U.S. patent application number 10/359557 was filed with the patent office on 2003-07-24 for method and apparatus for improved plasma processing uniformity.
Invention is credited to Mitrovic, Andrej S., Parsons, Richard, Quon, Bill H., Sirkis, Murray D., Strang, Eric J., Tsukamoto, Yuji.
Application Number | 20030137251 10/359557 |
Document ID | / |
Family ID | 22838144 |
Filed Date | 2003-07-24 |
United States Patent
Application |
20030137251 |
Kind Code |
A1 |
Mitrovic, Andrej S. ; et
al. |
July 24, 2003 |
Method and apparatus for improved plasma processing uniformity
Abstract
A method and apparatus for generating and controlling a plasma
(130) formed in a capacitively coupled plasma system (100) having a
plasma electrode (140) and a bias electrode in the form of a
workpiece support member (170), wherein the plasma electrode is
unitary and has multiple regions (R.sub.i) defined by a plurality
of RF power feed lines (156) and the RF power delivered thereto.
The electrode regions may also be defined as electrode segments
(420) separated by insulators (426). A set of process parameters
A={n, .tau..sub.i, .PHI..sub.i, P.sub.i, S; L.sub.i} is defined,
herein n is the number of RF feed lines connected to the electrode
upper surface at locations L.sub.i, .tau..sub.i is the on-time of
the RF power for the i.sup.th RF feed line, .PHI..sub.i is the
phase of the i.sup.th RF feed line relative to a select one of the
other RF feed lines, P.sub.i is the RF power delivered to the
electrode through the i.sup.th RF feed line at location L.sub.i,
and S is the sequencing of RF power to the electrode through the RF
feed lines. One or more of these parameters are adjusted so that
operation of the plasma system results in a workpiece (176) being
processed with a desired amount or degree of process
uniformity.
Inventors: |
Mitrovic, Andrej S.;
(Phoenix, AZ) ; Strang, Eric J.; (Chandler,
AZ) ; Sirkis, Murray D.; (Tempe, AZ) ; Quon,
Bill H.; (Brea, CA) ; Parsons, Richard;
(Phoenix, AZ) ; Tsukamoto, Yuji; (Wilmington,
MA) |
Correspondence
Address: |
PILLSBURY WINTHROP, LLP
P.O. BOX 10500
MCLEAN
VA
22102
US
|
Family ID: |
22838144 |
Appl. No.: |
10/359557 |
Filed: |
February 7, 2003 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
10359557 |
Feb 7, 2003 |
|
|
|
PCT/US01/24491 |
Aug 6, 2001 |
|
|
|
60223834 |
Aug 8, 2000 |
|
|
|
Current U.S.
Class: |
315/111.21 ;
315/39 |
Current CPC
Class: |
H01J 37/32009 20130101;
H01J 37/32532 20130101; H01J 37/32935 20130101; H01J 37/32091
20130101; H01J 37/32082 20130101 |
Class at
Publication: |
315/111.21 ;
315/39 |
International
Class: |
H01J 019/80 |
Claims
What is claimed is:
1. An electrode apparatus for use in plasma processing, comprising:
a) a unitary electrode; b) a RF power supply; and c) a RF
multiplexer electrically connected to said RF power supply and to a
plurality of locations on said unitary electrode via a
corresponding plurality of RF feed lines thereby establishing a
plurality of electrode regions corresponding to said plurality of
RF feed lines.
2. An apparatus according to claim 1, further comprising a
plurality of match networks arranged one in each of said plurality
of RF feed lines.
3. An apparatus according to claim 2, further including a control
system electrically connected for controlling the operation of said
RF power supply and said RF multiplexer.
4. A plasma reactor system for processing a workpiece, comprising:
a) a plasma chamber having sidewalls, an upper wall and a lower
wall defining an interior region capable of supporting a plasma; b)
a unitary electrode having a plurality of electrode regions,
arranged within said interior region adjacent said upper wall; c) a
RF multiplexer electrically connected to the plurality of electrode
regions of said unitary electrode via a corresponding plurality of
RF feed lines; d) corresponding to said plurality of RF feed lines;
and e) a workpiece support member, arranged in the interior region
adjacent said lower wall, for supporting the workpiece.
5. A system according to claim 4, further comprising: f) a control
system electrically connected to said RF multiplexer, for
controlling the operation of said RF multiplexer when processing
the workpiece.
6. A system according to claim 5, further comprising: g) a
plurality of match networks each arranged one in a respective one
of said plurality of RF feed lines.
7. A system according to claim 6, further comprising: h) a gas
supply system in pneumatic communication with said chamber interior
region, for supplying gas to said chamber interior region.
8. A system according to claim 7, further comprising: i) a
workpiece support member RF power supply electrically connected to
said workpiece support member, for electrically biasing said
workpiece support member.
9. A system according to claim 8, further comprising: j) a vacuum
system pneumatically connected to said chamber interior region.
10. A system according to claim 9, further comprising: k) a
workpiece handling system in operative communication with said
workpiece support member, for providing the workpiece to the
workpiece support member.
11. A system according to claim 5, further including a database
electrically connected to said control system.
