Ex Parte Cross et al

Board of Patent Appeals and Interferences

Feb 28, 2012

10323272 (B.P.A.I. Feb. 28, 2012)

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.
10/323,272 12/18/2002 Elfido Cross JR. 2162.097800/TT5243 7506
10742 7590 02/28/2012
GLOBALFOUNDRIES INC.
c/o Williams, Morgan & Amerson
10333 Richmond , Suite 1100
Houston, TX 77042
EXAMINER
RIAD, AMINE
ART UNIT PAPER NUMBER
2113
MAIL DATE DELIVERY MODE
02/28/2012 PAPER
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.
PTOL-90A (Rev. 04/07)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________________

BEFORE THE BOARD OF PATENT APPEALS
AND INTERFERENCES
____________________

Ex parte ELFIDO CROSS JR., ERNEST D. ADAMS III, ROBERT J.
CHONG, HOWARD E. CASTLE, THOMAS J. SONDERMAN, and
ALEXANDER J. PASADYN
____________________

Appeal 2010-001756
Application 10/323,272
Technology Center 2100
____________________

Before JOSEPH L. DIXON, THU A. DANG, and ANDREW J. DILLON,
Administrative Patent Judges.

DANG, Administrative Patent Judge.

DECISION ON APPEAL

Appeal 2010-001756
Application 10/323,272

2
I. STATEMENT OF THE CASE
Appellants appeal under 35 U.S.C. § 134(a) from a Final Rejection of
claims 1-22. We have jurisdiction under 35 U.S.C. § 6(b).
We affirm-in-part.

A. INVENTION
Appellants’ invention relates generally to an industrial process and,
more particularly, to performing fault detection in parallel in a
semiconductor fabrication process (Spec. 2, ll. 4-5).

B. ILLUSTRATIVE CLAIM
Claim 1 is exemplary:
1. A method, comprising:

receiving first data associated with processing of a
workpiece by a first processing tool;

receiving second data associated with processing of a
workpiece by a second processing tool; and

comparing at least a portion of at least one of the first
received data and the second received data to a fault
model common to both of the processing tools to
determine if a fault associated with at least one of the
processing of the workpiece by the first processing tool
and processing of the workpiece by the second
processing tool occurred.

C. REJECTION
The prior art relied upon by the Examiner in rejecting the claims on
appeal is:
Appeal 2010-001756
Application 10/323,272

3
Coss US 6,954,883 B1 Oct. 11, 2005
(filed Jan. 11, 2002)

Claims 1-22 stand rejected under 35 U.S.C. § 102(e) as being
anticipated by Coss.

II. ISSUE
The dispositive issue before us is whether the Examiner has erred in
finding that Coss teaches receiving “first data” from “a first processing
tool,” receiving “second data” from “a second processing tool” and
“comparing at least a portion of at least one of the first received data and the
second received data to a fault model common to both the processing tools to
determine if a fault associated with at least one of the processing of the
workpiece … occurred” (claim 1, emphasis added). In particular, the issue
turns on whether Coss teaches comparing a portion of data from at least one
of two processing tools with a model common to both processing tools to
determine if a fault associated with the processing has occurred.

III. FINDINGS OF FACT
The following Findings of Fact (FF) are shown by a preponderance of
the evidence.
Coss
1. Coss discloses performing fault detection using real-time or near real-
time data from a database (Abstract).
2. Coss’s Figure 6 is reproduced below:
App
App

Cos
sign
and
on t
and
met
proc
ana
inte
data
faul
data
eal 2010-0
lication 10
s’s Figure
als on line
processed
he line 62
processed
rology dat
essing too
lysis unit 6
grated met
base 320 f
t detection
from the
01756
/323,272
6 disclose
623 that a
by the fir
3 that are i
by the sec
a are colle
ls 610a an
60 sends m
rology too
or storage
unit 670 f
database 3
s a compu
re intende
st machine
ntended fo
ond mach
cted by the
d 610b (co
etrology
l 310 and
; and data
or perform
20 (col. 7,
4
ter system
d for proc
interface
r processi
ine interfa
integrate
l. 7, ll. 49
data and o
the offline
from the d
ing fault
l. 62 to co
630 wher
essing too
615a and c
ng tool B
ce 615b (c
d metrolog
-51); the m
ffline met
metrolog
atabase 32
detection a
l. 8, l. 8).
ein contro
l A 610a a
ontrol inp
610b are re
ol. 7, ll. 32
y tool 310
etrology
rology dat
y tool 650
0 are extr
nalysis ba

l input
re receive
ut signals
ceived
-38);
within th
data
a from the
to the
acted by a
sed upon
d
e

