`Tel: 571-272-7822 Entered: May 1, 2014
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`_______________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`_______________
`
`CORNING INCORPORATED
`Petitioner
`
`v.
`
`DSM IP ASSETS B.V.
`Patent Owner
`_______________
`
`Case IPR2013-00043 (Patent 7,171,103 B2)
`Case IPR2013-00044 (Patent 6,961,508 B2)
`_______________
`
`Before FRED E. McKELVEY, GRACE KARAFFA OBERMANN,
`JENNIFER S. BISK, SCOTT E. KAMHOLZ, and ZHENYU YANG,
`Administrative Patent Judges.
`
`McKELVEY, Administrative Patent Judge.
`
`FINAL WRITTEN DECISION
`35 U.S.C. § 318(a) and 37 C.F.R. § 42.73(b)
`
`
`
`
`
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`I.
`
`INTRODUCTION
`A. Background
`Petitioner, Corning Incorporated (“Corning”) filed ten petitions in
`
`November of 2012, challenging patents owned by DSM Assets B.V.
`(“DSM”).
`All ten petitions were at least partially granted, and therefore,
`progressed into the trial phase of an inter partes review.
`This is the final written decision for IPR2013-00043 and
`IPR2013-00044, both of which raise common issues.
`1. IPR2013-00043
`
`The petition in IPR2013-00043 (Paper 3) challenges claims 1-18
`(all of the claims) of U.S. Patent No. 7,171,103 B2 (Ex. 1001 (“the ’103
`patent”)).
`Patent Owner, DSM, filed a preliminary response on February 21,
`2013. Paper 13 (“Prelim. Resp. 43”).
`On May 13, 2013, the Board granted the petition as to all of the
`proposed grounds. Paper 14.
`The Board found that there was a reasonable likelihood that Corning
`would prevail with respect to the claims challenged in the petition on the
`following grounds:
`
` 2
`
`
`
`
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`
`Claims Challenged Basis Reference(s)1
`1-15
`§ 102 Szum ’157
`1-15
`§ 103 Szum ’157 and Szum ’041
`16 and 17
`§ 103 Szum ’157 and Yamazaki
`16 and 17
`§ 103 Szum ’157, Szum ’041, and Yamazaki
`18
`§ 103 Szum ’157, Yamazaki, and Winningham
`18
`§ 103 Szum ’157, Szum ’041, Yamazaki, and
`Winningham
`After institution of trial, DSM filed a patent owner response (Paper 43
`(“PO Resp. 43”)) and a supplemental response (Paper 75).
`DSM also filed a motion to amend claims submitting proposed new
`claim 19 for claim 12. Paper 45.
`Corning filed (1) a reply to the patent owner response (Paper 64), (2) a
`supplemental reply (Paper 76), and (3) an opposition to DSM’s motion to
`amend (Paper 63).
`DSM then filed a reply in support of its motion to amend. Paper 77.
`2. IPR2013-00044
`The petition in IPR2013-00044 (Paper 2) challenges claims 1-22 (all
`of the claims) of U.S. Patent No. 6,961,508 B2 (Ex. 1001 (“the ’508
`patent”)).
`DSM filed a preliminary response on February 21, 2013. Paper 11.
`On May 13, 2013, the Board granted the petition as to all of the
`proposed grounds. Paper 12.
`
`
`1 The references are: (1) WO 98/21157 (Ex. 1002) (“Szum ’157” also
`referred to in the record as “Szum ʼ21157”); (2) U.S. Patent No. 5,664,041
`(Ex. 1003) (“Szum ’041”); (3) EP 0 874 012 A1 (Ex. 1004) (“Yamazaki”);
`and (4) WO 01/49625 A1 (Ex. 1005) (“Winningham”).
`
` 3
`
`
`
`
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`
`The Board found that there was a reasonable likelihood that Corning
`would prevail with respect to the claims challenged in the petition on the
`following grounds:
`Claims Challenged Basis Reference(s)2
`1-8, 10-13, and 15-22 § 103 Szum ’157 and Szum ’041
`9 and 14
`§ 103 Szum ’157, Szum ’041, and Edwards
`After institution of trial, DSM filed (1) a patent owner response
`(Paper 42), and (2) a supplemental response (Paper 71).
