Opinion
No. C 01-4925 SBA, (Related to Case No. C 05-2133), [Docket No. 502].
February 24, 2006
ORDER
This matter comes before the Court on Defendants' Motion to Exclude the Expert Report and Testimony of Plaintiff's Proposed Expert Luciano Mule'stagno. Having read and considered the arguments presented by the parties in the papers submitted to the Court, the Court finds this matter appropriate for resolution without a hearing. The Court hereby GRANTS Defendants' Motion to Exclude the Expert Report and Testimony of Plaintiff's Proposed Expert Luciano Mule'stagno [Docket No. 502].
BACKGROUND
In this patent infringement suit, Plaintiff MEMC Electronics Materials, Inc. ("MEMC") has accused Defendants Mitsubishi Materials Silicon Corp., Mitsubishi Silicon America Corp., Sumitomo Mitsubishi Silicon Corp., SUMCO USA Corp. and SUMCO USA SALES Corp. (collectively, "SUMCO") of actively inducing the infringement of U.S. Patent No. 5,919,302 (the "`302 Patent"). The '302 Patent was issued on July 6, 1999 and claims an invention related to the manufacture of silicon crystals for use in pure silicon wafers.
The asserted claims from the '302 Patent, claims 1-6 and 9-12, are directed to a silicon wafer that contains a "first axially symmetric region in which vacancies are the predominant intrinsic point defect and which is substantially free of agglomerated vacancy intrinsic point defects wherein the first axially symmetric region comprises the central axis or has a width of at least about 15 mm." The claims also require a "second axially symmetric region in which silicon self-interstitial atoms are the predominant intrinsic point defect and which is substantially free of agglomerated silicon self-interstitial intrinsic point defects."
Per the Court's Markman ruling, "substantially free of agglomerated vacancy intrinsic point defects" means "a concentration of such agglomerated defects which is less than the detection limit of these defects, which is currently about 1000 defects/cm3."
On May 1, 2003, the Court held a claim construction hearing for the purpose of construing the claims of the '302 Patent. On October 17, 2003, MEMC served SUMCO with the expert report of Luciano Mule'Stagno (the "Expert Report"). The Expert Report consisted of the results of testing Mule'Stagno performed on silicon wafers produced during discovery by SUMCO. It comprised seven pages of narrative opinion and almost 500 pages of materials relating to testing methodology and results. Based upon the testing performed and described in the Expert Report, Mule'Stagno concluded that SUMCO's silicon wafers infringe the '302 Patent and U.S. Patent No. 6,287,380 (the "'380 Patent") literally and under the doctrine of equivalents. Up until the disclosure of the Expert Report, MEMC had not alleged infringement of the '380 Patent.
On December 8, 2003, SUMCO filed a Motion to Exclude the Expert Report and Testimony of Plaintiff's Proposed Expert Luciano Mule'Stagno and to Preclude Plaintiff's Infringement Claims ("SUMCO's First Motion to Exclude Expert").
On March 1, 2004, the Court issued an Order granting SUMCO's First Motion to Exclude Expert. The Court granted SUMCO's motion to exclude Mule'Stagno's Expert Report and testimony to the extent that it related to infringement of the '380 Patent. The Court also granted SUMCO's motion to preclude MEMC from pursuing its claim of infringement under the doctrine of equivalents and SUMCO's motion to bar MEMC from offering any expert report or testimony that SUMCO has infringed the '302 Patent under the doctrine of equivalents. Additionally, the Court granted SUMCO's motion to strike the "claim charts" in Mule'Stagno's Expert Report but denied SUMCO's motion to preclude MEMC's infringement claims.
With respect to Mule'Stagno's Expert Report as it pertained to the '302 Patent, the Court noted several deficiencies that rendered the testimony inadmissible pursuant to Rule 702. Specifically, the Court found that:
The Report merely identified the tests that Mr. Mule'Stagno used to test the wafers but does not explain in detail the methodology he used to test the wafers. For example, the Report states that `[t]he Light Point Defect Scans performed at sizes greater than 0.12 m clearly show lots of counts in a region extending from the center to about 10mm from the wafer edge. . . . Looking at these LPD maps, it is clear we are looking at wafers which have a large area covered by COPs which manifest themselves as LPDs in this test.' MEMC's Opp. App. A at ii. It is not evident from this statement how the Light Point Defect Scans are performed, what such scans purport to show and how the results are relevant to Mule'Stagno's conclusions.
