitude and phase of complex impedance
`Z need to be equal to 50S2 and Zero degrees, respectively,
`in order to minimize power reflection. The set of values of
`the capacitors C1 and C2 at which the complex impedance
`Z equals the output impedance of the rf generator 106 is
`referred to as a “tune' point or target point. Accordingly, the
`tune or target point is where the power reflection is at a
`minimum.
`Several techniques are known for tuning variable imped
`ance elements in an rf matching network. FIG. 3 illustrates
`a flow chart of a conventional method for tuning capacitors
`C1 and C2. The method starts in operation 302 and proceeds
`to operation 304, where the plasma processing system 100
`including the rf generator 106 is activated. At this time, the
`capacitors C1 and C2 are usually not set properly to the tune
`point. Thus, Some rf power is reflected back.
`Then in operation 306, the magnitude and phase of Z are
`determined by measuring a Voltage V, a current I, and an
`angle 0 between the Voltage and current in accordance with
`well known equation Z=|Ze", with Z=VI/I). In opera
`tion 308, it is determined whether the magnitude of Z is
`equal to a tune point value, for example, of 5092. If not, the
`operation proceeds to operation 310, where the variable
`capacitor C2 is adjusted by means of a computer and DC
`motors to match the impedance of the rf generator 106. If the
`magnitude of impedance Z is greater than the impedance
`of the rf generator 106, then the capacitance of C2 is
`increased. Conversely, if the magnitude of impedance of Z.
`is less than the impedance of the rf generator 106, then the
`capacitance of C2 is decreased.
`After adjusting capacitor C2 in operation 310 or if mag
`nitude is equal to the tune point in operation 308, the method
`proceeds to operation 312, where it is determined if the
`phase is equal to Zero degrees. If the phase is determined to
`be non-Zero, the method proceeds to operation 314, where
`capacitor C1 is adjusted to change the phase to reach the
`target impedance phase of Zero degrees. For example, if the
`phase of the impedance Z is less than Zero, then the
`capacitance of C1 is increased. Conversely, if the phase is
`greater than Zero, the capacitance of C1 is decreased. It
`should be noted that the variable capacitors C1 and C2 are
`adjusted by means of a computer and DC motors.
`Specifically, a computer may drive the DC motors to adjust
`the capacitance of the capacitors C1 and C2 So as to reach
`the target tune point.
`After adjusting capacitor C1 in operation 314 or if the
`phase is equal to Zero in operation 312, the method proceeds
`to operation 316, where it is determined whether the plasma
`processing is complete. If So, the method terminates opera
`tion 318. Otherwise, the method proceeds back to operation
`306 to continue tuning the capacitors C1 and C2 in the
`
`Page 15 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`3
`continually varying conditions (e.g., varying load, drifting
`tuning motors) of the plasma processing System.
`Unfortunately, the method described in FIG. 3 may not
`efficiently tune the capacitors to the target point for certain
`ranges of capacitance values. The problem is illustrated
`more clearly in FIGS. 4A and 4B. FIG. 4A illustrates an
`exemplary graph 400 plotting the magnitude 402 of imped
`ance Z and reflected power 404 as a function of the value
`of capacitor C2 for capacitor C1 held fixed at its tune value.
`The tune value of capacitor C2 is about 231 pF.
`A line 406 indicates the impedance tune value of 5092.
`The magnitude 402 and the tune line 406 intersect at points
`408 and 410. The point 410 represents the tune point at
`which the power reflected is at a minimum while the point
`408 corresponds to a capacitance value of about 255 pF. At
`capacitance C2 values of less than 255 pF, the tuning method
`of FIG.3 works efficiently by increasing C2 if Z is greater
`than 502 and decreasing C2 if Z is less than 502.
`However for C2 values above 255 pF of point 408, the above
`method adjusts the value of C2 in the wrong direction. For
`instance at C2 values above 255 pF, the method increases C2
`even though C2 is already above its tune value, thereby
`moving away from the tune point.