`DSM also filed a motion to amend claims by submitting proposed
`new claim 19 for claim 12. Paper 44.
`Corning filed (1) a reply to the patent owner response (Paper 60), (2) a
`supplemental reply (Paper 72), and (3) an opposition to DSM’s motion to
`amend (Paper 59).
`DSM then filed a reply in support of its motion to amend. Paper 73.
`3. Summary
`Oral argument for both cases took place on February 11, 2014. See
`IPR2013-00043, Paper 94; IPR2013-00044, Paper 91 (Transcripts of Oral
`Argument).
`The Board has jurisdiction under 35 U.S.C. § 6(c).
`This final written decision is issued pursuant to 35 U.S.C. § 318(a)
`and 37 C.F.R. § 42.73.
`
`
`2 The references are: (1) WO 98/21157 (Ex. 1002) (“Szum ’157” also
`referred to in the record as “Szum ʼ21157”); (2) U.S. Patent No. 5,664,041
`(Ex. 1003) (“Szum ’041”); and (3) U.S. Patent No. 5,416,880 (Ex. 1004)
`(“Edwards”).
`
` 4
`
`
`
`
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`
`Corning has failed to show by a preponderance of evidence that any of
`challenged claims 1-18 of the ’103 patent and challenged claims 1-22 of the
`’508 patent are unpatentable.
`Because we do not find any of the challenged claims unpatentable, we
`need not consider DSM’s motions to amend claims, and therefore, the
`motions to amend claims in both IPR2013-00043 and IPR2013-00044 are
`dismissed as moot.
`B. Related Proceedings
`Corning and DSM are simultaneously involved in eight other inter
`partes reviews based on patents claiming similar subject matter:
`(1) IPR2013-00045; (2) IPR2013-00046; (3) IPR2013-00047;
`(4) IPR2013-00048; (5) IPR2013-00049; (6) IPR2013-00050;
`(7) IPR2013-00052; and (8) IPR2013-00053.
`C. The ʼ103 Patent
`The ’103 patent is titled “Coated Optical Fibers” and relates to coated
`optical fibers having primary and secondary coatings and to radiation-
`curable primary coating compositions. Ex. 1001, 1:14-16.
`The patent explains that the “soft ‘cushioning’ ” primary coating is
`usually in contact with the fiber, while the “relatively hard” secondary
`coating surrounds the primary coating. Id. at 1:23-26.
`The coatings confer “microbending” resistance on the optical fiber,
`thereby helping to reduce attenuation of optical power along the fiber. Id. at
`1:27-29.
`The patent is directed, in particular, to coated optical fibers in which
`the primary coating provides “good microbending resistance,” and
`
` 5
`
`
`
`
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`simultaneously, has a “high cure speed” that will not unduly limit production
`rates. Id. at 1:34-37.
`Claims 1 and 16, reproduced below, illustrate the claimed subject
`matter (dispositive limitation in italics):
`1. An inner primary coating composition having:
`(a) an in-situ modulus (after cure) of less than 0.6 MPa;
`(b) a cure dose to attain 95% of the maximum attainable
`modulus of less than 0.65 J/cm2; and
`(c) a modulus retention ratio (after cure) of at least 0.6
`after hydrolytic aging; wherein said composition
`comprises:
`(i) 20–98 wt. % relative to the total weight of the
`composition of a radiation curable urethane (meth)
`acrylate oligomer having polyether polyol
`backbone;
`(ii) 0–80% wt. % relative to the total weight of the
`composition of one or more reactive diluents;
`(iii) 0.1–20 wt. % relative to the total weight of the
`composition of one or more photoinitiators; and
`(iv) 0–5 wt. % relative to the total weight of the
`composition of additives.
`16. A coated optical fiber comprising:
`(a) an optical fiber;
`(b) a primary coating obtained by curing the coating
`composition according to claim 1;
`(c) a secondary coating, wherein said secondary coating
`has:
`(i) a Tg of about 60° C. or higher;
`(ii) an elongation at break of at least 20%; and
`(iii) a tensile modulus of at least 500 MPa.
`D. The ’508 Patent
`The ’508 patent is titled “Coated Optical Fibers” and relates to coated
`optical fibers having primary and secondary coatings and to radiation-
`curable, primary coating compositions. 44 Ex. 1001, 1:12-16.
