Ex Parte Srinivasan et al

Patent Trial and Appeal Board

Feb 22, 2017

12469441 (P.T.A.B. Feb. 22, 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.
2950.118US01 1909
EXAMINER
BARTON, JEFFREY THOMAS
ART UNIT PAPER NUMBER
1757
MAIL DATE DELIVERY MODE
12/469,441 05/20/2009
62274 7590 02/22/2017
DARDI & HERBERT, PLLC
Moore Lake Plaza, Suite 205
1250 East Moore Lake Drive
Fridley, MN 55432
Uma Srinivasan
02/22/2017 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 PATENT TRIAL AND APPEAL BOARD
Ex parte UMA SRINIVASAN, XIN ZHOU,
HENRY HIESLMAIR, and NEERAJ PAKALA
Appeal 2015-006350
Application 12/469,441
Technology Center 1700
Before ROMULO H. DELMENDO, JAMES C. HOUSEL, and
CHRISTOPHER L. OGDEN, Administrative Patent Judges.
HOUSEL, Administrative Patent Judge.
DECISION ON APPEAL1
Pursuant to 35 U.S.C. § 134(a), Appellants2 appeal from the
Examiner’s decision finally rejecting claims 1, 3—12, 17—19, 22—24, and 26.
We have jurisdiction over the appeal under 35 U.S.C. § 6(b).
1 Our decision refers to the Specification (Spec.) filed May 20, 2009,
Appellants’ Appeal Brief (Appeal Br.) filed November 19, 2014, the
Examiner’s Answer (Ans.) mailed April 16, 2015, and Appellants’ Reply
Brief (Reply Br.) filed June 11, 2015.
2 According to Appellants, the real party in interest is NanoGram
Corporation, a wholly owned subsidiary of Teijin Limited of Japan. Appeal
Br. 3.
Appeal 2015-006350
Application 12/469,441
We AFFIRM-IN-PART.
STATEMENT OF THE CASE
The invention relates to a photovoltaic cell having an n-doped domain
and a p-doped domain at the same level along a surface of a semiconductor
layer. Spec. 1:30—2:1. Appellants disclose that the doped domains may
each have an average depth from about 100 nm to about 5 pm, and an edge-
to-edge spacing at one or more locations from about 10 pm to about 500 pm.
Id. at 2:1—4. Moreover, the doped domains may have a planar extent along
the surface in the shape of a stripe having a length to width ratio at least
about a factor of 10. Id. at 2:7—9. The photovoltaic cell may also have a
dielectric layer over the doped domains and a plurality of patterned metal
interconnects, wherein the dielectric layer can include windows exposing
about 5—80% of the doped domains and the metal interconnects can cover
the windows with an area at least about 20% greater than the window area.
Id. at 2:16—20. In addition, the top 10 % of the doped domain thickness can
have an average dopant concentration that is at least a factor of 4 greater
than that of the doped domain thickness at the level of 20—30% from the top.
Id. at 25:16-21.
Claim 1, reproduced below from the Claims Appendix of the Appeal
Brief, is illustrative of the subject matter on appeal.
1. A photovoltaic cell comprising a semiconductor layer, a first
metal current collector, a second metal current collector, an n-
doped domain and an adjacent p-doped domain along a surface
of the semiconductor layer at the same level as each other;
wherein the doped domains each have a pattern along the
surface with an average depth below the surface from about 100
nm to about 5 microns and an edge-to-edge spacing between the
n-doped domain and the p-doped domain has a value at one or
2
Appeal 2015-006350
Application 12/469,441
more locations from about 15 microns to about 500 microns;
wherein the doped domains have an average dopant
concentration at the top 10% of the thickness that is at least a
factor of four greater than the average dopant concentration at
the position 20-30% of the contact depth from the top of the
contact;
wherein the first metal current collector is aligned along
the n-doped domain pattern and is in contact with the n-doped
domain;
wherein the second metal current collector is aligned
along the p-doped domain pattern and is in contact with the p-
doped domain;
wherein the semiconductor layer has an average
thickness from about 20 microns to about 300 microns;
wherein the semiconductor layer comprises elemental
silicon, elemental germanium, silicon-germanium alloys or
blends thereof.
