Ex Parte Ikari et al

Board of Patent Appeals and Interferences

Nov 22, 2011

10515730 (B.P.A.I. Nov. 22, 2011)

UNITED STATES PATENT AND TRADEMARK OFFICE
_______________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
_______________

Ex parte YOSHIMITSU IKARI,
HIROSHI TAMAGAKI, and TOSHIMITSU KOHARA
______________

Appeal 2010-010975
Application 11/515,730
Technology Center 1700
_______________

Before CHARLES F. WARREN, CATHERINE Q. TIMM, and
MICHAEL P. COLAIANNI, Administrative Patent Judges.

WARREN, Administrative Patent Judge.

DECISION ON APPEAL
Applicants appeal to the Board from the decision of the Primary
Examiner finally rejecting claims 1-3, 5, 6, and 16-18 in the Office Action
mailed November 17, 2009. 35 U.S.C. §§ 6 and 134(a) (2002); 37 C.F.R.
§ 41.31(a) (2009).
An oral hearing was held November 10, 2011.1

1 An appeal, whether on brief or heard, is decided on the record. 37 C.F.R.
§ 41.37(c)(1)(vii) (2006) provides in pertinent part: “Any arguments or
authorities not included in the brief or reply brief filed pursuant to § 41.41
Appeal 2010-010975
Application 11/515,730
2
We reverse the decision of the Primary Examiner.
Claim 1 illustrates Appellants’ invention of a method for reactive
sputtering using a reactive sputtering apparatus including a sputtering
vaporization source having a metal target, and a sputtering power source,
and is representative of the claims on appeal:
1. A method for reactive sputtering using a reactive sputtering
apparatus including a sputtering vaporization source provided with a metal
target disposed in a vacuum chamber, a sputtering power source to drive the
sputtering vaporization source, and an introduction mechanism to introduce
an inert gas for sputtering and a reaction gas for forming a compound with
sputtering metal into the vacuum chamber, comprising the steps of:
performing reactive sputtering film formation on a substrate disposed
in the vacuum chamber, wherein a plasma emission is generated forward of
the sputtering vaporization source;
performing target voltage control to set a target voltage Vs for the
sputtering vaporization source based on the spectrum of the plasma emission
generated forward of the sputtering vaporization source; and
performing constant-voltage control to control the voltage of the
sputtering power source to be the target voltage Vs by comparing an actual
output voltage of the sputtering power source to the target voltage Vs, and
issuing an output voltage control command based on a difference between
the actual output voltage of the sputtering power source and the target
voltage Vs, wherein a control speed of the constant voltage control is higher
than a control speed of the target voltage control,
wherein the control response time Tv of the constant-voltage control
is within the range of 0.1 milliseconds to 0.1 seconds and the control
response time To of the target voltage control is within the range of 0.1
seconds to 60 seconds.
Appellants request review of the grounds of rejection under 35 U.S.C.
§ 103(a) advanced on appeal by the Examiner: claims 1, 5, 6, and 17 over

