Ex Parte Mao et al

Patent Trial and Appeal Board

Sep 14, 2017

13290879 (P.T.A.B. Sep. 14, 2017)

United States Patent and Trademark Office
UNITED STATES DEPARTMENT OF COMMERCE
United States Patent and Trademark Office
Address: COMMISSIONER FOR PATENTS
P.O.Box 1450
Alexandria, Virginia 22313-1450
www.uspto.gov
APPLICATION NO. FILING DATE FIRST NAMED INVENTOR ATTORNEY DOCKET NO. CONFIRMATION NO.
13/290,879 11/07/2011 Yi-Chao Mao TSMl 1-0539 8640
43859 7590 09/18/2017
SLATER MATSIL, LLP/TSMC
17950 PRESTON ROAD, SUITE 1000
DALLAS, TX 75252
EXAMINER
CARLSON, MARC
ART UNIT PAPER NUMBER
3723
NOTIFICATION DATE DELIVERY MODE
09/18/2017 ELECTRONIC
Please find below and/or attached an Office communication concerning this application or proceeding.
The time period for reply, if any, is set in the attached communication.
Notice of the Office communication was sent electronically on above-indicated "Notification Date" to the
following e-mail address(es):
docketing @ slatermatsil. com
PTOL-90A (Rev. 04/07)
UNITED STATES PATENT AND TRADEMARK OFFICE
BEFORE THE PATENT TRIAL AND APPEAL BOARD
Ex parte YI-CHAO MAO, JUI-PIN HUNG, JING-CHENG LIN,
SHIN-PUU JENG, and CHEN-HUA YU
Appeal 2016-002278
Application 13/290,8791
Technology Center 3700
Before LINDA E. HORNER, CHARLES N. GREENHUT, and
PAUL J. KORNICZKY, Administrative Patent Judges.
HORNER, Administrative Patent Judge.
DECISION ON APPEAL
STATEMENT OF THE CASE
Yi-Chao Mao et al. (Appellants) seek our review under 35 U.S.C.
§ 134(a) of the Examiner’s decision rejecting claims 7-13 and 20-28 under
35 U.S.C. § 103(a) as unpatentable over Takahashi (US 5,830,041, issued
November 3, 1998). Final Office Action (January 8, 2015) (hereinafter
1 Appellants identify Taiwan Semiconductor Manufacturing Co., Ltd. as the
real party in interest. Appeal Brief 3 (June 22, 2015) (hereinafter “Appeal
Br.”).
Appeal 2016-002278
Application 13/290,879
“Final Act.”). Claims 1-6 and 14-19 are canceled. Id. at 2. We have
jurisdiction under 35 U.S.C. § 6(b).
We affirm-in-part.
CLAIMED SUBJECT MATTER
Appellants’ claimed subject matter relates to “[a] method of detecting
end points in . . . grinding processes.” Spec. 114. Claims 7 and 20 are the
independent claims on appeal and are reproduced below.
7. A method comprising:
grinding a sample wafer using a grind wheel;
during the step of grinding the sample wafer, monitoring
wheel currents for driving the grind wheel;
inspecting the sample wafer to determine an optimal end
point of the grinding of the sample wafer;
recording one of the wheel currents corresponding to the
optimal end point as a target wheel current;
performing a grinding process on a production wafer
substantially identical to the sample wafer;
during the step of grinding the production wafer,
monitoring wheel currents for grinding the production wafer; and
stopping the grinding process after the target wheel current
is reached.
20. A method comprising:
predetermining a target wheel current;
pre-selecting an extended grinding process;
performing a grinding on a wafer;
during the grinding, monitoring a wheel current of a
grinding wheel used for grinding the wafer;
starting the pre-selected extended grinding process once
the wheel current reaches the target wheel current; and
stopping the grinding once the pre-selected extended
grinding process is finished.
Appeal Br. 20, 21 (Claims Appendix).
2
Appeal 2016-002278
Application 13/290,879
ANALYSIS
Independent Claim 7
The Examiner found that Takahashi teaches the method of claim 7
including the final step of “stopping the grinding process after the target
wheel current is reached.” Final Act. 5 (citing Takahashi, col. 6,11. 5-15).
Appellants contest this finding and assert that “in Takahashi, the
determination of the endpoint is based on the change of currents and elapsed
time, not whether a specific (target) wheel current is reached or not.”
Appeal Br. 10. Appellants further argue that “Takahashi uses the change of
current (friction) of the production wafer (not a sample wafer) to determine
the endpoint (the time for stopping the grind process).” Id. at 11 (emphasis
added) (Appellants stating that “Takahashi cares when (reference time Tl)
the wheel current becomes stable, and then stops grinding after a time
elapses from reference time Tl”).