`
` 6
`
`
`
`
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`
`The patent explains that the “soft ‘cushioning’ ” primary coating is
`usually in contact with the fiber, while the “relatively hard” secondary
`coating surrounds the primary coating. Id. at 1:19-26.
`The coatings confer “microbending” resistance on the optical fiber,
`thereby helping to reduce attenuation of optical power along the fiber. Id. at
`1:23-26.
`The patent is directed, in particular, to coated optical fibers in which
`the primary coating provides “good microbending resistance” and
`simultaneously has a “high cure speed” that will not unduly limit production
`rates. Id. at 1:30-34.
`Claims 1 and 20, reproduced below, illustrate the claimed subject
`matter (dispositive limitations in italics):
`1. A coated optical fiber comprising:
`(i) an optical fiber;
`(ii) a primary coating; and
`(iii) a secondary coating;
`wherein
`(a) said coated optical fiber has an attenuation
`increase of less than 0.650 dB/km at 1550 nm;
`(b) said primary coating has a modulus retention ratio
`after hydrolytic aging of at least 0.5 and/or a glass
`transition temperature (Tg) below –35° C.; and
`(c) said primary coating is obtained by curing a
`primary coating composition having a cure dose to
`attain 95% of the maximum attainable modulus of
`less than 0.65 J/cm2.
`20. An inner primary coating composition having:
`(a) an in-situ modulus (after cure) of less than 0.6 MPa;
`(b) a cure dose to attain 95% of the maximum attainable
`modulus of less than 0.65 J/cm2; and
`(c) a modulus retention ratio (after cure) of at least 0.6
`after hydrolytic aging.
`
` 7
`
`
`
`
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`
`
`II. ANALYSIS
`A. Claim Construction
`As a step in our analysis for determining whether the challenged
`claims are unpatentable, we determine the meaning of the claims.
`The Board interprets claims of an unexpired patent using the broadest
`reasonable construction in light of the specification of the patent. See
`37 CFR § 42.100(b); Office Patent Trial Practice Guide, 77 Fed. Reg.
`48,756, 48,766 (Aug. 14, 2012).
`The dispositive claim limitation in the ʼ103 patent is limitation (b) of
`claim 1: “a cure dose to attain 95% of the maximum attainable modulus of
`less than 0.65 J/cm2.”
`The dispositive claim limitation in the ʼ508 patent is limitation (c) of
`claim 1: “said primary coating is obtained by curing a primary coating
`composition having a cure dose to attain 95% of the maximum attainable
`modulus of less than 0.65 J/cm2.”
`Corning has failed to prove that prior art compositions inherently
`attain 95% of the maximum attainable modulus at a cure dose of less than
`0.65 J/cm2.
`DSM has failed to prove that the prior art compositions do not attain
`95% of the maximum attainable modulus at a cure dose of less than 0.65
`J/cm2.
`Accordingly, there is no occasion to construe further the language of
`limitation (b) of claim 1 of the ʼ103 patent or limitation (c) of the ʼ508
`patent, hereinafter the “95% limitations.”
`B. Testimony and documentary evidence
`
` 8
`
`
`
`
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`The testimony and documentary evidence relevant to the “95%
`
`limitations” in both IPRs are essentially the same.
`
`We discuss the testimony and evidence, citing to the evidence of
`record in IPR2013-00043.
`Nevertheless, Table 1, reproduced below, correlates chronologically
`by exhibit number the common testimony and documentary evidence
`submitted, and upon which we have relied in deciding both IPR2013-00043
`and IPR2013-00044.