will be refused consideration by the Board, unless good cause is shown.”
See also Manual of Patent Examining Procedure (MPEP) §§ 1205.02 and
1209 (8th ed., Rev. 3, August 2005; 1200-14 and 1200-48).
Appeal 2010-010975
Application 11/515,730
3
Terry (US 6,106,676); claim 2 over Terry in view of Nakamura (US
4,936,964) and Tanaka (US 4,894,132); claim 3 over Terry in view of
Nakamura; claim 16 over Terry in view of Tanaka and Brooks (US
5,963,320); and claim 18 over Terry in view of Tanaka and Brooks. App.
Br. 4-5; Ans. 4, 9, 11, and 12.
Opinion
The dispositive issue is whether the Examiner erred in determining
that one of ordinary skill in the art following the teachings of Terry would
have recognized that in a method for reactive sputtering, target voltage
control can be performed separately from constant voltage control and at
lower control speed than the constant voltage control, and would have
arrived at a control response time To of 0.1 seconds to 60 seconds for
performing target voltage control, as specified in independent claim 1, and at
a response time of 0.1 seconds to 60 seconds for detecting a plasma emission
spectrum and comparing the plasma emission spectrum with a target value
of the plasma emission as specified in independent claim 17. Ans. 14; App.
Br. 9-10; Reply Br. 2-3.
The plain language of claim 1 specifies a method for reactive
sputtering using a reactive sputtering apparatus that has, in a vacuum
chamber, a sputtering vaporization source with a metal target, a sputtering
power source that drives the sputtering vaporization source, and a
mechanism that introduces an inert gas and a reaction gas. The claimed
method comprises at least the steps of, among other things, performing
constant voltage control of the sputtering power source, within a control
response time Tv of 0.1 milliseconds, that is, 0.0001 seconds, to 0.1 seconds,
that is, 100 milliseconds, by (1) comparing an actual output voltage of the
Appeal 2010-010975
Application 11/515,730
4
sputtering power source to the target voltage Vs for the sputtering power
source, and (2) issuing an output voltage control command based on the
difference between the actual output voltage of the sputtering power source
and the target voltage Vs to control the voltage of the sputtering power
source to be the target voltage Vs. The claimed method comprises the
further step of performing target voltage control to set the target voltage Vs
for the sputtering vaporization source within a control response time To of
0.1 seconds, that is, 100 milliseconds, to 60 seconds, that is, 60,000
milliseconds, based on the spectrum of the plasma emission generated
forward of the sputtering vaporization source. We interpret claim 1 as
specifying that the output voltage command is issued to the sputtering power
source, and that target voltage control includes detecting the plasma
emission and generating the emission spectrum. See Spec., e.g., 17:21 to
18:1, 36:12 to 38:25, and Fig. 14.
The plain language of claim 17 similarly specifies a method for
reactive sputtering using a reactive sputtering apparatus that has, in a
vacuum chamber, a sputtering vaporization source with a metal target, a
sputtering power source, and a mechanism that introduces an inert gas and a
reaction gas. The claimed method comprises at least the steps of, among
other things, outputting a target voltage Vs, and outputting a control
command to the sputtering power source based on a comparison of detected
output voltage of the sputtering power source and the target voltage Vs.
Claim 17 specifies, with respect to outputting the target voltage Vs, that the
same is based on comparing the measured spectrum of the plasma emission
generated forward of the sputtering vaporization source with a target value
of the plasma emission, “wherein the steps of detecting the plasma emission
Appeal 2010-010975
Application 11/515,730
5
and comparing the measured spectrum together have a response time with
the range of 0.1 seconds to 60 seconds.” Claim 17 does not specify a
response time for outputting a control command to the sputtering power
source. Claim 17 further specifies the steps of detecting the output power of
the sputtering power source, comparing the same with a target measure of
the output power of the sputtering power source, and controlling a reaction
gas flow parameter based on the comparison. See Spec., e.g., 17:21 to 18:1,
36:12 to 38:25, and Fig. 14.
We find that Terry would have disclosed to one of ordinary skill in the
art a method for reactive sputtering using a reactive sputtering apparatus that
has, in a vacuum chamber, a sputtering vaporization source with a metal
target, a sputtering power source that drives the sputtering vaporization
source, and a mechanism that introduces an inert gas and a reaction gas.
Terry, e.g., cols. 5-6 and Fig. 2. Terry discloses an “inner” or “fast” control