`Similarly, the method of FIG. 3 does not efficiently tune
`the impedance angle 0 in some range of C1 values. FIG. 4B
`shows a graph 450 plotting the phase 0452 of the impedance
`Z and reflected power 454 as a function of the capacitor
`C1. In the graph 450, the capacitor C2 is held fixed at its tune
`value of 231 pF. A tune line 456 represents the 0 value of
`Zero degrees. The phase line 452 and the tune line 456
`intersect at points 458 and 460. The point 460 represents a
`tune point where the power reflected is at a minimum. The
`value of C1 at the point 460 is about 837 pF,
`The point 458 corresponds to a capacitance value of about
`800 pF. At C1 values of greater than the capacitance at the
`point 458, the method of FIG. 3 will increase C1 if 0 is
`smaller than 0 degrees and decrease C1 if 0 is larger than 0
`degrees. However, for C1 values below the capacitance at
`the point 458, the method decreases C1 even though C1 is
`already below its tune value. Hence, the method adjusts C1
`in the wrong direction. Thus, by adjusting C1 and C2 values
`into wrong directions for certain capacitance value ranges,
`the method may never find tune point or may Substantially
`delay the determination of the tune point.
`Another method for controlling a matching network is
`described in U.S. Pat. No. 5,689,215 by Richardson et al.,
`which is incorporated herein by reference. This method
`physically changes the values of C1 and C2 in a prescribed
`order and measures the percentage of reflected power as a
`function of capacitor value as each capacitor is varied. In So
`doing, the method finds two pairs of values (C1, C2) that
`correspond to local minima in the reflected power. A Straight
`line connecting these two pairs of values points generally
`toward the true tune point. The method varies the values of
`C1 and C2 to follow the straight line. This process is then
`repeated until the desired level of reflected power is
`obtained.
`While the method described in U.S. Pat. No. 5,689,215
`Works well, it may require Substantial time to find a tune
`point. This is because physically varying the capacitors
`generally requires more time than computer implemented
`calculations.
`U.S. Pat. No. 5,793,162 by Barnes et al., which is incor
`porate herein by reference, also describes a technique for
`controlling matching network of a vacuum plasma proces
`Sor. The technique for controlling a matching network also
`
`15
`
`25
`
`35
`
`40
`
`45
`
`50
`
`55
`
`60
`
`65
`
`US 6,259,334 B1
`
`4
`relies on measurements of reflected power. Specifically, the
`technique involves varying a capacitor in one direction. If
`the reflected power increases, the adjustment direction of the
`capacitor is reversed. On the other hand, if the reflected
`power decreases, the adjustment direction of the capacitor
`remains the Same as long as the reflected power continues to
`decrease. While this technique works well, it may not
`produce tune points in a speedy manner. For example, the
`reversals of directions may further add to the time required
`to find the tune point.
`Yet another method for tuning a matching network using
`a predictor-corrector control system is described in U.S. Pat.
`No. 5,187,454 by Collins et al., which is incorporated herein
`by reference. This method involves estimating the values of
`matching network values at their tuned condition and mov
`ing the network values toward the estimated tune position.
`Unfortunately, this method also adjusts the variable imped
`ance elements of the match network in the wrong directions
`under Some circumstances. This can occur, for example,
`when the load impedance is not constant but is itself a
`rapidly varying function of the values of the variable imped
`ance elements.
`In View of the foregoing, what is needed are methods and
`Systems for more rapidly and Stably tuning an rf matching
`network to deliver maximum power to a vacuum plasma
`processing chamber.
`
`SUMMARY OF THE INVENTION
`Broadly Speaking, the present invention fills these needs
`by providing methods and Systems for tuning an rf matching
`network. It should be appreciated that the present invention
`can be implemented in numerous ways, including as a
`process, an apparatus, a System, a device, a method, or a
`computer readable medium. Several inventive embodiments
`of the present invention are described below.
`In accordance with one embodiment, the present inven
`tion provides a method for tuning an impedance matching
`network toward a tune point where power reflection is at a
`minimum. The impedance matching network is coupled
`between an rf generator and a load to transmitrf power to the
`load. The impedance matching network includes a set of
`variable impedance elements. The method includes (a) mea
`Suring a network impedance value of the impedance match
`ing network including the load at current values of the
`variable impedance elements; (b) computing directions (i.e.,
`increasing or decreasing) and relative rates of change for the
`variable impedance element values in response to the net
`work impedance of the network Such that the directions and
`relative rates of change for the variable impedance elements
`are adapted to change the reflected power in the direction of
`the most rapid decrease in reflected power; (c) driving the
`variable impedance elements by adjusting the variable
`impedance elements in the computed directions by the
`computed relative rates of change Such that the variable
`impedance elements are driven to new current values in the
`direction of the most rapid decrease in reflected power; and
`(d) repeating operations (a) through (c) until a desired level
`of tuning precision is obtained at the current values of the
`variable impedance elements.