The Rejections
The Examiner maintains, and Appellants request review of, the
following grounds of rejection:
1) Claim 26 under 35U.S.C. § 112, first paragraph (pre-AIA), as
failing to comply with the written description requirement;
2) Claims 1, 4—9, and 17 under 35 U.S.C. § 103(a) as unpatentable
over Lammert3 in view of Chiang4 and Horzel;5
3) Claims 3 and 7 under 35 U.S.C. § 103(a) as unpatentable over
Lammert in view of Chiang and Horzel, further in view of
Nagashima;6
3 Michael D. Lammert and Richard J. Schwartz, The Interdigitated Back
Contact Solar Cell: A Silicon Solar Cell for Use in Concentrated Sunlight,
24 IEEE Transactions on Electron Devices 4, 337-342 (April 1977).
4 US 4,133,698, issued January 9, 1979.
5 US 6,552,414 Bl, issued April 22, 2003.
6 US 6,927,417 B2, issued August 9, 2005.
3
Appeal 2015-006350
Application 12/469,441
4) Claims 18 and 22 under 35 U.S.C. § 103(a) as unpatentable
over Lammert in view of Chiang and Horzel, further in view of
Husher;7
5) Claim 19 under 35 U.S.C. § 103(a) as unpatentable over
Lammert in view of Chiang and Horzel, further in view of
Korevaar;8 and
6) Claims 10-12, 23, 24, and 26 under 35 U.S.C. § 103(a) as
unpatentable over Lammert in view of Husher, Chiang, and
Kaschmitter.9
ANALYSIS
Rejection 1: Written Description under 35 U.S.C. § 112, first paragraph
We affirm the stated written description rejection for substantially the
fact findings and the reasons set forth by the Examiner in the non-Final
Office Action and in the Examiner’s Answer. We offer the following for
emphasis only.
We first note that claim 26 was added by an Amendment filed on
October 24, 2013. The Examiner finds the recitation, “wherein the p-doped
domain has a sheet resistance of less than about 50 Ohms per square,”10 of
7 US 2004/0025932 Al, published February 12, 2004.
8 US 2008/0173347 Al, published July 24, 2008.
9 US 5,538,564, issued July 23, 1996.
10 As Appellants correctly note, claim 26 actually recites “wherein the p-
doped domain has a sheet resistance from about 10 Ohms per square to
about 50 Ohms per square.” Appeal Br. 10—11; Ans. 28 (emphasis omitted).
Regardless, the limitation that the Examiner finds lacks adequate written
descriptive support, i.e., “about 50 Ohms per square,” is present in claim 26.
As such, the Examiner’s error is irrelevant to the rejection and, therefore, is
harmless.
4
Appeal 2015-006350
Application 12/469,441
claim 26 lacks adequate written descriptive support with regard to the
reference to “about 50 Ohms per square” as an end point for the range of
sheet resistances. Ans. 3. In particular, the Examiner finds that no data is
provided at or about 50 ohms per square, particularly in Appellants’ Figure 9
or Example 1 (Spec. 48:25—51:7). Ans. 3.
Appellants contend that the Examiner fails to meet the burden of
showing the claimed invention is not described in the Specification. Appeal
Br. 10. Appellants assert that “there is certainly data of the device described
in the specification and the drawing consistent with the claim language.” Id.
at 11. Appellants argue that “[t]he cited data from the specification
implicitly provides a range spanning the presently claimed range, and thus
directly indicates possession of the range to a person of ordinary skill in the
art.” Id. at 12. In particular, Appellants assert that Figure 9 shows eleven
data point, nine of which are below and two of which are above 50 Ohms
per square. Id. at 13. Appellants urge that the data points “spanning ‘about
50 ohms per square’ with systematic variation based on processing
conditions clearly demonstrates to a person of ordinary skill in the art
possession of claimed devices with the sheet resistance range.” Id.
Moreover, Appellants argue that the Examiner failed to apply the
correct legal standard for compliance with the written description
requirement: “whether the disclosure of the application as originally filed
reasonably conveys to the artisan that the inventor had possession at that
time of the later claimed subject matter.” Id. at 11. Appellants also argue
that each of In re Wertheim, 541 F.2d 257 (CCPA 1976) and In re
Eickmeyer, 602 F.2d 974 (CCPA 1979) supports Appellants’ position. Id. at
12-13.