12. A method of determining a set of optimum plasma process
parameters A*={n*, .tau..sub.i*, .PHI..sub.i*, P.sub.i*, S*;
L.sub.i*} for plasma processing, with a high degree of uniformity,
a workpiece in a plasma reactor chamber having an electrode with an
upper surface as part of a plasma reactor system, wherein n is the
number of RF feed lines connected to the electrode upper surface at
locations L.sub.i, t.sub.i is the on-time of the RF power for the
i.sup.th RF feed line, .PHI..sub.i is the phase of the i.sup.th RF
feed line relative to a select one of the other RF feed lines, Pi
is the RF power delivered to the electrode to location L.sub.i
through the i.sup.th RF feed line, and S is the sequencing of RF
power to the electrode through the RF feed lines, the method
comprising the steps of: a) setting initial values for process
parameters n, .tau..sub.i, .PHI..sub.i, P.sub.i, and S; and b)
processing one or more workpieces while varying one or more of said
process parameters to determine the optimized set of process
parameters A*={n*, .tau..sub.i*, .PHI..sub.i*, P.sub.i*, S*} that
achieve a process non-uniformity less than a predetermined
standard.
13. A method according to claim 12, wherein said step b) includes
the steps of: i) forming a first plasma within the reactor chamber
having characteristics corresponding to said process parameters and
processing a first workpiece for a predetermined process time; ii)
measuring the workpiece process uniformity; and iii) comparing the
workpiece process uniformity to a predetermined standard.
14. A method according to claim 13, wherein said step b) further
includes the step of: iv) reducing the workpiece process
non-uniformity by changing at least one of said process parameters
and repeating said steps i) through iii) using one of said first
workpiece and a workpiece other than said first workpiece, until
the workpiece process non-uniformity is less than said
predetermined standard.
15. A method according to claim 12, wherein in said steps a) and
b), the locations L.sub.i of the RF feed lines are parameters in
the set A of process parameters that can be varied.
16. A method according to claim 12, wherein the initial process
parameter values are determined with the assistance of computer
modeling.
17. A method according to claim 12, wherein said step b) includes
use of a linear processing model as a basis for varying at least
one of the process parameters.
18. A method according to claim 12, wherein said step b) includes
use of a nonlinear processing model as a basis for varying at least
one of the process parameters.
19. A method according to claim 12, wherein said step b) includes
providing RF power P.sub.i to a plurality of electrode segments in
a multi-segment electrode by multiplexing RF power via RF power
multiplexing.
20. A method according to claim 19, wherein said RF power
multiplexing is accomplished by programming a control system
electrically connected to a plurality of RF power supplies and
electronically controlling the activation of the RF power
supplies.
21. A method according to claim 12, wherein said step b) involves
providing RF power P.sub.i to a unitary electrode via RF power
multiplexing.
22. A method of processing a workpiece to be processed according to
claim 12, further including the steps, after said step b), of: c)
providing the workpiece to be processed in the reactor chamber; d)
forming an optimized plasma with the process chamber using the set
of optimized process parameters determined in said step b); and e)
processing the workpiece to be processed with the optimized
plasma.
23. A method of determining a set of optimum plasma process
parameters A*={n*, .tau..sub.i*, .PHI..sub.i*, P.sub.i*, S*;
L.sub.i*} for plasma processing, with a desired degree of
uniformity, a workpiece in a plasma reactor chamber having an
electrode with an upper surface as part of a plasma reactor system,
wherein n is the number of RF feed lines connected to the electrode
upper surface at locations L.sub.i, .tau..sub.i is the on-time of
the RF power for the i.sup.th RF feed line, .tau..sub.i is the
phase of the i.sup.th RF feed line relative to a select one of the
other RF feed lines, P.sub.i is the RF power delivered to the
electrode to location L.sub.i through the i.sub.th RF feed line,
and S is the sequencing of RF power to the electrode through the RF
feed lines, the method comprising the steps of: a) setting initial
values for process parameters n, .tau..sub.i, .PHI..sub.i, P.sub.i,
and S; and b) processing one or more workpieces while varying one
or more of said process parameters to determine the optimized set
of process parameters A*={n*, .tau..sub.i*, .PHI..sub.i*, P.sub.i,
S} that achieve a desired process uniformity.
24. A method according to claim 23, wherein said step b) includes
the steps of: i) forming a first plasma within the reactor chamber
having characteristics corresponding to said process parameters and
processing a first workpiece for a predetermined process time; ii)
measuring the workpiece process uniformity; and iii) comparing the
workpiece process uniformity to a predetermined standard.
25. A method according to claim 23, wherein said step b) further
includes the step of: iv) reducing the workpiece process
non-uniformity by changing at least one of said process parameters
and repeating said steps i) through iii) using one of said first
workpiece and a workpiece other than said first workpiece, until
the workpiece process non-uniformity is less than said
predetermined standard.
26. A method of processing a workpiece to be processed according to
claim 23, further including the steps, after said step b), of: c)
providing the workpiece to be processed in the reactor chamber; d)
forming an optimized plasma with the process chamber using the set
of optimized process parameters determined in said step b); and e)
processing the workpiece to be processed with the optimized plasma.
Description
[0001] This is a continuation of International Application No.
PCT/US01/24491, filed on Aug. 6, 2001, and also claims benefit of
U.S. application No. 60/223,834, filed Aug. 8, 2000, the contents
of both of which are incorporated herein in their entirety by
reference.