Takahashi discloses a two-step polishing process in which, in a first
step, the uneven surface of the wafer is polished to a flat surface, and in a
second step, polishing may continue to remove additional material from the
flat surface of the wafer. Takahashi, col. 4,1. 55 - col. 5,1. 4, Figs. 2A, 3.
Takahashi discloses that the first step begins at time TO and ends when the
frictional force between the wafer and the polishing cloth becomes relatively
stable, which indicates that the wafer has been polished to a flat surface. Id.
at col. 4,11. 55-59. A change in frictional force can be measured as a change
in torque, which is indirectly measured as a change in electric current of the
electric motor turning the turntable on which the wafer is attached. Id. at
col. 5,11. 35^12.
3
Appeal 2016-002278
Application 13/290,879
Takahashi discloses that the second step begins at the end of the first
step and continues for a predetermined amount of time to an endpoint. Id. at
col. 4,1. 67 - col. 5,1. 4. The predetermined amount of time for the second
step is determined by “dividing a desired amount of material removed from
the semiconductor wafer after the flat surface is obtained, by an estimated
polishing rate.” Id. at col. 5,11. 4-7. Takahashi teaches that the estimated
polishing rate is determined based on the amount of material removed and
the polishing time of at least one semiconductor wafer which has been
polished immediately before polishing the current wafer. Id. at col. 5,11. 7-
16; see also id. at col. 6,11. 6-16 (describing estimating the polishing rate
using an average polishing rate of a plurality of wafers which have been
polished immediately beforehand). Takahashi teaches using an estimated
polishing rate because the “polishing rate generally changes with
deterioration with age of the polishing cloth, and other unknown factors.”
Id. at col. 5,1. 63 - col. 6,1. 2.
Takahashi further teaches that the time for the second step “may be in
the range of substantially zero second[s] to several minutes.” Takahashi,
col. 7,11. 21-24. As such, we understand Takahashi to disclose an
embodiment of the method in which, when the time for the second step is
zero, the wafer surface is polished until the surface is flat and then the
process ends without removing any additional material from the wafer. In
this embodiment, the method includes only the first step discussed above
because the time for the second step is zero. As such, the point at which the
monitored electric current becomes relatively stable (e.g., reference time T1
in Figure 3) becomes the endpoint of the process.
4
Appeal 2016-002278
Application 13/290,879
The question is, thus, whether the endpoint T1 disclosed in Takahashi
occurs at “the target wheel current” of claim 7. We agree with Appellants
that in Takahashi, the reference time T1 is determined based on monitoring
the wheel currents of the wafer currently being processed (e.g., the
production wafer) and is not determined based on a recorded wheel current
corresponding to an optimal end point of a sample wafer, as called for in
claim 7. For instance, Takahashi discloses that by monitoring frictional
force, i.e., wheel current, to determine when the uneven surface of the
workpiece is polished to flat, “an endpoint for the workpiece can be
determined on a real-time basis during polishing.” Takahashi, col. 2,11. 34-
39 (emphasis added); see also id. at col. 7,11. 3-7.
The Examiner explains that Takahashi uses current detectors, a signal
processing device, and a controller to accumulate historical data for the
workpieces and has learned that the current values change over the
processing period and has correlated the endpoint of a polishing process
based on these measured values. Final Act. 12. Figure 3 of Takahashi may
represent prior experiments on sample wafers that show a correlation
between changes in wheel current during polishing and an endpoint of the
process such that the reference time T1 can be defined as the point at which
the current becomes stable. Claim 7, however, recites “recording one of the
wheel currents corresponding to the optimal end point as a target wheel
current.” Appeal Br. 20 (emphasis added). Thus, the “target wheel current”
defined in claim 7 is a specific wheel current that corresponds to the optimal
end point based on grinding and inspection of a sample wafer.
5
Appeal 2016-002278
Application 13/290,879
The Examiner further finds that “targeting a current change is
inherently the same as measuring the endpoint current.” Final Act. 13. We
agree with the Examiner that Takahashi measures the wheel current
throughout the polishing process and thus, measures the current at the point
at which the process ends. However, as stated above, Takahashi is
measuring this endpoint current in real time, and not recording the endpoint
wheel current based on grinding and inspection of a sample wafer, as called
for in claim 7.
The Examiner further states that “[t]he recording of historical data
using the same combination of device, operating parameters, and workpieces
will make it obvious to obtain the desired endpoint current value by routine
experimentation.” Final Act. 13. The Examiner does not provide adequate
reasoning to explain why one skilled in the art would have been led to
establish a set endpoint current value, which is contrary to Takahashi’s
teaching of determining the endpoint in real-time during the polishing
process. The Examiner’s determination seems counterproductive to
Takahashi’s goal, in that, if a wafer has a different amount of unevenness to
its surface than the wafers used to determine the set endpoint current value,
then polishing to a set endpoint current value may result in not enough or too
much polishing of the workpiece.