loop that “controls the current at the target” which “operates at a high speed
relative to that of the outer [or “slow”] control loop that controls the gas
flow.” Terry col. 3, ll. 25-43.
Terry discloses that when “the optical emission of a titanium metal
transition” is monitored, “[a]s the [optical emission of titanium metal
transition] varies from an empirically determined setpoint, the control
system makes very small changes to the power supply current” via the inner
fast control loop. Terry col. 3, ll. 48-64. “This inner (fast) loop can be
executed very quickly, such as once every five milliseconds.” Terry col. 3,
ll. 64-66.
We find that Terry discloses that “[t]he outer (slow) control loop
monitors the average power over a longer time period, such as a period of
Appeal 2010-010975
Application 11/515,730
6
one second,” and “[i]f the average power deviates from a pre-established
power setpoint, the outer control loop changes the concentration of the
oxygen gas,” thus inducing “a change in the optical emission of titanium,
causing a response by the inner control loop.” Terry col. 4, ll. 2-8.
We find that Terry discloses that “[t]he invention uses the
responsiveness of the power supply electronic circuitry to have a fast control
while relying on the gas flow controller to respond on the order of seconds
to maintain an average constant sputter power.” Terry col. 4, ll. 53-57.
Terry discloses that “[t]he cycle time of the fast loop is preferably from one
to fifty milliseconds, and in a preferred embodiment is 25 milliseconds,”
wherein in each cycle of the inner fast control loop, the optical emission is
measured, compared with an emission setpoint using a control algorithm
which then sends an output signal to the power supply to adjust the current
and thus the optical emission. Terry col. 6, l. 65 to col. 7, l. 19, and Fig. 3.
Terry discloses that the control algorithm can be a proportion-integral-
derivative control algorithm. Terry col. 7, l. 46 to col. 8, l. 7, and Fig. 4.
Terry discloses that the outer slow control loop “maintains the power at a
desired level by measuring the current, voltage and/or power from the power
supply controller and manipulating the oxygen flow in the process to bring
the power to a desired setpoint.” Terry col. 7, ll. 22-26.
We find that Terry discloses that “[t]he inner control loop reacts to
this change in emission very quickly by increasing the power supply current
to bring the optical emission level back to the setpoint. The inner [fast] loop
completes multiple cycles for each cycle of the outer [slow] loop.” Terry
col. 7, ll. 31-34.
Appeal 2010-010975
Application 11/515,730
7
The Examiner submits that Terry’s inner fast control loop and outer
slow control loop performs “constant-voltage control to control the voltage
of the sputtering power source at a target voltage” as specified in claim 1.
Ans. 4, citing Terry abstract and col. 7, ll. 22-26. The Examiner further
relies on Terry’s inner fast control loop response time of 1 to 50
milliseconds to establish that “control response time of the constant-voltage
control is within the range of 0.1 milliseconds to 0.1 seconds” specified in
claim 1, and Terry’s outer slow control loop response time of 100 to 1000
milliseconds to establish that “the control response time of the target voltage
control is within the range of 0.1 to 60 seconds.” Ans. 5, citing Terry col. 7,
ll. 13-15 and 22-25, and patent claim 9. The Examiner also relies on Terry’s
outer slow control loop to establish that the output target voltage “response
time is in the range of 0.1 to 60 seconds” as specified in claim 17. Ans. 7,
citing Terry col. 7, ll. 22-26. The Examiner concludes that it would have
been obvious to one of ordinary skill in the art to vary the response time of
each of Terry’s inner fast and outer slow control loops,” and to “vary the
times as to reverse the inner/outer relation between the two loops, on the
basis that Terry teaches that “[t][he purpose of the control mechanisms . . . is
to stabilize the process with respect to target poisoning,” and that the
discovery of the optimum value of a result effective variable involves only
routine experimentation. Ans. 6, citing Terry col. 3, ll. 57-61.
Appellants submit that Terry’s teaching that the response time of 1 to
50 milliseconds for the inner fast control loop is less than half of the
minimum control response time of 0.1 seconds, that is, 100 milliseconds, for
the target voltage control specified in claims 1 and 17. App. Br. 9.
Appellants contend that the claimed 0.1 to 60 seconds response time for the
Appeal 2010-010975
Application 11/515,730
8
target voltage control would not have been obvious over “the 100 to 1000
millisecond response time of the ‘slow loop’ that controls the gas flow rate