BACKGROUND OF THE INVENTION
[0002] The present invention pertains to plasma processing of
workpieces, and in particular pertains to a method and apparatus
for improving the uniformity of plasma processing.
[0003] Ionized gas or "plasma" may be used during processing and
fabrication of semiconductor devices, flat panel displays and other
products requiring etching or deposition of materials. Plasma may
be used to etch or remove material from semiconductor integrated
circuit wafers, or to sputter or deposit material onto a
semiconducting, conducting or insulating surface. Creating a plasma
for use in manufacturing or fabrication processes typically is done
by introducing a low-pressure process gas into a chamber
surrounding a workpiece such as an integrated circuit (IC) wafer.
The atoms or molecules of the low-pressure gas in the chamber are
ionized to form plasma by a radio frequency energy (power) source
after the gas molecules enter the chamber. The plasma then flows
over and interacts with the workpiece. The chamber is used to
maintain the low pressures required for plasma formation, to
provide a clean environment for processing the semiconductor
devices and to serve as a structure for supporting one or more
radio frequency energy sources.
[0004] Plasma may be created from a low-pressure process gas by
inducing an electron flow that ionizes individual gas atoms or
molecules by transfer of kinetic energy through individual
electron-gas molecule collisions. Typically, electrons are
accelerated in an electric field such as one produced by radio
frequency (RF) energy. This RF energy may be low frequency (below
550 KHz), high frequency (e.g., 13.56 MHz), or microwave frequency
(e.g., 2.45 GHz).
[0005] The two main types of etching in semiconductor processing
are plasma etching and reactive ion etching (RIE). A plasma etching
system typically includes a radio frequency energy source and a
pair of electrodes. A plasma is generated between the electrodes,
and the workpiece (i.e., substrate or wafer) to be processed is
arranged parallel to one of the electrodes. The chemical species in
the plasma are determined by the source gas(es) used and the
desired process to be carried out.
[0006] A problem that has plagued prior art plasma reactor systems
is the control of the plasma to obtain uniform etching and
deposition. In plasma reactors, the degree of etch or deposition
uniformity is determined by the design of the overall system, and
in particular by the design of the RF feed transmission and the
associated control circuitry.
[0007] In a plasma reactor system, the electrode is connected to a
RF power supply. The technological trend in plasma reactor design
is to increase the fundamental RF driving frequency of the RF power
supply from the traditional value of 13.56 MHz to 60 MHz or higher.
Doing so improves process performance, but increases the complexity
of reactor design.
[0008] One approach to improving etch and deposition uniformity has
been to use a multi-segment electrode. With reference to FIG. 1,
plasma reactor system 8 comprises reactor chamber 10 having an
interior 12, within which is arranged a segmented electrode 16
having separate thick conducting electrode segments 18 each with an
upper surface 18U and a lower surface 18L. A silicon slab or
"facing" (not shown) may be attached to each of the lower surfaces
18L of the segmented electrode by suitable attachment means to
control contamination due to sputtering of the metal electrode.
Electrode segments 18 are separated by an insulator 20 and are
powered by corresponding RF power supplies 26 via RF feed lines 30
connected to respective electrode segments. Power control to
electrode segments 18 is provided by a main control unit 36. Match
networks 40 arranged between RF power supplies 26 and electrode
segments 18 are tuned to provide the best match to the load
associated with a plasma 50 formed in interior region 12, so as to
optimize power transfer to the plasma.
[0009] Reactor system 8 includes a workpiece support member 60
opposite segmented electrode 16, upon which a workpiece 66, such as
a wafer, is supported. The design of segmented electrode 16 is such
that lower surfaces 18L of electrode segments 18 interfaces with a
vacuum region 70 in interior region 12. This puts electrode
segments 18 directly in contact with plasma 50 formed in vacuum
region 70, although if silicon facings as mentioned above (not
shown) are used, the surfaces of the silicon electrode facings will
be directly in contact with plasma 50. Numerous seals (not shown)
are required between insulators 20 and the electrode segments, and
between the chamber 10 and insulators 20, to isolate vacuum region
70.
[0010] Current plasma reactor systems can perform an etch process
with approximately 5% non-uniformity. This level of performance is
sufficient to meet near-term needs for state-of-the-art process
performance, but will soon be inadequate as the demands on the
manufacturing process increase to require, on a routine basis,
non-uniformity below 5%.
[0011] In light of the demands on improving process speed, one
technological trend in plasma reactor design is to increase the
fundamental RF frequency from the traditional value of 13.454 MHz
to 60 MHz or higher, as mentioned above. Doing so improves process
performance, but increases the complexity of reactor design. A
second trend in reactor design is to have multiple electrodes,
i.e., electrode segments, such as those discussed above in
connection with FIG. 1. However, multiple electrodes combined with
increased operating frequencies mean that delivering the correct
amount of RF power becomes more complicated because of capacitive
coupling between the electrode segments and greater sensitivity to
parasitic capacitive and inductive elements. This effect is
exacerbated by the shorter wavelengths of higher fundamental
frequencies. The result is increased difficulty in reducing process
non-uniformity.