`
`Table 1
`Description of testimony or
`Ex. number in
`document
`IPR2013-00043
`U.S. Patent 5,416,880 “Edwards”
`None
`ʼ508 Patent
`None
`ʼ103 Patent
`1001
`WO 98/21157 “Szum ʼ157”
`1002
`U.S. Patent 5,664,041 “Szum ʼ041”
`1003
`Winningham declaration
`1006
`Kouzmina declaration
`1007
`Reichmanis declaration
`1028
`Bowman cross-examination
`1033
`Anderson
`1038
`Schmid
`1047
`Chawla
`1050
`Kouzmina cross-examination
`2022
`Kouzmina cross-examination
`2024
`Winningham cross-examination
`2028
`Bowman declaration
`2030
`Dose-segment modulus data
`2049
`Dose-segment modulus data
`2050
`Dose-segment modulus data
`2051
`Dose-segment modulus data
`2052
`Curve fit and statistical analysis
`2053
`Proc. of the Int’l Soc. for Optical Eng’g
`2058
`
`Ex. number in
`IPR2013-00044
`1004
`1001
`None
`1003
`1002
`1005
`1006
`1026
`1031
`1036
`1045
`1048
`2021
`2022
`2027
`2029
`2048
`2049
`2050
`2051
`2052
`2057
`
` 9
`
`
`
`
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`In support of its petitions, Corning offered the testimony of Ms. Inna
`
`I. Kouzmina (Ex. 1007).
`According to Kouzmina, compositions were made in accordance with
`Example 10-1 and Example 10-2 as described by Szum ʼ157 (Szum ʼ157,
`118:10-21; Table 15).
`
`Kouzmina also testified (Ex. 1007 ¶ 29) concerning “[t]ests . . .
`conducted on primary coatings as described in Example 10-1 and
`Example 10-2 of Szum [ʼ157], to measure the cure dose to attain 95% of the
`maximum attainable modulus of the cured coatings in accordance with
`procedures set forth in the ʼ103 patent at 9:21-43.”
`
`Kouzmina explains (Ex. 1007 ¶ 30):
`Specifically, cure dose was determined by Dose vs.
`Modulus curve analysis. Six radiation-cured sample films of
`each composition were prepared, with each sample film being
`obtained by applying and curing, at room temperature under a
`nitrogen atmosphere, a composition having a thickness of
`approximately 75 microns on a glass plate. Each composition
`was cured with a different dose: 0.2, 0.3, 0.5, 0.75, 1.0, and 2.0
`J/cm2 respectively. Six specimens were cut from the center
`portion of each prepared sample film. MTS Tensile Tester of
`MTS Systems Corporation was used to measure the 2.5%
`secant modulus of each specimen. The dose-modulus curve
`was then created by plotting the modulus values vs. the dose
`and by fitting a curve through the data points. The “cure dose”
`of the coating composition was determined to be the dose at
`which 95% of the ultimate secant modulus was attained.
`The experimental cure dose to attain 95% of the maximum attainable
`
`modulus is described in Kouzmina Table B (Ex. 1007 ¶ 31):
`
`
`
`
`10
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`
`
`In a light most favorable to Corning, we will assume the “modulus”
`
`mentioned in paragraph 31 and Table B is a secant modulus.
`
`In support of its petitions, Corning also offered the testimony of
`Dr. Michael Winningham (Ex. 1006).
`According to Winningham (id. ¶ 92):
`
`Element (b) of [ʼ103 patent] claim 1 requires the primary
`coating composition to have “a cure dose to attain 95% of the
`maximum attainable modulus of less than 0.65 J/cm2.” The
`primary coating composition[s] of Examples 10-1 and 10-2 of
`Szum [ʼ157] have a cure dose of 0.37 J/cm2 and 0.22 J/cm2,
`respectively, to attain 95% of the maximum attainable modulus.
`Kouzmina Decl. [Ex. 1007] ¶ 31. Thus, the limitation of
`element (b) of claim 1 is met by the disclosure of Szum [’157].
`Kouzmina’s cross-examination reveals that she oversaw, but did not
`
`personally conduct, the experimental work reported in her direct testimony.
`
`Kouzmina testified that (1) she instructed two Corning scientists to
`make oligomers per information described in the prior art (presumably
`including Szum ʼ157), and (2) asked them to make compositions described
`in the prior art and report back to her when their experimental work was
`done. Ex. 2022, 433:17–435:2.
`In due course, the two scientists “generated results and reported
`back . . . those results [to Kouzmina].” Id. at 484:21-23.
`When asked how she knew the two Corning scientists accurately
`made the oligomers, Kouzmina testified that “I had their notes and they
`
`
`
`
`11
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`reported to me for this project, so I have their information.” Id. at
`436:20-22.
`What Kouzmina means by “notes” is not clear; we will assume that
`the “notes” are laboratory-like notebooks and related documents.