5
Appeal 2015-006350
Application 12/469,441
Appellants’ arguments are not persuasive of reversible error. As the
Examiner finds (Ans. 29), Figure 9 depicts three different data sets of
measured sheet resistances for a p-doped region after laser doping, each for a
different laser frequency (250 kHz, 375 kHz, and 500 kHz). At 500 kHz, all
three measurements are at or below 25 Ohms/square; at 375 kHz, three of
four measurements are at or below 25 Ohms/square (and one is between 75
and 100 Ohms/square); at 250 kHz, two measurements are below 25
Ohms/square, one is between 25 and 50 Ohms/square, and one is between 50
and 75 Ohms/square. Appellants present no data point at or near 50
Ohms/square, nor do Appellants identify 50 Ohms/square as a desirable
threshold in either the drawings or the Specification. Indeed, 50
Ohms/square appears to be an arbitrary end point for Appellants recited
range.11 The only basis of support Appellants proffer is that some data
points are below this end point. That fact however is true of any arbitrary
end point above 10 Ohms/square (the lower recited end point). Possession
of the end point of a claimed range must have more written descriptive
support than Figure 9 provides for 50 Ohms/square. Purdue Pharma L.P. v.
FauldingInc., 230 F.3d 1320, 1326 (Fed. Cir. 2000) (holding that district
court did not err in finding lack of written description, where examples in
11 Wertheim, 541 F.2d at 262 (“it is our considered opinion that the board
was correct in saying:
Not having been specifically named or mentioned in any manner, one
is left to selection from the myriads of possibilities encompassed by
the broad disclosure, with no guide indicating or directing that this
particular selection should be made rather than any of the many others
which could also be made.)
6
Appeal 2015-006350
Application 12/469,441
the specification indicated values of a ratio above and below a claim limit,
and there was nothing in the specification indicating which examples
embody the invention and which do not).
In addition, Appellants’ reliance on Wertheim and Eickmeyer is
misplaced. The Wertheim case centered, in relevant part, on the support for
solids content ranges of “at least 35%” in claim 1 and “between 35% and
60%” in claims 2 and 4, which the PTO held were not entitled to the benefit
of the appellants’ Swiss priority date. Wertheim, 541 F.2d at 260. However,
the Swiss priority application disclosed the range “25 to 60% solid matter,”
as well as specific examples “having solids contents of 36% and 50%.” Id.
at 262. Based on this description, the Wertheim court held that the solids
content range of “at least 35%” was not support by the Swiss priority
application because Appellants had not shown that this range was inherently
limited by the disclosed upper limit of 60%. Id. at 263—264. However,
because of the disclosed examples at 36% and 50%, and the disclosed range
of 25—60%, the Wertheim court held that the solids content range of 35—60%
was supported therein. Id. at 264—265. The Eickmeyer court, similarly, held
that where an application originally disclosed examples at 56°C and 80°C,
along with knowledge in the art that comparable processes could be operated
above 80°C, a range of at least 56°C was supported. Unlike Wertheim and
Eickmeyer, Appellants have not disclosed examples of sheet resistances near
50 Q/square or any range of desired or acceptable sheet resistances.
Accordingly, we sustain the Examiner’s rejection of claim 26 for
failing to comply with the written description rejection.
7
Appeal 2015-006350
Application 12/469,441
Rejection 2: Obviousness over Lammert in view of Chiang and Horzel
The Examiner finds Lammert teaches a back contact photovoltaic cell
substantially as recited in claim 1, except for the average depth of the doped
domains below the back surface of the cell wherein the doped domains have
an average dopant concentration at the top 10% of domain thickness that is a
factor of at least four greater than the average dopant concentration at a
position 20—30% of the contact depth from the top of the contact. Ans. 4—6.
With regard to the first missing feature, the Examiner relies on Chiang for
teaching an average depth of the doped domain of 300 nm. Id. at 6. In view
of Chiang’s teaching, the Examiner concludes it would have been obvious to
have modified Lammert’s structure to provide doped domain depth of 300
nm. Id.