[0012] In addition, current multi-segmented electrode plasma
reactors require a power supply for each electrode. Thus, if there
are five electrode segments, there must be five corresponding power
supplies (or separate amplifiers). This leads to high cost and
increased maintenance requirements and thus high wafer processing
costs. This cost and increased maintenance might be worthwhile if
there were a way to improve the performance of such a system to
provide a higher degree of etch or deposition uniformity beyond the
present limits of existing plasma processing systems.
BRIEF SUMMARY OF THE INVENTION
[0013] The present invention pertains to plasma processing of
workpieces, and in particular pertains to a method and apparatus
for improving the uniformity of the plasma processing.
[0014] The present invention is a method of and apparatus for
generating and controlling a plasma formed in a capacitively
coupled plasma system having a plasma electrode and a bias
electrode, wherein the plasma electrode has multiple regions
defined by RF power feed lines, with the size of each region being
dependent on the amount of RF power delivered thereto. The
electrode regions can also be defined as electrode segments
separated by insulators. The RF power to each electrode region is
independently controlled. In particular, the amplitude, phase,
frequency, and/or "on-time" during which the RF power is applied to
each RF feed line of the electrode can be varied, thereby affecting
the spatial distribution of the plasma-exciting electric field and
the plasma density (i.e., ion density) of the plasma.
[0015] Accordingly, a first aspect of the invention is an electrode
apparatus for use in plasma processing. The apparatus comprises a
unitary electrode having an upper surface and a lower surface. A
unitary electrode is an electrode, usually planar, for which the
entire electrode comprises either a single conductor or a plurality
of conductors that are interconnected by means of low resistance
ohmic contacts. A silicon facing; i.e., a so-called silicon
electrode, can be attached by various attachment means to the lower
surface of the unitary electrode in accordance with common
practice. An RF multiplexer is electrically connected to a
plurality of locations on the electrode upper surface via a
corresponding plurality of RF feed lines. The unitary electrode has
a plurality of electrode regions corresponding to the plurality of
RF feed lines. These electrode regions are akin to electrode
segments of a segmented electrode, differing therefrom in that the
electrode regions of a unitary electrode are not separated by
insulators. The apparatus can preferably include a plurality of
match networks arranged one in each RF feed line in said plurality
of RF feed lines. The apparatus can also preferably include a
control system electrically connected to the RF multiplexer, for
controlling the operation of said RF multiplexer when performing RF
multiplexing.
[0016] A second aspect of the present invention is a plasma reactor
system for processing a workpiece in a manner that achieves a high
degree of process uniformity. The system comprises a plasma chamber
having sidewalls, an upper surface and a lower surface defining an
interior region capable of supporting plasma. A unitary electrode
having an upper surface is arranged within the interior region
adjacent the upper wall. A RF multiplexer is electrically connected
to a plurality of locations on the upper surface of the unitary
electrode via a corresponding plurality of RF feed lines. The
electrode includes a plurality of electrode regions corresponding
to the plurality of RF feed lines. Also included in the interior
region adjacent the lower wall is a workpiece support member for
supporting the workpiece.
[0017] A third aspect of the present invention is a method of
determining a set of optimum plasma process parameters A*={n*,
.tau..sub.i*, .PHI..sup.i*, P.sub.i*, S*; L.sub.i} for plasma
processing a workpiece with a high degree of uniformity. The method
is carried out in a plasma reactor chamber having an electrode with
an upper surface as part of a plasma reactor system. The parameters
are defined as follows: n is the number of RF feed lines connected
to the electrode upper surface at locations L.sub.i, .tau..sub.i is
the on-time of the RF power for the i.sup.th RF feed line,
.PHI..sup.i is the phase of the i.sup.th RF feed line relative to a
selected one of the other RF feed lines, P.sub.i is the RF power
delivered to the electrode at location L.sub.i through the i.sup.th
RF feed line, and S is the sequencing of RF power to the electrode
through the RF feed lines. The method comprises a first step of
setting initial values for process parameters n, .tau..sub.i,
.PHI..sub.i, P.sub.i, and S, and then a second step of processing
one or more workpieces while varying one or more of said process
parameters to determine the optimized set of process parameters
A*={n*, .tau..sup.i*, .PHI..sup.i*, P.sub.i*, S*} that achieve a
process non-uniformity less than the predetermined standard. The
second step includes the steps of forming a first plasma within the
reactor chamber having characteristics corresponding to the initial
process parameters and processing a first workpiece for a
predetermined process time, measuring the workpiece process
uniformity, and comparing the workpiece process uniformity to a
predetermined standard. If the workpiece process non-uniformity is
greater than the predetermined standard, then at least one of the
process parameters is changed and the above-steps repeated (using
either the first workpiece or a different workpiece) until the
workpiece process non-uniformity is less than the predetermined
standard.
[0018] A fourth aspect of the invention involves processing a
workpiece using the optimized process parameters deduced as
described above in connection with the third aspect of the
invention.
[0019] A fifth aspect of the invention is the method of the first
aspect of the invention, but carried out to achieve a desired
degree of process uniformity. Such a method might be performed
where there is a need to purposely provide a certain amount of
process non-uniformity to counter other process effects.
BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING
[0020] FIG. 1 is a schematic cross-sectional diagram of a prior art
plasma reactor system having a segmented electrode comprising a
plurality of electrode segments with corresponding RF power
supplies connected thereto;
[0021] FIG. 2A is a schematic cross-sectional diagram of the plasma
reactor system of the present invention having a unitary electrode
with multiple RF feed lines connected at one end to various
locations on the unitary electrode upper surface, and connected at
the opposite ends to a RF power multiplexer;
[0022] FIG. 2B is a block diagram of an embodiment of one component
of the system of FIG. 2A;
[0023] FIG. 3 is a flow diagram of the method according to the
present invention of optimizing the process parameters of the
plasma reactor system of FIG. 2;
[0024] FIG. 4A is a schematic diagram of a first alternate
embodiment for the RF power sources and electrode of the present
invention, wherein multiple RF power supplies feed power through
multiple match networks to electrode segments of a multi-segment
electrode;
[0025] FIG. 4B is a schematic diagram of a second alternate
embodiment for the RF power sources and electrode of the present
invention, wherein a single RF power supply feeds power through a
match network and a multiplexer distributing power to electrode
segments in a multi-segment electrode;
[0026] FIG. 4C is a schematic diagram of the first alternate
embodiment for the RF power sources and electrode of the present
invention, wherein multiple RF power supplies feed power through
multiple match networks to a unitary electrode; and
[0027] FIG. 4D is a schematic diagram of the second alternate
embodiment for the RF power sources and electrode of the present
invention, wherein a single RF power supply feeds power through a
match network and a multiplexer distributing power to electrode
regions R.sub.i of a unitary electrode.
DETAILED DESCRIPTION OF THE INVENTION
[0028] The present invention pertains to plasma processing of
workpieces, and in particular pertains to a method and apparatus
for improving the uniformity of the plasma processing.
[0029] With reference to FIG. 2A, plasma reactor system 100
comprises reactor chamber 102 with sidewalls 104, an upper wall 108
and a lower wall 112 defining an interior region 120 capable of
supporting plasma 130. Arranged within interior region 120 near
upper wall 108 is a unitary electrode 140 having an upper surface
140U and a lower surface 140L and a periphery 144. Electrode 140 is
referred to as the "plasma electrode". Insulators 146 are arranged
between electrode periphery 144 and sidewalls 104 to electrically
isolate electrode 140 from chamber 102. System 100 further includes
a RF power multiplexer 150. A RF power supply 152 is included in RF
multiplexer 150. Alternatively, an external RF power supply (as
illustrated by RF power supply 154) is provided.
[0030] FIG. 2B shows an exemplary layout of a multiplexer 150
acting to distribute RF power from a single RF generator 154
sequentially to a plurality of electrodes or electrode regions.
Although only two electrodes are illustrated for the sake of
simplicity, it will be understood that any number of electrodes can
be supplied. In FIG. 2B a central computer, or controller, 230
communicates with RF power generator 154, multiplexer 150 and one
or more match networks 160. Multiplexer 150 is composed of an
isolator and a coax switch. The isolator can be a commercially
available product of the type distributed by HD Communications
Corporation (1 Comac Loop, Ronkonkoma, N.Y. 11779) who specializes
in RF, microwave, fiber optic and wireless communications products.
In particular, that company distributes and custom builds UHF, VHF,
high power circulators and isolators. For instance, devices in the
frequency ranges of (VHF) 66-88, 144-225 MHz and (FM) 88-108 MHz
are currently available. The coax switch can be a commercially
available product distributed by Texas Towers, in particular the
Ameritron RCS-8V. The specifications for the Ameritron RCS-8V are
as follows: a frequency range of 0.5 to 450 MHz, power handling of
5000 W at 30 MHz, power handling of 1000 W at 150 MHz, <0.05 dB
insertion loss and 50 .OMEGA. impedance. Generator 154, match
networks 160 and the coax switch of multiplexer 150 are all
connected to be controlled by computer 230 to distribute power to
the various electrodes in any desired pattern.
[0031] Reverting to FIG. 2A, system 100 further includes multiple
(n) RF feed lines 156 (e.g., 156.sub.i, where i=1 to n) connected
to multiplexer 150 and that pass through tipper wall 108 and
connect at their opposite ends to different locations L.sub.i
(e.g., L.sub.1, L.sub.2, . . . L.sub.n) on electrode upper surface
140U. RF power multiplexer 150 is a device that directs RF power to
one or more of the RF feed lines 156 at a given time. The locations
L.sub.i of RF feed lines 156 define corresponding electrode regions
R.sub.i (e.g., R.sub.1, R.sub.2, . . . . R.sub.n) to which RF power
is provided. The sizes of regions R.sub.i depend on the amount of
RF power P.sub.i provided, relative to the amount of power provided
to adjacent regions but typically each of these regions extends
over an area of the order of the total electrode area divided by n.
For a segmented electrode, when only a single electrode segment is
powered, it is known that the highest process rate (etch or
deposition rate) on a wafer occurs directly under that powered
segment. Although a similar result occurs with a unitary electrode
when a RF feed with its L.sub.i axially located is the only powered
RF feed, when a RF feed with its L.sub.i located off-axis is the
only powered RF feed, the highest process rate on a wafer occurs at
a location diametrically opposite to that L.sub.i.
[0032] System 100 also preferably includes match networks 160
arranged in RF feed lines 156 between RF multiplexer 150 and
electrode 140. Match networks 160 are tuned to provide the best
match to the load presented by plasma 130 formed within interior
region 120 so as to optimize power transfer to the plasma.