`When asked “[w]hat level of scrutiny did you give the oligomer
`synthesis,” Kouzmina responded “I . . . instructed the [Corning scientists]
`making oligomers to follow the . . . prior art as closely as possible. And I
`trusted their judgment on executing this.” Ex. 2024, 866:25–867:8.
`
`Winningham testified that he did not “review any lab notebooks” and
`“I don’t recall seeing any other data.” Ex. 2028, 690:18-19.
`
`Data and other documents related to Corning’s tests did not
`accompany the Petition.
`
`As a result of “additional discovery” (37 C.F.R. § 41.150(c)), DSM
`obtained some laboratory notebooks and other documents. PO Resp. 43, 9.
`
`After analysis of all the evidence available to it, DSM challenges the
`accuracy of the “Results” set out in Table B. Id. at 20-21.3
`
`In support of its response, DSM submitted the direct declaration
`testimony of Dr. Christopher N. Bowman. Id. at 21.
`
`
`3 On page 20, footnote 2, of the patent owner’s response (PO Resp. 43),
`DSM attempts to incorporate by reference arguments made in its preliminary
`response (Prelim. Resp. 13). We decline to consider arguments incorporated
`by reference. Incorporation by reference is an unacceptable means of
`permitting a party to exceed page limits set out in the rules. 37 C.F.R.
`§ 42.24; see also DeSilva v. DiLeonardi, 181 F.3d 865, 866-67 (7th Cir.
`1999) (“[A]doption by reference amounts to a self-help increase in the
`length of the appellate brief.” “[I]ncorporation [by reference] is a pointless
`imposition on the court’s time. A brief must make all arguments accessible
`to the judges, rather than ask them to play archaeologist with the record.”).
`
`12
`
`
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`DSM’s challenge is based on two rationales: (1) its own counter-
`
`testing (id. at 21-25); and (2) Corning’s cure dose analysis, which is said to
`be statistically invalid (id. at 26-28).
`
`Bowman testified that he understood that DSM prepared compositions
`described in Examples 10-1 and 10-2 of Szum ʼ157. Ex. 2030 ¶ 71.
`
`Bowman further testified that DSM determined cure doses to attain
`95% maximum attainable modulus of the compositions described in
`Examples 10-1 and 10-2 of Szum ʼ157. Id. ¶¶ 113-14.
`
`According to Bowman (Ex. 2030 ¶ 114):
`Six sample films of each of Szum ’157 Examples 10-1
`and 10-2 were prepared on a plate and exposed to a dose of 0.2
`J/cm2, 0.3 J/cm2, 0.5 J/cm2, 0.75 J/ cm2, 1.0 J/ cm2, and 2.0
`J/cm2, respectively, from a 600 W “D”-lamp under a blanket of
`nitrogen to ensure maximum cure at each dose. Prior to
`exposure of the samples, the dose was measured three times
`using a calibrated ILT490 broadband UVA/UVB radiometer
`from International Light Technologies, which is calibrated at
`least once every six months, to confirm proper dosage. After
`each sample was exposed to the relevant dose, it was
`conditioned at about 23±1ºC and at a relative humidity of about
`50% for sixteen to twenty-four hours. The center of each
`sample was then cut into specimens having a width of 12.7 mm.
`The thickness of each specimen was measured at five different
`locations to confirm that each specimen exhibited uniform
`thickness. This data, as well as the average thickness of each
`specimen, is summarized in Exhibits 2049-2052.
`Exhibits 2049-52 describe DSM “mean” thickness data as follows:
`
`
`
`
`
`
`13
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`
`Segment or Secant Modulus
`Mean thickness
`Exhibit
`Segment
`0.086
`2049
`Secant
`0.090
`2050
`Segment
`0.095
`2051
`Secant
`0.099
`2052
`DSM’s average segment and secant moduli of reproduced Examples
`
`10-1 and 10-2 as a function of cure dose (J/cm2) are reported in Table 3 of
`Bowman’s declaration (Ex. 2030 ¶ 115).
`Table 3
`
`Error in Table 3 is reported as one standard deviation.
`Data described in Table 3 and Exhibits 2049-52 was used to plot
`
`segment and secant modulus (MPa) as a function of dose (J/cm2) curves.