Appeal 2010-001756
Application 10/323,272

5

IV. ANALYSIS
Claims 1, 5, 6, and 15-17
Appellants contend that Coss does not teach “the claimed feature of
comparing data received from processing tool(s) to a common fault model
(that is, a fault model common to the first processing tool and second
processing tool)” as required by claim 1 (App. Br. 5). According to
Appellants, “Coss discloses determining fault detection using parameters
particular to that processing step, and not based on conditions, calculations,
and/or expectations that are common to the processing steps of both the
processing tools A & B” (App. Br. 7). In particular, though Appellants
admit that, in Coss, “the fault detection is based on values that are particular
to the processing step being performed on the wafer by processing tool A
and processing tool B,” Appellants assert that the comparison is “not based
on any values (or threshold values) common to different processing steps”
(App. Br. 6). Further, Appellants contend that Coss “does not use a common
fault model … to determine faults” (App. Br. 7), but rather “the ‘fault
detection analysis’ is performed to calculate or detect faults even before any
‘threshold value’ is ever utilized” (App. Br. 7-8).
However, the Examiner explains that, in Coss, “[t]he criterias [sic] for
the two processing tools are the same,” “the data collected at the database
level comes [from] both processing tools” and thus “it is logical to come to
the conclusion that the manufacturing model is the base of the comparison
between the data stored in the database and the recipe through item 670
(Figure 6; fault detection unit)” (Ans. 9). The Examiner finds that “the
Appeal 2010-001756
Application 10/323,272

6
recipe is common because the manufacturing parameters cannot change
from one wafer to the other” (id.).
Although Appellants contend that Coss does not teach determining
fault detection “based on conditions, calculations, and/or expectations that
are common to the processing steps of both the processing tools A & B”
(App. Br. 7), such argument is not commensurate in scope with the language
of claim 1. That is, claim 1 does not require any such “conditions,
calculations, and/or expectations that are common to the processing steps”
(id., emphasis added), as Appellants argue. Rather, claim 1 merely requires
“a fault model” that is “common to both of the processing tools” (emphasis
added).
In determining whether Coss teaches receiving “first data” from “a
first processing tool,” receiving “second data” from “a second processing
tool” and “comparing at least a portion of at least one of the first received
data and the second received data to a fault model common to both the
processing tools” as specifically recited in representative claim 1, we give
the claim its broadest reasonable interpretation consistent with the
Specification. See In re Morris, 127 F.3d 1048, 1054 (Fed. Cir. 1997).
However, we will not read limitations from the Specification into the claims.
In re Van Geuns, 988 F.2d 1181, 1184 (Fed. Cir. 1993).
Claim 1 does not place any limitation on what the term “fault model”
means, includes, or represents other than a portion of the received data is
compared to the model, wherein the model is common to both of the
processing tools. Thus, we give “fault model” its broadest but reasonable
interpretation as standard parameters common to processing tools, to which
received data is compared, as defined in the claim. Furthermore, since claim
Appeal 2010-001756
Application 10/323,272