With regard to the second missing feature, the Examiner relies on
Horzel for teaching a dopant gradient made by “increasing the surface
concentration of the dopant in the area where the metal contacts are formed
facilitates carrier transport towards these collecting contacts.” Id. The
Examiner also finds Horzel teaches that the higher the surface dopant
concentration, the lower the electrical resistance. Id. at 7. Therefore, the
Examiner finds “Horzel discloses that the concentration profile of the dopant
from the metal contact into the solar cell is a results effective variable,” and
concludes that it would have been obvious to modify Lammert to include a
dopant concentration gradient by increasing dopant concentration in the top
region of the doped domain. Id. Moreover, the Examiner concludes that the
ordinary artisan would have optimized this concentration gradient such that
the concentration at the top 10% is at least a factor of four greater that the
concentration at the position 20—30% of the contact depth from the top of the
8
Appeal 2015-006350
Application 12/469,441
contact “in order to facilitate carrier transport towards the collecting contacts
as well as to reduce electrical resistance in the system, absent a showing of
criticality or unexpected result.” Id.
Appellants argue, inter alia, that the Examiner’s interpretation of
Horzel’s teaching is erroneous because Horzel teaches dopant concentration
varies laterally between weakly doped regions 15 and deeply diffused
regions 12, rather than varying in the depth of the domain as required by
claim 1. Appeal Br. 14—15. In addition, Appellants argue that Horzel fails
to teach that dopant profile is a result-effective variable, nor provide any
teaching about a desired dopant profile for a back contact solar cell. Id. at
16.
Appellants’ argument is persuasive of reversible error. Although the
Examiner does not disagree that Horzel describes a concentration gradient
across the surface of the device (rather than depth of the device), the
Examiner nonetheless notes that Horzel’s teaching was not bodily
incorporated into Lammert. Ans. 34. Instead, the Examiner indicates that
“Horzel was merely relied on to provide motivation for forming the contact
region of Lammert to have a concentration gradient.” Id. However, the
Examiner’s position is not supported by a preponderance of the evidence.
The Examiner does not direct our attention to any reference teaching or
other evidence of record that Horzel’s teaching of a lateral dopant
concentration gradient would suggest a depth dopant concentration gradient
to one of ordinary skill in the art. Nor has the Examiner directed our
attention to any teaching supporting that such a gradient would have been
optimized to create an average dopant concentration at the top 10% of the
domain that is at least four times greater than that at a position 20—30% from
9
Appeal 2015-006350
Application 12/469,441
the top. “[Rejections on obviousness grounds cannot be sustained by mere
conclusory statements; instead, there must be some articulated reasoning
with some rational underpinning to support the legal conclusion of
obviousness.” In re Kahn, 441 F.3d 977, 988 (Fed. Cir. 2006) quoted with
approval in KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398, 418 (2007).
Accordingly, we do not sustain the Examiner’s obviousness rejection
of claim 1. We likewise do not sustain this rejection as to claims 4—9 and 17
which depend directly or indirectly from claim 1.
Rejections 3—5: Obviousness over Lammert in view of Chiang and Horzel,
further in view of Nagashima, Husher, or Korevaar
These rejections apply to dependent claims 3,7, 18, 19, and 22, which
depend directly or indirectly from claim 1. The Examiner does not rely on
any of the additional references, Nagashima, Husher, or Korevaar, to remedy
the deficiency in the combination of Lammert in view of Chiang and Horzel
discussed above. Therefore, we do not sustain these rejections for the same
reasons given above.
Rejection 6: Obviousness over Lammert in view of Husher, Chiang, and
Kaschmitter
After review of the opposing positions articulated by Appellants and
the Examiner, the applied prior art, and Appellants’ claims and Specification
disclosures, we determine that the Appellants’ arguments are insufficient to
identify reversible error in the Examiner’s obviousness rejections. In re
Jung, 637 F.3d 1356, 1365 (Fed. Cir. 2011). Accordingly, we affirm the
stated obviousness rejections for substantially the fact findings and the
10
Appeal 2015-006350
Application 12/469,441
reasons set forth by the Examiner in the Examiner’s Answer. We offer the
following for emphasis only.