[0033] Reactor system 100 further includes a workpiece support
member 170 arranged adjacent lower wall 112 opposite segmented
electrode 140, capable of supporting a workpiece 176, such as a
wafer, to be processed (e.g., etched or coated) by means of plasma
130.
[0034] System 100 also includes a workpiece handling system 180 in
operative communication with plasma chamber 102 (see dashed line
182), for placing workpieces 176 onto and removing workpieces 176
from workpiece support member 170. Also included is a gas supply
system 190 in pneumatic communication with chamber 102 via a gas
supply line 194 for supplying gas to chamber interior 120 to purge
the chamber and to create plasma 130. The particular gases included
in gas supply system 190 depend on the application. However, for
plasma etching applications, gas supply system 190 preferably
includes such gases as chlorine, hydrogen-bromide,
octafluorocyclobutane, and various other fluorocarbon compounds,
etc. For chemical vapor deposition applications, gas supply system
190 preferably includes silane, ammonia, tungsten-tetrachloride,
titanium-tetrachloride, and the like.
[0035] Further included in system 100 is a vacuum system 200 in
pneumatic communication with chamber 102 via a vacuum line 204.
Also included in system 100 is a workpiece support power supply 210
electrically connected to workpiece support member 170, for
electrically biasing the workpiece support member. This electrical
connection allows workpiece support member 170 to serve as a lower
electrode, also referred to as the "bias electrode".
[0036] System 100 also includes a main control system 230, which is
in electronic communication with and controls and coordinates the
operation of workpiece handling system 180, gas supply system 190,
vacuum system 200, workpiece support power supply 210, and RF power
multiplexer 150 through electrical signals. Main control system 230
thus controls the operation of system 100 and the plasma processing
of workpieces 176 in the system, as described below.
[0037] In a preferred embodiment, main control system 230 is a
computer with a memory unit MU having both random-access memory
(RAM) and read-only memory (ROM), a central processing unit CPU
with a microprocessor (e.g., a PENTIUM.TM. processor from Intel
Corporation), and a hard disk HD, all electrically connected. Hard
disk HD serves as a secondary computer-readable storage medium, and
may be, for example, a hard disk drive for storing information
corresponding to instructions for control system 230 to carry out
the present invention, as described below. Control system 230 also
preferably includes a disk drive DD, electrically connected to hard
disk HD, memory unit MU and central processing unit CPU, wherein
the disk drive is capable of accepting and reading (and even
writing to) a computer-readable medium CRM, such as a floppy disk
or compact disk (CD), on which is stored information corresponding
to instructions for control system 230 to carry out the present
invention. It is also preferable that control system 230 has data
acquisition and control capability. A suitable control system 230
is a computer, such as a DELL PRECISION WORKSTATION 610.TM.,
available from Dell Corporation, Dallas, Tex.
[0038] System 100 also includes a database 240 electrically
connected to (or alternatively, integral to) control system 230 for
storing data pertaining to the plasma processing of workpiece 176,
and for also including predetermined sets of instructions (e.g.,
computer software) for operating system 100 via control system 230
to process the workpieces.
[0039] Method of Operation
[0040] The operation of system 100 involves setting numerous
process-related parameters that can be modified in optimizing RF
power delivery to electrode 140 in a manner that allows the etch or
deposition rate to be controlled to obtain a high degree of etch or
deposition uniformity (i.e., non-uniformity less than 5%).
[0041] These process parameters are the number n of RF feed lines
156 providing power to electrode 140, the power on-time .tau..sub.i
for the i.sup.th RF feed line (i=1 to n), the phasing .PHI..sub.i
of the i.sup.th RF feed line or combinations of RF feed lines
relative to a select one of the RF feed lines, the amount of power
P.sub.i delivered to the i.sup.th RF feed line, and the sequence S
of power P.sub.i delivered to the electrode via each RF feed line
and hence to each electrode section R.sub.i. An additional
parameter, which is typically fixed but in certain cases can be
varied, is the location L.sub.i (L.sub.1, L.sub.2, . . . L.sub.n)
at which each of RF feed lines 156 is attached to upper surface
140U of electrode 140. These process parameters form a process
parameter set A={n, .tau..sub.i, .PHI..sub.i, P.sub.i, S; L.sub.i}.
Any of the parameters in set A can be combined and varied
independently or in concert to achieve the desired workpiece
uniformity requirements. These parameters can also be utilized to
achieve differential workpiece processing when desired. In either
case, the parameter set A that contains an optimized set of
parameters is A*={n*, .tau..sub.i*, .PHI..sub.i*, P.sub.i S*;
L.sub.i*}.