7
1 only defines a “fault” as “associated with” the processing of the
workpiece, we give the limitation “to determine … a fault associated with …
the processing of the workpiece” its broadest but reasonable interpretation as
determining any undesirable variation from standard values related in any
way to the processing of a workpiece.
Coss discloses performing fault detection (FF 1) wherein data from
processing tools 610a and 610b are sent to the database and then extracted
by a fault detection unit for performing fault detection analysis (FF 2). We
find Coss’s data received from processing tool 610a to comprise “first data”
and data received from processing tool 610b to comprise “second data,”
wherein at least a portion of the received data is retrieved from the database
and received by the fault detection unit for fault detection analysis.
In Coss, the received data from the tools is examined by the fault
detection unit to determine whether there is fault (FF 1). We find such
examination for fault to comprise comparing the received data with such
standard parameters to determine if there is fault. That is, in order for the
fault detection unit to determine there is fault, the data must necessarily be
examined in comparison to standard parameters to determine if there is any
undesirable variation. In fact, even Appellants admit that, in Coss, “the fault
detection is based on values that are particular to the processing step being
performed on the wafer by processing tool A and processing tool B” (App.
Br. 6, emphasis added). That is, we find Coss discloses determining whether
there’s any undesirable variation from standard values particular to the
processing step. In view of our broad but reasonable claim interpretation
above, we find Coss discloses comparing at least a portion of at least one of
Appeal 2010-001756
Application 10/323,272

8
the “first received data” and the “second received data” to a “fault model” as
required by claim 1.
Since the processing step is performed by both processing tools A and
B, we find the standard values particular to the step to be “common” to the
processing tools, as required by claim 1. In fact, we find no error in the
Examiner’s finding that the criteria for the two processing tools “are the
same” and that “manufacturing parameters cannot change from one wafer to
the other” (Ans. 9). In view of our broad but reasonable interpretation
above, we find Coss to disclose a fault model “common” to both the
processing tools.
Further, since Coss’s standard values are particular to the processing
step being performed on the wafer, we find that any variation from the
standard values determined in Coss to be related to the processing step. In
view of our broad but reasonable claim interpretation, we find Coss to
disclose determining a fault “associated with” the processing of the
workpiece.
Accordingly, we agree with the Examiner that Coss discloses
“comparing at least a portion of at least one of the first received data and the
second received data to a fault model common to both of the processing
tools to determine if a fault associated with at least one of the processing of
the workpiece … occurred” as required by claim 1. In view of the record,
we find that Appellants have not shown that the Examiner erred in rejecting
representative claim 1, independent claims 15 and 16 falling therewith (App.
Br. 7), and claims 5, 6, and 17 depending respectively from claims 1 and 16
over Coss.

Appeal 2010-001756
Application 10/323,272

9
Claims 8, 9, 21, and 22
As for independent claim 8, Appellants repeat the argument that
“Coss does not teach detecting a fault using a ‘fault model’ that is
representative of at least two processing tools” and add that “Coss does not
teach that the threshold value represents an acceptable operational range”
(App. Br. 9). However, the Examiner points out that “it is not possible, for
fault detection unit to detect fault if it does not have a certain way of
knowing that there is an error or an error occurred” (Ans. 10). We agree
with the Examiner.
As discussed above, we find Coss discloses standard values
“common” to both the processing tools that are used to determine whether
there is any undesirable variation from these standard values. Thus, we find
Coss to disclose standard values that are representative of acceptable
operations by both processing tools, i.e., “at least two processing tools,” as
recited in claim 8.
Since claim 8 does not define as to what is an “acceptable operational
range,” we give the term its broadest but reasonable definition as any
operational range outside of which there is an unacceptable/undesirable
variation. That is, we agree with the Examiner that “it is not possible, for
fault detection unit to detect fault if it does not have a certain way of
knowing that there is an error or an error occurred” (Ans. 10). Thus, we find
Coss’s “fault model” comprises the operational range outside of which there
is an unacceptable variation/fault. Thus, in view of our broad but reasonable
interpretation, we find Coss discloses “a fault model representative of an
acceptable operational range of at least the first and second processing tools”
as required by claim 8.
Appeal 2010-001756
Application 10/323,272