`In another embodiment, the present invention provides a
`method for tuning an impedance matching network to a tune
`point where power reflection is at a minimum. The imped
`ance matching network is coupled between an rf generator
`and a load to transmit rf power to the load. The impedance
`matching network includes a set of variable impedance
`elements. The method includes measuring a network imped
`
`Page 16 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`US 6,259,334 B1
`
`15
`
`25
`
`S
`ance value of the impedance matching network including the
`load at current values of the variable impedance elements.
`The method also includes computing a load impedance in
`response to the network impedance value and the variable
`impedance element values. In addition, the method includes
`determining a tune equation adapted to approximate a tune
`line through the tune point. The tune line includes a plurality
`of Sets of values for the variable impedance elements.
`Furthermore, the method includes adjusting the impedance
`element values So as to reach a nearest Set of values for the
`variable impedance elements on the tune line. The method
`further includes adjusting the impedance element values
`along the tune line in the direction of decreasing reflected
`power.
`In yet another embodiment, the present invention pro
`vides a System for tuning an impedance matching network to
`a tune point where power reflection is at a minimum. The
`System includes an rf generator, a cable, an impedance
`matching network, and an impedance tuning apparatus. The
`rf generator is adapted to generate rf power and has an
`impedance value of Zo. The cable is coupled to the rif
`generator to transmit the rf power and also has the charac
`teristic impedance value of Zo. The impedance matching
`network is coupled to receive the rf power for delivery to a
`load and includes a Set of variable impedance elements that
`are capable of being tuned. The impedance tuning apparatus
`is coupled to tune the impedance matching network. The
`impedance tuning apparatus including means for measuring
`a network impedance value of the impedance matching
`network including the load at current values of the variable
`impedance elements. The impedance tuning apparatus fur
`ther includes means for determining directions and relative
`rates of change for the variable impedance element values in
`response to the network impedance of the network Such that
`the directions and relative rates of change for the variable
`impedance elements are adapted to change the reflected
`power in the direction of the most rapid decrease in reflected
`power. Additionally, the impedance tuning apparatus
`includes means for driving the variable impedance elements
`by adjusting the variable impedance elements in the com
`40
`puted directions by the computed relative rates of change
`Such that the variable impedance elements are driven to new
`current values in the direction of the most rapid decrease in
`reflected power. The driving means is adapted to adjust the
`variable impedance elements until a desired level of power
`reflection is obtained.
`In accordance with yet another embodiment, the present
`invention provides a System for tuning an impedance match
`ing network to a tune point where power reflection is at a
`minimum. The System includes an rf generator, a cable, an
`impedance matching network, and an impedance tuning
`apparatus. The rf generator is adapted to generate rf power
`and is characterized by an impedance value of Zo. The cable
`is coupled to the rf generator to transmit the rf power and
`also has the impedance value of Zo. The impedance match
`ing network is coupled to receive the rf power for delivery
`to a load and includes a set of variable impedance elements
`that are capable of being tuned. The impedance tuning
`apparatus is coupled to tune the impedance matching net
`work. The impedance tuning apparatus includes means for
`measuring a network impedance value of the impedance
`matching network including the load at a set of current
`values of the variable impedance elements. The impedance
`tuning apparatus also includes means for computing a load
`impedance in response to the network impedance value and
`the variable impedance element values. In addition, the
`impedance tuning network includes means for determining
`
`6
`a tune equation adapted to approximate a tune line through
`the tune point, the tune line including a plurality of Sets of
`values for the variable impedance element values.
`Furthermore, the impedance tuning network includes means
`for adjusting the impedance element values So as to reach a
`nearest Set of values for the variable impedance elements on
`the tune line. Additionally, the impedance tuning network
`includes means for adjusting the impedance element values
`along the tune line in the direction of decreasing reflected
`power.
`Advantageously, the present invention tunes the rf match
`ing network in the direction of the Steepest decrease in
`reflected power So that Substantial time is Saved in approach
`ing the tune point. In addition, the methods and Systems of
`the present invention calculate the correct directions in
`which to change the variable impedance elements without
`requiring time-consuming preliminary physical movements
`of the elements. These and other advantages of the present
`invention will become apparent upon reading the following
`detailed descriptions and Studying the various figures of the
`drawings.