[0042] The optimum parameter set A* can be achieved empirically
using the method comprising the following steps. With reference to
the flow diagram 300 of FIG. 3, once a workpiece is loaded onto
workpiece support member 170, the first step 301 is setting the
parameters n, .tau..sub.i, .PHI..sub.i, P.sub.i, S, and L.sub.i to
initial values. As mentioned above, L.sub.i is typically fixed, but
could be varied if the need arises, i.e., if varying the other
process parameters does not result in an acceptable process
uniformity. These initial values can be decided upon based on data
from previously run experiments stored in database 240 or through
the assistance of computer modeling. The second step 302 is forming
plasma 130 within interior region 120 based on the initial
parameter values set in step 301 and processing workpiece 176 with
the plasma. Plasma 130 will have time-varying characteristics
(e.g., plasma density, energy, etc.) corresponding to the initial
values of the process parameters. The plasma characteristics (i.e.
plasma density, energy, etc.) will vary in time in that, in
particular, their spatial distributions will vary in time
throughout the process run due to the RF power and/or other process
parameter sequencing through the process run. For instance, a first
electrode or electrode region is initially powered for a short time
and corresponding to this condition there is a particular spatial
distribution of the plasma properties. This short period of time is
followed by a second short period of time during which a second
electrode or electrode region is powered with a different
corresponding distribution of plasma properties. Each of these
short steps adds to form the entire process run.
[0043] In second step 302, workpiece 176 is processed with plasma
130 for a process time T.sub.P. During this time, RF multiplexer
150, under the direction of controller 230, delivers an amount of
RF power Pi to each of the n RF feed lines 156i for an
"on-time",.tau..sub.i<<.tau..sub.P. This process is referred
to herein as "RF power multiplexing." Moreover, the sequencing S
and phasing .PHI..sub.i of the n RF feed lines is also varied. The
sequencing S can be such that only one location L.sub.i at a time
is powered, or multiple locations L.sub.i at one time are
powered.
[0044] Thus, the etch or deposition distribution for the entire
process time .tau..sub.P can be considered as a linear
superposition or a non-linear combination of the etch distributions
achieved by applying RF power P.sub.i to each RF feed line 156
based on the initial set of process parameters A. Here, the various
parameters, such as the on-time .tau..sub.i can be different or the
same for each RF feed line 156.sub.i. Further, the amount of RF
power P.sub.i delivered can the same for all RF feed lines
156.sub.i as well.
[0045] With continuing reference to flow chart 300 of FIG. 3, the
next step 303 is measuring the workpiece etch or deposition (i.e.,
process) uniformity. This is preferably accomplished using
well-known optical interferometric techniques. The workpiece
uniformity measurement is based on the greatest etch
depth/deposition thickness minus the smallest etch depth/deposition
thickness divided by two times the mean etch depth/deposition
thickness of a large number of sites on the workpiece surface. For
a wafer with a diameter of about 20 centimeters, a reasonable
number of measurement sites is approximately 50 to 100. This
results in a quantitative workpiece uniformity value M.sub.U.
[0046] The next step 304 is assessing whether or not the workpiece
uniformity measurement M.sub.U is acceptable, i.e., less than a
certain predetermined standard (e.g., threshold) T.sub.U. It can be
the case that a certain degree of non-uniformity is sought to
counter other processing effects, such as thin-film thickness
variations across the surface of the workpiece. Such variations can
be accounted for by measuring the spatial variation in uniformity
as a function M.sub.U(x,y) and comparing it to a predetermined
standard represented by a spatially dependent function T(x,y) that
corresponds to the non-uniformity profile sought. Here, it is
assumed that the x-y plane lies in the plane of the workpiece
surface being processed.
[0047] If the result in step 304 is that the workpiece uniformity
is not acceptable (i.e., the non-uniformity is greater than a
predetermined standard), then in the next step 305, based on the
results of step 304, at least one of the initial process parameters
n, .tau..sub.i, P.sub.i, S and L.sub.i is changed in the effort to
converge to the optimum set of process parameters, A*={n*,
.tau..sub.i*, .PHI..sub.i*, P.sub.i*, S*; L.sub.i*}. Changing at
least one of the process parameters results in plasma 130 that has
time-varying characteristics different from the first-formed plasma
130. In this sense, plasma 130 represents first, second, third,
etc. plasmas similar to but different from one another that may
need to be formed in the process of deducing the optimum set of
process parameters. As mentioned above, one or more of the process
parameters in set A can be changed with the assistance of a
computer program or algorithm that models the plasma etch or
deposition process.
[0048] One approach to finding the optimum set of parameters A* is
through the use of a linear superposition or a non-linear
combination of etch or deposition rates based on each electrode
region R.sub.i For a linear superposition, this can be stated
as:
PR(x,y)=(1/T.sub.P).SIGMA..tau..sub.iW.sub.i[PR.sub.i(x,y)], Eq.
1
[0049] where PR(x,y) is the overall process rate, PR.sub.i(x,y) is
the process rate for each electrode region R.sub.i, T.sub.P is the
total process time (i.e., the sum of the .tau..sub.i), and W.sub.i
is a weighting coefficient. The summation is performed by summing
from 1 to n. The weighting coefficient W.sub.i is a function of at
least one of the above-defined process parameters. For a linear
optimization, W.sub.i will typically be in the range of 0.9 to 1.1.
Here, the weighting coefficients W.sub.i may be determined
empirically based on the measurements made in step 303 by varying
one or more of the parameter values to arrive at the required
weighting coefficients W.sub.i. Other more sophisticated computer
models of the plasma etch process can also be used to assist in
converging to the optimum process parameter set A*. Also, it will
be apparent to one skilled in the art that non-linear equations can
also be formed and solved in the manner similar to Eq. 1 to find
the set of optimum parameters A*.