This rejection applies to remaining independent claim 10 and its
dependent claims. Only dependent claim 26 is separately argued.
Consistent with 37 C.F.R. § 41.37(c)(iv) (2014), we select claim 10 to
address Appellants’ arguments generally, and will discuss Appellants’
arguments regarding claim 26 separately. The remaining dependent claims
11, 12, 23, and 24 stand or fall with claim 10.
Claim 10 is reproduced below from the Claims Appendix to the
Appeal Brief.
10. A photovoltaic cell comprising a semiconductor layer, an n-doped
domain and an adjacent p-doped domain along a surface of the
semiconductor layer and at the same level as each other, wherein the
doped domains each have a planar extent along the surface comprising
a stripe having a ratio of the average length that is at least about a
factor of 10 greater than the average width, a dielectric layer over the
doped domains and a plurality of patterned metal interconnects
contacting the doped domains;
wherein the dielectric layer has a thickness from about 10 nm to
800 nm and comprises a plurality of windows that expose a total of
from about 10 percent to about 80 percent of each of the doped
domains;
wherein each of the metal interconnects over the windows have
an area at least about 20 percent greater than the area of the windows
and connect through a plurality of windows to the corresponding
doped domain; and
wherein the n-doped domain and p-doped domain are separated
by 40 microns to about 500 microns and have an average depth from
about 100 nm to about 2.5 microns.
The Examiner finds Lammert teaches a back contact solar cell
substantially as recited in claim 10, except for the dimensions of the metal
11
Appeal 2015-006350
Application 12/469,441
interconnects (aluminum contacts) and the dielectric layer (Si02), and the
average depth of the doped domains. Ans. 13—18. With regard to the
average depth of the doped domains, the Examiner relies on Chiang for
teaching an average depth of the doped domains of 300 nm for a back
contact solar cell. Id. at 18. In view of Chiang’s teaching, the Examiner
concludes it would have been obvious to have modified Lammerf s structure
to provide doped domain depth of 300 nm. Id.
With regard to the thickness of the dielectric layer, the Examiner
relies on Kaschmitter for teaching a back contact solar cell with a dielectric
(SiCE) layer covering the back surface of the cell with electrical contacts
penetrating through this layer to contact highly doped domains, wherein the
o
suitable thickness of this layer is from 500 to 1000 A (50—100 nm). Id. In
view of Kaschmitter’s teaching, the Examiner concludes it would have been
obvious to have modified Lammerf s structure to provide a dielectric layer
thickness of from 50—100 nm. Id. at 19. Alternatively, the Examiner
determines that, absent a showing of criticality or unexpected results, it
would have been obvious to optimize the thickness of Lammerf s dielectric
layer within the range of 10—800 nm in order to provide sufficient
passivating protection to prevent shorts and maximize the efficiency of the
solar cell. Id.
With regard to the window and contact areas, the Examiner relies on
Husher for teaching a back contact solar cell having metal interconnects
(contacts). Id. at 15. The Examiner also finds Husher teaches that a portion
of the interconnect penetrates through a dielectric (SiCE) layer in order to
make electrical contact with n-doped and p-doped domains (regions) and
another portion of the interconnect overhangs the dielectric layer, similar to
12
Appeal 2015-006350
Application 12/469,441
Lammert’s structure shown in Figure 1. Id. In addition, the Examiner finds
Husher teaches that the width of the penetrating portion should be widened
to lower metal sheet resistance and improve efficiency, but that too much
width reduces solar cell density. Id. As such, the Examiner finds Husher
teaches that the contact width (and thereby the window area) should be
optimized for sheet resistance for a given doped domain depth. Id.
Therefore, the Examiner concludes it would have been obvious to modify
Lammerf s structure by optimizing the contact width (window area) in order
to improve the cell efficiency for a given doped domain depth. Id. at 15—16.
Moreover, the Examiner finds Husher teaches providing additional
metal on the interconnects above the dielectric layer in order to lower the
sheet resistance. Id. at 16. The Examiner concludes it would have been
obvious to modify Lammerf s structure by optimizing the interconnect width
overhanging the dielectric layer in order to lower sheet resistance while
balancing the area needed to provide another junction. Id.