10
Accordingly, we also find that Appellants have not shown that the
Examiner erred in rejecting claim 8 and claim 21, falling therewith, and
claims 9 and 22 depending respectively therefrom (App. Br. 9) over Coss.
Claims 2, 3, 10, 11, 18, and 19
As for claim 2, Appellants also contend that, in Coss, “there is not a
single suggestion that a plurality of fault models is used for fault detection,
much less a plurality of common fault models” (App. Br. 9). However, as
discussed above, we find Coss discloses standard values [i.e., more than one
value] “common” to both of the processing tools that are used to determine
whether there is any detraction from these standard values. That is, we find
Coss discloses a “plurality” of common standard values used for fault
detection. Thus, we see no error in the Examiner’s finding that Coss
discloses “a plurality of common fault models” are required by claim 1.
Accordingly, we also find that Appellants have not shown that the
Examiner erred in rejecting claim 2 and claim 10 and 18, falling therewith,
and claims 3, 11, and 19 depending respectively therefrom (App. Br. 9) over
Coss.
Claims 4, 7, 12-14, and 20
As for claims 4, 13, and 20, Appellants contend that Coss does not
disclose “processing of a workpiece by a third processing tool” (App. Br.
10). Appellants contend that “the clause ‘data acquired on at least one’ [of
Coss] (quoted by the Examiner) refers to acquiring data either on a real-time
or near real-time basis, and does not describe receiving data from three
different processing tools” (App. Br. 9-10). Though the Examiner explains
that “it is possible to [have] more than 3 stations” (Ans. 11), the section of
Coss cited by the Examiner merely discloses “as the wafers 105 would pass
Appeal 2010-001756
Application 10/323,272

11
through other processing stations (not shown) in the processing tool 610”
(col. 4, ll. 47-49). That is, this section of Coss merely discloses that the
wafer passes through a section of one of two (not three) stations. There is no
teaching in this section of Coss that a third station is necessarily included.
As for claim 12, Appellants contend that in Coss, “there is simply no
teaching of [a] plurality of fault models having a[] hierarchical order” (App.
Br. 12). Though the Examiner explains that “fault models according to Coss
are contained in the database which is by definition an organized body of
related information” (Ans. 6), the sections of Coss cited by the Examiner
merely disclose comparing data from the database to standard values. That
is, there is no teaching in these cited sections of Coss that the standard
values are necessarily arranged in a hierarchical order.
As for claims 7 and 14, Appellants contend “claim 7 calls for updating
the ‘fault model,’” and thus “the Examiner must now show that Coss teaches
updating this so-called ‘threshold’ in Coss’s fault detection system” (App.
Br. 13). Appellants contend that the passage cited by the Examiner “does
not even relate to fault detection analysis” (id.). Though the Examiner finds
that “Coss not only updates the fault model constituted by database 320[,]
fault detection unit 670, and manufacturing model 640, but Coss updates it
in real time” (Ans. 12), the section of Coss cited by the Examiner merely
disclose “[d]ata is acquired on at least one of a real-time basis or near
real-time basis”(col. 2, ll. 43-44). That is, this section of Coss merely
discloses that data in the database can be updated. However, the fault model
cannot be data in the database, since, as Appellants correctly contend, “the
claims call for comparing a portion of the received data to a ‘fault model’”
Appeal 2010-001756
Application 10/323,272

12
(Reply Br. 11). There is no teaching in Coss that updating the fault models
is necessarily included.
Since the Examiner has not made a clear distinction as to what
teachings of Coss comprise a “third processing tool,” “a hierarchical order”
for the fault models, or “updating the fault model based on at least a portion
of at least one of the first received data and the received second data” (claim
7), the Examiner has failed to meet the initial burden of proof required for
the rejection of claims 4, 7, 12-14, and 20 pursuant to 35 U.S.C. § 102.
Accordingly, we are constrained to reverse the Examiner’s rejection
of representative claims 4, 7, 12-14, and 20 over Coss.

V. CONCLUSION AND DECISION
The Examiner’s rejection of claims 1-3, 5, 6, 8-11, 15-19, 21, and 22
under 35 U.S.C. § 102(e) is affirmed. However, the Examiner’s rejection of
claims 4, 7, 12-14, and 20 under 35 U.S.C. § 102(e) is reversed.
No time period for taking any subsequent action in connection with
this appeal may be extended under 37 C.F.R. § 1.136(a)(1)(iv).

AFFIRMED-IN-PART

peb