in Fig. 3 of [Terry], since the slow loop in [Terry] does not include the
detection of the emission spectrum and is not a voltage control loop.” App.
Br. 9-10.
The Examiner responds that Terry’s inner fast control loop has a
response time of 1 to 50 milliseconds “with respect to information received
from the optical detector,” and “[o]ne of ordinary skill in the art would
recognize that in a process control system the characteristics of the response
are dependent upon the time of the response,” which is thus a result effective
variable. Ans. 14, citing Terry col. 7, l. 50 to col. 8, l. 7. The Examiner
concludes that the claimed control response time To of 0.1 to 60 seconds for
the target voltage control is patentably indistinct from Terry’s response time
of 0.001 to 0.05 seconds for the inner fast control loop because “the prior art
range is well within an order of magnitude of the claimed range and is a
result effective variable.” Ans. 14. Appellants reply that Terry teaches the
inner fast loop responds “very quickly” to changes in emission “in order to
complete multiple cycles for each [outer] slow [control] loop.” Reply Br.
2-3, citing Terry col. 3, ll. 64-65, and col. 7, ll. 31-34.
On this record, we agree with Appellants that the Examiner erred in
determining that Terry would have led one of ordinary skill in the art to
modify the disclosed method for reactive sputtering by uncoupling the steps
of measuring the emission spectrum and comparing the same to an emission
setpoint to set a power supply control target, from Terry’s inner fast control
loop which controls the power supply, and performing the separated target
power supply control steps in a response time slower than the response time
Appeal 2010-010975
Application 11/515,730
9
for performing the steps of adjusting the power supply. In this respect, we
agree with Appellants that Terry’s disclosure of the outer slow control loop
would not have suggested such a modification of Terry’s inner fast control
loop as the Examiner contends in the statement of the ground of rejection
based on Terry alone. Indeed, Terry’s outer slow control loop monitors only
the average power for the purposes of changing the concentration of oxygen
gas, thus interacts with the inner fast control loop only in this manner. See,
e.g., B.F. Goodrich Co. v. Aircraft Braking Sys. Corp., 72 F.3d 1577,
1582-83 (Fed. Cir. 1996) (“When obviousness is based on a particular prior
art reference, there must be a showing of a suggestion or motivation to
modify the teachings of that reference. This suggestion or motivation need
not be expressly stated.” (citation omitted)).
Furthermore, even if one of ordinary skill in the art performed the
steps of measuring the emission spectrum and comparing the same to an
emission setpoint to set a power supply control target separately from the
other steps in Terry’s inner fast control loop, that person would have been
led by Terry to perform the steps at a response time to support the disclosed
response time for the inner fast control loop of about 1 to 50 milliseconds,
which is illustrated by response times of 5 milliseconds and 25 milliseconds.
Indeed, Terry discloses that the response time of the inner fast control loop
is at high speed relative to the response time of the outer slow control loop
which Terry teaches can be a period of 1.0 seconds. On this record, one of
ordinary skill in the art would have reasonably inferred from Terry that the
workable or optimum range for the response time for performing the inner
fast control loop is within the range of 1 to 50 milliseconds. Indeed, the
Examiner has not adduced evidence or scientific reasoning establishing that
Appeal 2010-010975
Application 11/515,730
10
one of ordinary skill in the art would have reasonably sought a greater
workable or optimum range for the inner fast control loop than taught by
Terry. See, e.g., In re Sebek, 465 F.2d 904, 907 (CCPA 1972) (“Where, as
here, the prior art disclosure suggests the outer limits of the range of suitable
values, and that the optimum resides within that range, and where there are
indications elsewhere that in fact the optimum should be sought within that
range, the determination of optimum values outside that range may not be
obvious.”); cf., e.g., In re Aller, 220 F.2d 454, 456 (CCPA 1955) (“[W]here
the general conditions of a claim are disclosed in the prior art, it is not
inventive to discover the optimum or workable ranges by routine
experimentation.”).
A discussion of Nakamura, Tanaka, and Brooks is not necessary to
our decision as the Examiner does not contend that these references address
the issues we considered above. Ans. 8-13.
Accordingly, in the absence of a prima facie case of obviousness, we
reverse the grounds of rejection under 35 U.S.C. § 103(a) advanced on
appeal.
The Primary Examiner’s decision is reversed.
REVERSED

sld