With regard to the spacing between the n-doped and p-doped
domains, the Examiner finds Lammert teaches that the widths of these
domains (and ratio of the widths) should be optimized in order to maximize
device efficiency. Id. at 17. Additionally, the Examiner finds Husher
teaches that the spacing between adjacent doped domains needs to account
for the depth of these domains and the depth of the depletion region in order
to prevent adjacent depletion regions from making contact. Id. Therefore,
the Examiner concludes it would have been obvious to modify Lammerf s
structure by optimizing the domain widths and the spacing between them in
order to maximize device efficiency, while separating the depletion regions.
Id.
13
Appeal 2015-006350
Application 12/469,441
Appellants contend that the Examiner’s proposed combination of
Lammert, Husher, Chiang, and Kaschmitter fails to teach or suggest each
feature of claim 10 and that these references fail to support their
combination. Appeal Br. 19. With regard to Chiang, Appellants assert that
Chiang fails to teach a dielectric layer. Id. Appellants also argue that
Chiang only teaches a dopant depth for the n-type domain, and “does not
seem to teach a dopant depth for the [p]-type region.” Id. at 20. Regarding
Husher, Appellants argue that all of Husher’s doped region is exposed
regardless of the size of the opening through the patterning material
(dielectric layer). Id. at 21. Appellants assert that Husher is silent regarding
the concept of adjusting window size. Id.
Appellants’ arguments regarding Chiang and Husher are not
persuasive of reversible error in the Examiner’s rejection. Each reference
cited by the Examiner must be read, not in isolation, but for what it fairly
teaches in combination with the prior art as a whole. See In re Merck & Co.,
800 F.2d 1091, 1097 (Fed. Cir. 1986). While Appellants are correct that
Chiang only specifically teaches the depth of the n-doped domain (compare
Ans. 32 with Appeal Br. 14 and 20), the Examiner finds Chiang discloses
that the n-doped and p-doped domains have the same depth. Ans. 32, citing
Chiang, Fig. 1. Appellants do not challenge this finding with any
particularity. Further, while the measurements of features in a drawing have
little value when a reference does not disclose drawings as drawn to scale,12
drawings may evaluated for what they reasonably disclose and suggest to
12 Hockerson-Halberstadt, Inc. v. Avia Group Inti, 222 F.3d 951, 956 (Fed.
Cir. 2000).
14
Appeal 2015-006350
Application 12/469,441
one of ordinary skill in the art. In re Aslanian, 590 F.2d 911, 914 (CCPA
1979). The fact that Chiang is silent as to the depth of the p-doped domain,
other than that shown in the drawings, supports the conclusion that the depth
of this domain was within the ordinary skill in the art. Given that Chiang
depicts both the n-doped and p-doped domains in the drawings as having
approximately the same depth, one of ordinary skill in the art reasonably
would have expected both domains to have approximately the same depth,
e.g., 300 nm. Appellants do not direct our attention to any disclosure in
Chiang, or any other evidence of record, leading to a different conclusion.
Turning to Husher, we note Appellants do not address the Examiner’s
findings that Husher teaches that the contact width (and thereby the window
area) should be optimized for sheet resistance for a given doped domain
depth and that the amount of metal on the interconnects above the dielectric
layer should be optimized in order to lower the sheet resistance. Although
Appellants argue that Husher teaches a structure inconsistent with Lammert,
the Examiner’s findings provide sufficient support for suggesting that the
window area and contact area in Lammert should be optimized to lower
sheet resistance and improve device efficiency. In addition, we note that the
range of window area recited in claim 10, from about 10% to about 80% of
the doped domains, covers a substantial amount of the range of possible
window areas. As such, although Husher does not specify a range over
which the window area should be optimized, the ordinary artisan reasonably
would expect routine optimization to occur within the recited range.
Appellants next criticize Lammert’s Figure 1, arguing that the device
depicted therein is clearly not to scale. Appeal Br. 20. However, as
Appellants concede {id.), Lammert’s Figure 1 shows structural relationships.