For these reasons, we do not sustain the Examiner’s rejection of claim
7, and dependent claims 8-13 and 27, as unpatentable over Takahashi.
Independent Claim 20
Appellants argue that independent claim 20 is allowable over
Takahashi for the same reasons presented above in response to the rejection
6
Appeal 2016-002278
Application 13/290,879
of independent claim 7. Appeal Br. 18. Unlike claim 7, however, claim 20
does not define “target wheel current” with reference to “one of the wheel
currents corresponding to the optimal end point” determined by grinding and
inspecting a sample wafer. Rather, claim 20 simply calls for
“predetermining a target wheel current.”2 As explained above, in Takahashi,
the target wheel current is the point at which the current becomes stable (i.e.,
the current corresponding to reference time Tl). Takahashi, col. 4,1. 55 -
col. 5,1. 4, Figs. 2A, 3. Takahashi also pre-selects an extended grinding
process (i.e., the pre-selected time period for the second step of Takahashi’s
process). Id. at col. 4,1. 67 - col. 5,1. 7. Takahashi performs grinding on a
wafer and during the grinding, monitors a wheel current and starts the
second step once the wheel current reaches the target wheel current (i.e., the
end of step one), and stops the grinding once the predetermined time for the
second step expires. Id.
Appellants argue that “Takahashi does not know when the grinding
will start to stabilize (reference time Tl), and does not know what the wheel
current will be at reference time Tl. Accordingly, Takahashi does not have
any pre-determined target wheel current.” Reply Br. 5. This argument is
based on the wording of claim 7, which defines the target wheel current as a
specific value. This argument is, however, not commensurate with the
language of claim 20, which we do not read as requiring a specific target
wheel current value.
2 Dependent claim 26 further limits the method of claim 20 to define the step
of “predetermining the target wheel current” to comprise steps similar to
those of claim 7. Appeal Br. 22-23 (Claims Appendix).
7
Appeal 2016-002278
Application 13/290,879
Appellants contend that the Examiner’s reading of “target wheel
current” to encompass that of a difference of two measured current values
over time is “against the commonly accepted mean[ing] of these terms.”
Reply Br. 6. Appellants fail, however, to provide a definition of these terms
that would distinguish them from the Examiner’s interpretation or
demonstrate that the Examiner’s interpretation is unreasonably broad.
Takahashi’s reference time T1 is based on when the wheel current stabilizes,
at which point, the measured wheel currents over a time increment should be
roughly the same. Thus, the stabilized wheel current is the target wheel
current. See Ans. 45 (“Once that [stabilized] current value is measured or
the stability of current measurement is identified, or a difference in current is
measured to be a certain amount between measurement periods, the
operation enters into the pre-selected grinding period.”) In other words,
Takahashi teaches that whatever the actual wheel current is at the point in
time at which the current stabilizes, that wheel current is the target wheel
current that triggers the start of the extended grinding process.
Appellants’ Specification describes:
In general, in accordance with embodiments, a target wheel
loading corresponding to the optimal end point may be found, for
example, by performing a grinding process on a sample wafer
and find a target wheel loading (such as the 10 amps current in
Figure 4). On the production wafers having the same structure
as the sample wafer, the optimal end point may be when the
target wheel loading is reached.
Spec. 121. The target wheel loading can be defined by reference to current.
Appellants chose in their embodiment to define the predetermined target
wheel current statically by selecting a specific target current value and
assuming that other similar wafers could be processed in the same manner as
8
Appeal 2016-002278
Application 13/290,879
the sample wafer to achieve the desired grinding results. Takahashi chose to
select a dynamic pre-determined target wheel current that is based on the
point at which the wheel current stabilizes for each wafer being processed.
We decline to limit the “predetermined” target wheel current to the specific
embodiment disclosed in Appellants’ Specification, because claim 20 does
not define the target wheel current as being predetermined based on a static
current value discerned from testing a sample wafer. As such, we find no
error in the Examiner’s determination that Takahashi’s teaching of
predetermining a dynamic target wheel current meets the limitations of
claim 20.
For these reasons, we sustain the Examiner’s rejection of independent
claim 20, and its dependent claims 22-25 and 28, as unpatentable over
Takahashi. We do not, however, sustain the rejection of dependent claim
26, for the same reasons discussed supra in our analysis of claim 7.
DECISION
The decision of the Examiner rejecting claims 7-13, 26, and 27 is
reversed.
The decision of the Examiner rejecting claims 20-25 and 28 is
affirmed.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a). See 37 C.F.R.
§ 1.136(a)(l)(iv).
AFFIRMED-IN-PART
9