[0050] It should be noted here that though there can be a total of
n RF feed lines 156 connected to upper surface 140U, fewer than all
n RF feed lines may be activated. Here, it is assumed that n is the
number of RF feed lines 156.sub.i chosen to be activated. This
number can be less than the total number of feed lines connected to
upper surface 140U of electrode 140 because optimization of the
process parameters can require that a certain group of the n RF
feed lines not be activated.
[0051] Steps 302 to 305 are repeated until the measured workpiece
uniformity M.sub.U is acceptable, thereby defining the optimum
process parameter set A*. Once the iteration of steps 301-305 is
completed, then in step 306 the optimum process parameter set A* is
defined and recorded in database 240 and/or in control system 230
in memory unit MU. In the next step 307, a workpiece 176, such as a
semiconductor wafer, is provided by placing the workpiece onto
workpiece support member 170 via workpiece handling system 180.
Then, in the next step 308, main control system 230 controls the
formation of an "optimized" version of plasma 130 and controls the
processing of workpiece 176 provided in step 307 according to the
optimum set of process parameters A* established in step 306 and
recorded in database 240 and/or control system 230. Step 308 is
carried out for one or more workpieces 176. If workpiece 176 needs
to be processed with a new plasma process step requiring plasma
different from a particular form of plasma 130, then the steps of
flow diagram 300 are repeated for the new plasma process.
[0052] By way of example, n can be five, but might also range
between two and ten. A typical process time requires on the order
of 60 seconds. If only one RF feed is excited at any given time,
then if four sequences S occur with five RF feeds, each sequence S
will last for 15 sec. If all of the .tau..sub.i are equal, each
.tau..sub.i will be three seconds. If the .tau..sub.i are too
short, the demands on the RF matching network(s) can be too great;
i.e., steady-state conditions may not be attained during the time
.tau..sub.i.
[0053] With five RF feeds, one RF feed can be located on the axis
of symmetry, with the other four located ninety degrees apart
centered on a circle with a diameter approximately equal to 2/3 the
wafer diameter. The parameter .PHI..sub.i is meaningful only if two
or more RF feeds are powered simultaneously. The value for P.sub.i
can be nearly equal, but need not be so. To provide high process
throughput, it is preferable that the value for Pi should be about
the same as the power that would be delivered via a single feed to
a conventional electrode.
[0054] Alternate Embodiments
[0055] The present invention, as described above, is a method and
apparatus for delivering power at different locations to a unitary
electrode. The process parameter optimization and operation method
described above can also be applied to a number of alternate
structural embodiments as shown in FIGS. 4A-D, below, including a
multi-segment electrode having a plurality of electrode
segments.
[0056] With reference to FIG. 4A, multiple RF power supplies 400
feed power through corresponding multiple RF feed lines 406 and
multiple match networks 410 to corresponding electrode segments 420
in a multi-segment electrode. Electrode segments 420 are separated
by insulators 426. In this first alternative embodiment, the number
of power supplies, match networks, and electrode segments can be
two or greater. A control system 440 similar if not identical to
control system 230 is programmed to control the operation of RF
power supplies 400 so that they are multiplexed in a manner that
replicates the RF power multiplexing operation of RF multiplexer
150 of apparatus 100.
[0057] With reference now to FIG. 4B, a single RF power supply 400
feeds power through a single match network 410 and a multiplexer
450 distributes power to electrode segments 420 of the
multi-segment electrode via multiple RF feed lines 460. The number
of electrode segments 420 can be two or greater. As described above
earlier in connection with system 100, control system 440, similar
if not identical to control system 230, controls the operation of
multiplexer 450 during the operation of system 100. In the system
shown in FIG. 4B, match network 410 may need to be to programmed to
adjust itself each time multiplexer 450 switches.
[0058] With reference now to FIG. 4C, multiple RF power supplies
400 feed power through multiple match networks 410 to a unitary
electrode 140, and control system 440, similar if not identical to
control system 230, is programmed to control the operation of RF
power supplies 400 so that they are multiplexed in a manner that
replicates the RF power multiplexing operation of RF multiplexer
150 of apparatus 100.
[0059] With reference now to FIG. 4D, there is shown a system
essentially the same as that described above in connection with
system 100, except that a single adjustable match network 410 is
used instead of a plurality of match networks. In the system of
FIG. 4D, a single RF power supply 400 feeds power through match
network 410 and a multiplexer 450 that distributes power to
electrode regions R.sub.i of a unitary electrode.
[0060] In alternate embodiments, the embodiments of FIGS. 4A and 4B
are extended to the use coil antennas wherein an electrode 420 can
be a RF antenna.
[0061] The many features and advantages of the present invention
are apparent from the detailed specification and thus it is
intended by the appended claims to cover all such features and
advantages of the described method that follow in the true spirit
and scope of the invention. Further, since numerous modifications
and changes will readily occur to those of ordinary skill in the
art, it is not desired to limit the invention to the exact
construction and operation illustrated and described. Moreover, the
methods and apparatus of the present invention, like related
apparatus and methods used in the semiconductor arts that are
complex in nature, are often best practiced by empirically
determining the appropriate values of the operating process
parameters, or by conducting computer simulations to arrive at the
optimum process parameters for a given application. Accordingly,
all suitable modifications and equivalents should be considered as
falling within the spirit and scope of the invention.
* * * * *