15
Appeal 2015-006350
Application 12/469,441
These relationships include window openings through the dielectric layer
permitting metal interconnects to penetrate to contact the doped domains. In
addition, the metal interconnect is clearly depicted as having an overhang on
the surface of the dielectric layer that has a greater area than that of the
windows. While specific dimensional relationships may not be depicted, the
Examiner has established a reasonable basis for expecting that such
dimensional relationships are known result-effective variables that should be
optimized for lower sheet resistance and improved device efficiency. As the
Examiner correctly observed, absent a showing of criticality or unexpected
results, such optimization would have been within the ordinary skill in the
art. The discovery of an optimum value of a variable in a known process is
normally obvious. In re Aller, 220 F.2d 454 (CCPA 1955). Exceptions to
this rule include (1) the results of optimizing a variable were unexpectedly
good and (2) the parameter optimized was not recognized in the prior art as
one which would affect the results. In re Antonie, 559 F.2d 618, 620 (CCPA
1977). Appellants do not direct our attention to any evidence that the results
of optimizing the window and contact areas were unexpectedly good, nor
have Appellants identified reversible error in the Examiner’s determination
that these parameters were recognized in the prior art as result-effective
variables.
Appellants next argue that “while Kaschmitter has doped regions on
the back surface of their cell, this cell has a distinct structure owing to the
use of amorphous silicon/microcrystalline silicon substrate.” Appeal Br. 21.
Appellants urge that Kaschmitter’s active layer has a thickness of 0.1—10 pm
with doped contacts extending at least half-way therethrough, as opposed to
Lammerf s 100 pm thick bulk layer with relatively thin doped domains. Id.
16
Appeal 2015-006350
Application 12/469,441
at 21—22. In addition, Appellants argue that the Examiner fails to provide
any motivation or reason to combine the teachings of Lammert and
Kaschmitter, especially given the significant differences between these
references. Id. at 22.
However, these arguments do not address the Examiner’s rejection, in
particular, that Kaschmitter is relied on solely as a suggestion for the
thickness of the dielectric layer on the back surface of a back contact solar
cell, and that given Lammert’s silence on the dielectric layer thickness, one
of ordinary skill in the art would have expected Kaschmitter’s suggested
thickness to have been a suitable thickness for the dielectric layer.
Appellants neither identify reversible error in the Examiner’s stated position
nor show that the recited dielectric layer thickness is critical or provides
unexpected results.
Finally, Appellants broadly argue that Lammert and Kaschmitter are
non-analogous prior art and, given the respective ages of these references,
their combination is only based on hindsight using Appellants’ own
invention as a guide. Appeal Br. 23. We do not find these arguments
persuasive. Appellants’ arguments merely direct attention to the structural
differences between the devices disclosed by these references. Such
arguments do not identity reversible error in the Examiner’s findings that
these references are from the same field of endeavor. Indeed, despite the
structural differences highlighted by Appellants, it is clear that each of the
applied references is from the same field of endeavor—solar cells, more
particularly, back contact solar cells. Moreover, Appellants’ assertion that
the fact that Lammert published more than 30 years ago and none of the
references advance the state of the art in the ensuing 30 years does not
17
Appeal 2015-006350
Application 12/469,441
support the conclusion that the claimed structure is an unexpected result. A
showing of unexpected results requires evidentiary support, generally
directed to the performance of the device being claimed as compared to the
closest prior art. Appellants’ argument does not come close to meeting this
burden. See In re De Blauwe, 736 F.2d 699, 705 (Fed. Cir. 1984)
(unexpected results must be established by factual evidence).
We turn now to claim 26, which depends from claim 10 and further
requires that the p-doped domain has a sheet resistance from about 10
Q/square to about 50 Q/square. The Examiner finds that though Lammert
does not disclose the sheet resistance of the p-doped domain specifically,
Lammert does disclose that the amount of doping is a result-effective
variable affecting the device efficiency. Ans. 20 and 48. The Examiner also
finds that sheet resistance is dependent on dopant concentration, decreasing
with increasing concentration. Id. As such, the Examiner concludes it
would have been obvious to optimize the dopant concentration to improve
device efficiency which would simultaneously reduce sheet resistance. Id.