`
`An example of a DSM curve of segment modulus as a function of
`dose for Szum ʼ157 Example 10-1 is set out below (Ex. 2030).
`
`
`
`
`
`
`14
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`
`
`
`
`
`Depicted is a graph of segment modulus v. dose
`for Szum ʼ157 Example 10-1
`According to Bowman (id. ¶ 117):
`After curve-fitting, the cure dose to attain 95% of the maximum
`attainable segment modulus was determined to be 0.74 J/cm2
`with a 95% confidence interval ranging from 0.72 J/cm2 to 0.76
`J/cm2. R2 for the fit was 0.9786 based on the average segment
`modulus at each dose and 0.9640 based on all data points at
`each dose. (See Ex. 2053[, 2 of 11 and 11 of 11]).
`Bowman explains the significance of “R2” as follows (Ex. 2030
`
`¶ 122):
`In statistics, R2 is known as the coefficient of determination.
`In a curve fitting analysis, R2 is a measure of the correlation
`between the fitted curve and the observed data. R2 generally
`falls within a range of 0 to 1. An R2 value close to 1 indicates a
`strong correlation between the fitted curve and the observed
`data, which means that the fitted curve accurately reflects the
`behavior of the observed data. When R2 is closer to 1, the fitted
`curve can be used to accurately predict the behavior of a
`system. In contrast, an R2 value close to 0 indicates very little
`correlation between the fitted curve and the observed data,
`which means that the fitted curve does not accurately reflect the
`behavior of the observed data. When R2 is closer to 0, the fitted
`
`15
`
`
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`
`curve cannot be used to accurately predict the behavior of a
`system.
`A summary of DSM cure dose data for segment and secant moduli for
`
`Szum ʼ157 appears in Table 4 of Bowman’s declaration testimony (id.
`¶ 117).
`
`Table 4
`
`
`
`According to the data in Table 4, DSM did not obtain 95% of the
`
`maximum attainable modulus with a dose of less than 0.65 J/cm2 in
`reproducing Szum ʼ157 Examples 10-1 and 10-2. Ex. 2030 ¶ 118.
`
`Bowman therefore found that Examples 10-1 and 10-2 cannot
`inherently describe subject matter within the scope of claim 1 of the ’103
`patent. Id. ¶ 119.
`
`Bowman also addressed the statistical validity of Corning’s
`experimental results. Id. ¶ 120.
`
`Unlike DSM, Corning did not provide underlying data with its
`petition to support cure doses reported by Kouzmina and Winningham.
`
`Bowman Table 5, reproduced below, compares cure dose measured
`by Corning and DSM for secant modulus. Id.
`
`
`
`
`16
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`
`Table 5
`
`
`
`While Corning believes its cure dose falls within the scope of claim 1
`
`of both the ’103 patent and the ʼ508 patent, the DSM cure dose does not.
`
`Through additional discovery, DSM was able to obtain the underlying
`data said to support Corning’s cure dose “results.”
`
`According to Bowman, the underlying data “does not support the
`conclusion that Examples 10-1 or 10-2 exhibit a cure dose that necessarily
`falls within the scope of the claims of the ʼ103 patent.” Ex. 2030 ¶ 120.
`
`Bowman Table 6, reproduced below, summarizes Corning’s
`underlying data produced to support its cure dose values. Id. ¶ 121.
`Bowman Table 6
`
`
`
`
`
`
`17
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`Bowman Figure 5, reproduced below, represents a curve fit for
`
`secant modulus (MPa) versus dose (J/cm2) for Example 10-1 of Szum ʼ157.
`Id. Fig. 5.
`
`
`
`Bowman Fig. 5 depicts a plot of secant modulus v. dose
`for Szum ʼ157 Example 10-1
`The R2 fitting reported by Corning is said to be 0.9765. Ex. 2030, 57.
`
`Bowman testified that upon review of the Corning R2 values, “I was
`
`immediately suspicious . . .” (id. ¶ 122), apparently because the “dose data
`. . . looks relatively constant” (id. ¶ 123).
`
`Bowman asked DSM to independently analyze Corning’s data using
`essentially the same approach used by DSM to analyze cure dose data of its
`experiments. Id. ¶ 124.