`BRIEF DESCRIPTION OF THE DRAWINGS
`The present invention will be readily understood by the
`following detailed description in conjunction with the
`accompanying drawings, and like reference numerals des
`ignate like Structural elements.
`FIG. 1 illustrates a representative plasma processing Sys
`tem for processing a Semiconductor wafer.
`FIG. 2 shows a circuit diagram of an rf matching network
`coupled to an unknown load, which is equivalent to the
`combination of an electroStatic chuck and plasma.
`FIG. 3 illustrates a flow chart of a conventional method
`for tuning capacitorS C1 and C2.
`FIG. 4A illustrates an exemplary graph plotting magni
`tude of impedance Z and reflected power as a function of
`the value of capacitor C2.
`FIG. 4B shows a graph plotting the phase 0 of the
`impedance Z and reflected power as a function of the value
`of capacitor C1.
`FIG. 5 illustrates an exemplary plasma processing System
`500 for processing a semiconductor wafer 502 in accordance
`with one embodiment of the present invention.
`FIG. 6 shows a more detailed circuit diagram of the rf
`matching network coupled to a load in accordance with one
`aspect of the present invention.
`FIG. 7A illustrates a flowchart of a method performed by
`the impedance tuning apparatuS 510 for tuning the rf match
`ing network 508 in accordance with one embodiment of the
`present invention.
`FIG. 7B shows a more detailed flowchart of an operation
`of determining an intermediate Set of values for the variable
`capacitors C1 and C2 in accordance with one embodiment of
`the present invention.
`FIG. 7C illustrates a more detailed flowchart of an opera
`tion of determining an intermediate Set of values for the
`variable capacitors C1 and C2 in accordance with another
`embodiment of the present invention.
`FIG. 8 shows a graph of a pair of tuning paths.
`FIG. 9 illustrates a flowchart of a method performed by
`the impedance tuning apparatus for tuning the matching
`network in accordance with one embodiment of the present
`invention.
`FIG. 10 shows a graph of a tuning path obtained by the
`method described in FIG. 9.
`
`35
`
`45
`
`50
`
`55
`
`60
`
`65
`
`Page 17 of 22
`
`ADVANCED ENERGY INDUSTRIES INC.
`Exhibit 1007
`
`

`

`US 6,259,334 B1
`
`15
`
`25
`
`35
`
`40
`
`7
`FIG. 11 illustrates a flow chart of the combined method
`implemented by the impedance tuning apparatus for tuning
`the matching network.
`DETAILED DESCRIPTION OF THE
`PREFERRED EMBODIMENTS
`An invention is described herein for methods and Systems
`for tuning an rf matching network. In the following
`description, numerous Specific details are Set forth in order
`to provide a thorough understanding of the present inven
`tion. It will be obvious, however, to one skilled in the art,
`that the present invention may be practiced without Some or
`all of these specific details. In other instances, well known
`proceSS Steps have not been described in detail in order not
`to unnecessarily obscure the present invention.
`FIG. 5 illustrates an exemplary plasma processing System
`500 for processing a semiconductor wafer 514 in accordance
`with one embodiment of the present invention. The plasma
`processing System 500 includes a plasma processing cham
`ber 502, rf generator 504, a cable 506, an rf matching
`network 508, and an impedance tuning apparatus 510. The
`rf generator 504 generates rf power for transmission through
`the cable 506. The cable 506 is coupled between the rf
`generator 504 and the rf matching network 508 for trans
`mitting rf power to the plasma processing chamber 502
`through the rf matching network 508. The impedance tuning
`device 510 is coupled to the rf matching network 508 for
`tuning variable elements in the rf matching network 508
`toward a tune point.
`The plasma processing chamber 502 is coupled to the rf
`matching network 508 to receive the rf power. The plasma
`processing chamber 502 includes an electrostatic chuck 512
`for Supporting and clamping a wafer 514 in place for plasma
`processing. The rf matching network 508 is coupled to
`deliver rf power from the rf generator 504 to the electrostatic
`chuck 512, which functions as an electrode. AS used herein
`in describing the invention, the terms electrode and electro
`Static chuck