Appellants argue that the Examiner erred because Lammert does not
disclose values including those of the claims. Appeal Br. 24—25, citing In re
Applied Materials, Inc., 692 F.3d 1289, 1295 (Fed. Cir. 2012) and In re
Sebek, 465 F.2d 904, 907 (CCPA 1972). Appellants argue that Lammert
does not teach the recited range of sheet resistance nor how to process solar
cell structures to obtain these resistances. Appeal Br. 25. Appellants urge
that Horzel’s teaching of sheet resistances no less than 80 Q supports
Appellants’ view that sheet resistances in the range recited in claim 26
would not have been obvious. Id. As such, Appellants contend that the
Examiner has not established that Lammert suggests any appropriate range
18
Appeal 2015-006350
Application 12/469,441
to optimize nor any technique that would be appropriate to achieve the
claimed range. Reply Br. 8.
We first note that Horzel’s teaching regarding resistance of no less
than 80 Q is not persuasive that Appellants’ recited range would not have
been obvious. Horzel’s teaching is limited to the local spreading resistance
on the surface whereas Appellants recite a sheet resistance, which appears to
be referring to a different metric. Further, Horzel was not applied in the
rejection and there is no indication in Horzel that lower resistances could not
be obtained. The data reported in Horzel, Figure 12, merely represents an
example of Horzel’s device and does not limit the teaching of Horzel with
regard to achieving reduced resistances at higher dopant concentrations
(Horzel 7:35—36 (“The higher the dopant concentration the lower the
electrical resistance.”)).
In addition, though Lammert does not teach a range of sheet
resistances for the p-doped domain, the Examiner has established on the
record that sheet resistance is dependent on dopant concentration (as we
noted above regarding Horzel). Appellants do not dispute that this
relationship was known. Moreover, Appellants do not disclose that the laser
doping process used in fabricating the recited solar cell is new or requires
any special conditions not known in the art. Indeed, all three laser
frequencies were able to achieve sheet resistances within the recited range,
further indicating that such resistances would be obtainable by routine
optimization.
Further, although Appellants urge that the cited cases hold that, absent
a teaching in a reference of a range over which to optimize, the prior art
cannot render obvious a claimed range, we disagree. Indeed, the Sebek court
19
Appeal 2015-006350
Application 12/469,441
stated “while it may ordinarily be the case that the determination of optimum
values for the parameters of a prior art process would be at least prima facie
obvious, that conclusion depends upon what the prior art discloses with
respect to those parameters.” Sebek, 465 F.2d at 907. Unlike here, the facts
in Sebek indicated that the optimum would be outside of the claimed range
and, therefore, the claimed range was not obvious. Id. Appellants do not
direct us to any disclosure in Lammert or the other applied references that
would suggest that the optimum dopant concentration, and therefore sheet
resistance, would be outside of the range of claim 26.
Also, the Applied Materials court stated that “[i]n cases in which the
disclosure in the prior art was insufficient to find a variable result-effective,
there was essentially no disclosure of the relationship between the variable
and the result in the prior art.” Applied Materials, 692 F.3d at 1297. Here,
the Examiner has established, without dispute, that the prior art discloses the
relationship between dopant concentration and electrical resistance, and
therefore sheet resistance, of a solar cell, as well as the relationship between
resistance and device efficiency. Therefore, Appellants have not identified
reversible error in either the Examiner’s finding that sheet resistance is a
result-effective variable or the conclusion that it would have been within the
ordinary skill in the art to perform routine optimization of this variable by
varying dopant concentration.
Accordingly, we will sustain the Examiner’s rejection of claims 10—
12,23,24, and 26.
20
Appeal 2015-006350
Application 12/469,441
DECISION
Upon consideration of the record, and for the reasons given above and
in the Answer, the decision of the Examiner rejecting claims 26 under
35 U.S.C. § 112, first paragraph, as failing to comply with the written
description requirement, and rejecting claims 10-12, 23, 24, and 26 under
35 U.S.C. § 103(a) as unpatentable over Lammert in view of Husher,
Chiang, and Kaschmitter is affirmed.
However, for the reasons given above and in the Appeal and Reply
Briefs, the decision of the Examiner rejection claims 1, 3—9, 17—19, and 22
under 35 U.S.C. § 103(a) as unpatentable over Lammert in view of Chiang
and Horzel, alone or further in view of any of Nagashima, Husher, and
Korevaar, 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).
AFFIRMED-IN-PART
21