`Bowman Table 7, reproduced below, compares Corning’s R2 values
`
`vis-à-vis DSM’s R2 values.
`
`
`
`
`18
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`
`Bowman Table 7
`
`
`
`
`
`As a result of Corning’s relatively lower R2 values, which Bowman
`
`characterizes as “significantly low R2 values,” Bowman opines (Ex. 2030
`¶ 126 (footnote omitted)):
`Upon seeing such low R2
` values, a person of ordinary skill in
`the art would have immediately recognized that . . . Corning’s
`data was subject to some type of systematic or experimental
`error (e.g., the use of a malfunctioning or uncalibrated
`radiometer) . . . . Accordingly, Corning’s data set cannot
`eliminate systematic and experimental errors as an explanation
`as to why its data does not conform to the modulus-dose
`behavior predicted by Equation 1 [(see Ex. 2030 ¶ 116)] and as
`typically observed. (Ex. 2058).
`
`
`
`
`19
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`Figure 1 of Exhibit 2058, reproduced below, shows equilibrium
`
`moduli ranging from about 65 to about 100% as a function of cure dose.
`
`
`
`Depicted is a curve of modulus as a function of cure energy.
`Corning’s modulus curves show moduli ranging from (1) about
`0.85 to about 1.2 MPa as a function of cure dose (Ex. 2030 Fig. 5), and
`(2) about 0.60 to about 0.65 MPa as a function of cure dose (Id. Fig. 6) for
`the compositions of Examples 10-1 and 10-2 of Szum ’157, respectively.
`According to Bowman, “[a]s indicated by Figure 1 in Exhibit 2058, as
`early as 1993, data for modulus as a function of dose was readily obtained[,]
`and data over a broad range of modulus values are indicated.” Id. ¶ 131.
`Corning has not pointed to any evidence that Winningham conducted
`any independent analysis of Corning’s curve fitting. Cf. Ex. 2028, 761:19 –
`762:9 (discussing, on cross-examination, an “Exhibit 130” and “Exhibit
`131”—exhibits we have not found in the record and, in any event, would not
`be properly numbered exhibits).
`
`
`
`
`20
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`In reply to Bowman’s curve-fitting analysis, Corning submitted
`
`testimony of Dr. Elsa Reichmanis. Ex. 1028.
`
`Corning relies on Reichmanis to make two points: (1) Corning’s
`curve-fitting is based on a proper analysis, while DSM’s curve-fitting is not;
`and (2) the difference in cure dose may be a result of the thickness of the
`films tested on behalf of Corning vis-à-vis those tested on behalf of DSM.
`
`A first observation by Reichmanis is that Corning’s modulus versus
`dose curves generally have the same overall appearance as typical modulus
`versus dose curves—steep initial rise then a generally leveling off. See, e.g.,
`id. ¶ 74 (referring to Ex. 1047 Fig. 4.13), ¶ 79 (referring to Ex. 1050 Fig. 6).
`
`Reichmanis therefore “find[s] nothing peculiar or suspicious about
`Corning’s modulus versus dose curves for Examples 10-1 and 10-2.”
`Ex. 1028 ¶ 80.
`A second observation is that Bowman, in determining R2 values,
`
`excluded the origin (data point 0,0) from the calculation. Id. ¶ 83.
`
`Reichmanis testified that “I have reviewed Corning’s recalculation of
`its own R2 values by excluding the origin (data point 0,0) from the
`calculation and I observed that they obtained R2 values similar to what DSM
`obtained.” Id.
`
`According to Reichmanis, “it makes [no] statistical sense to exclude
`the origin (data point 0,0) from the data.” Id. ¶ 84.
`
`Reichmanis points to no credible underlying data to support her
`testimony.
`Reichmanis notes that (id. ¶ 85):
`
`Regardless, the issues raised by DSM and Dr. Bowman
`regarding Corning’s R2 values are not dispositive, because what
`
`21
`
`
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`
`matters is what dose is required to attain 95% of the maximum
`attainable modulus and not how well the curve fits to the data.
`For example, if one digitizes the above modulus vs. dose curve
`from . . . Ex. 1050 [(presumably referring to Figure 6)] . . . and
`calculates the R2 with the origin (0,0), the adjusted R2 is
`approximately 0.994, and without the point of origin (0,0), the
`adjusted R2 is approximately 0.163. What I view in terms of
`Corning’s data and their modulus vs. dose curves are entirely
`consistent with what I would expect for coatings that have a
`cure dose to attain 95% of the maximum attainable modulus of
`0.37 J/cm2 and 0.22 J/cm2, respectively.
`Reichmanis does not explain what recalculation of R2 (presumably
`
`Figure 6 of Exhibit 1050) has to do with R2 values associated with curve-
`fitting of Corning’s and DMS’s reproductions of Szum ʼ157 Examples 10-1
`and 10-2.
`
`Moreover, unlike Bowman, who called attention to DSM calculations
`(see, e.g., Ex. 2053), Reichmanis does not call our attention to
`documentation underlying her calculation of adjusted R2.
`C. Discussion
`
`1. Arguments considered
`
`In resolving this case on the merits, we consider only the arguments
`
`made in the Petition, the Response, and the Reply.
`
`We find that:
`(1) Corning has not established by a preponderance of the
`evidence that Examples 10-1 and 10-2 inherently describe an inner
`primary coating meeting the “95% limitation” required by paragraph
`(b) of claim 1 of the ʼ103 patent or paragraph (c) of claim 1 of the
`ʼ508 patent, and
`
`
`
`
`22
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`
`(2) DSM has not established by a preponderance of the
`evidence that Examples 10-1 and 10-2 do not meet the required “95%
`limitation”.
`Accordingly, we do not find it necessary to consider DSM’s
`Supplemental Response (Paper 75) or Petitioner’s Reply thereto (Paper 76).
`
`In reaching our decision that Corning has failed to establish inherency
`as to the “95% limitation,” we have assumed that both Corning and DSM
`accurately reproduced the coatings of Szum ’157. In light of our findings, it
`is not necessary to determine whether the coatings were accurately
`reproduced.
`2. Corning’s proofs
`
`
`
`On the factual issue of whether Corning has established that Szum
`ʼ157 inherently describes a coating having the required cure dose, there is
`conflicting testimony of Bowman and Reichmanis.
`
`Both witnesses appear to be qualified scientists.
`We have no reason to question the good faith of either witness and we
`believe they have testified faithfully to their respective opinions.
`
`Nevertheless, to the extent that there is a conflict between the two
`witnesses with respect to Corning’s cure dose proofs, we credit Bowman
`over Reichmanis.
`
`The Bowman testimony is detailed and supported by underlying data,
`while the Reichmanis testimony is general and is not credibly supported by
`underlying data. 37 C.F.R. § 42.65(a), (b)(5).
`
`Corning argues that Bowman ignored the origin (0,0 data point).
`At a dose of zero (0) J/cm2, Bowman tells us that the modulus would
`be “close to zero.” Ex. 1034, 1097:25.
`
`23
`
`
`
`
`
`IPR2013-00043 (Patent 7,171,103 B2)
`IPR2013-00044 (Patent 6,961,508 B2)
`
`Reichmanis believes the origin should be included in any R2
`
`determination and maintains that the only thing that counts is the cure dose
`(not the R2 value). Ex. 1028 ¶¶ 84-85.
`
`Fitting data points to a curve is important.
`The data points, when properly fitted to a curve, permit one skilled in
`the art to determine whether 95% of the maximum attainable modulus is
`attainable at the relevant cure dose for a given composition.
`
`According to the ʼ103 patent, cure doses of 0.2, 0.3, 0.5, 0.75, 1.0, and
`2.0 J/cm2 are to be employed. ’103 patent 9:58-60; see also ’508 patent
`7:58-59.
`
`A “dose-modulus curve was then created by plotting the modulus
`values vs. the dose and by fitting a curve through the data points.”
`’103 patent 9:39-41.
`The ’103 patent and the ’508 patent do not indicate that the data
`points are to include the origin (0,0 data point).
`
`Reichmanis provides no credible explanation as to why the origin
`should be included in the face of the explicit disclosure of the ʼ103 patent
`and the ʼ508 patent.
`
`Once the data points are curve-fitted, the “maximum” modulus
`surfaces at the point where the curve flattens out. See, e.g., Ex. 2030, Fig. 1
`(essentially between a dose of 1.00 and 2.00 J/cm2).
`