March 1, 2004 Order at 11. Additionally, the Court noted that:
The Report also does not demonstrate that the research and analysis supporting the proffered conclusions have been subjected to normal scientific scrutiny through peer review and publication or identify an objective source demonstrating that the methods and premises that he used were generally accepted by or espoused by a recognized minority of scientists in his field. See Lust by Through Lust v. Merrell Dow Pharms., 89 F.3d 594, 597 (9th Cir. 1996). Mr. Mule'Stagno has submitted a declaration in connection with MEMC's Opposition that states that the test that he used, the bulk microdefect density test ('BMD test'), has been used throughout the silicon industry for 15 years to test silicon. Mathiowetz Decl. Exh. F ¶ 5. Furthermore, Mr. Mule'Stagno states that "it is routine practice in the industry to test a single strip of silicon (where the strip starts at the center of the wafer and ends at the outer edge)' in order to characterize the entire wafer. Id. But nowhere in his declaration does he point to an objective source, such as a learned treatise, the policy statement of a professional association, a published article in a reputable scientific journal or the like, demonstrating that his methods and premises were generally accepted by or espoused by a recognized minority of scientists in his field. Daubert II, 43 F.3d at 1319. Without objective verification that any of Mr. Mule'Stagno's methods were reliable, the Court cannot find that the proffered evidence is both relevant and reliable as required by Daubert.Id.
Pursuant to Federal Rule of Civil Procedure 26(a)(2)(C) and (e)(1), MEMC was given an opportunity to cure the defects in Mule'Stagno's Expert Report by providing objective verification of Mule'Stagno's methodology. In particular, MEMC was directed to show, via objective sources, what Mule'Stagno's tests were, what the results purported to show, and how the results were relevant to the conclusions. MEMC was also instructed to demonstrate, via objective sources, that Mule'Stagno's methodologies were generally accepted by or espoused by a recognized minority of scientists in the field.
In response to the Court's Order, on March 9, 2004, MEMC served SUMCO with the Second Declaration of Mule'Stagno. See Powers Decl. ISO Second Mot. to Exclude Expert ("Powers Decl.") at Ex. 3. However, on March 16, 2004, before SUMCO had a chance to respond to the Second Declaration of Mule'Stagno, this Court granted SUMCO's Motion for Summary Judgment of Zero Damages and specifically found that SUMCO was not liable for either direct infringement or inducement of infringement. Accordingly, on April 22, 2004, the court entered a Judgment of Non-infringement. MEMC subsequently appealed.
On August 22, 2005, the Federal Circuit issued its decision on the appeal and held that this Court did not err when it granted summary judgment on the issue of direct infringement. However, the Federal Circuit held that the Court did err when it granted summary judgment with respect to inducement of infringement because it found that there were genuine issues of material fact with respect to whether SUMCO induced infringement of the '302 Patent on the part of Samsung Austin. Accordingly, the Federal Circuit remanded the case to this Court for further proceedings.
On November 4, 2005, SUMCO filed the instant Motion to Exclude the Expert Report and Testimony of Plaintiff's Proposed Expert Luciano Mule'stagno ("Second Motion to Exclude Expert").
LEGAL STANDARD
Expert testimony is admissible pursuant to the Federal Rules of Evidence, primarily Rule 702. Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 589 (1993). Under Daubert, the district court acts as a "gatekeeper," excluding "junk science" that does not meet the standards of reliability required under Rule 702. Gen. Elec. Co. v. Joiner, 522 U.S. 136, 142 (1997); Kennedy v. Collagen Corp., 161 F.3d 1226, 1229-30 (9th Cir. 1998). The trial court accomplishes this goal through a preliminary determination that the proffered evidence is both relevant and reliable. Daubert, 509 U.S. at 589-95.
Scientific evidence is deemed reliable if the principles and methodology used by the expert proffering it are grounded in the methods of science. Id. at 592-95. In Daubert, the Supreme Court gave a non-exhaustive list of factors for determining whether scientific testimony is sufficiently reliable to be admitted into evidence, including: (1) whether the scientific theory or technique can be (and has been) tested; (2) whether the theory or technique has been subjected to peer review and publication; (3) whether there is a known or potential error rate; and (4) whether the theory or technique is generally accepted in the relevant scientific community. Id. at 593-94.
The Ninth Circuit has explained that if an expert did not conduct his or her own research, independent of the litigation, on the subject of the testimony, the district court must determine whether there exists any "objective, verifiable evidence that the testimony is based on `scientifically valid principles.'" Daubert v. Merrell Dow Pharms., Inc., 43 F.3d 1311, 1317-18 (9th Cir. 1995) (" Daubert II"). An expert may demonstrate the scientific validity of a theory or technique by showing that "the research and analysis supporting the proffered conclusions have been subjected to normal scientific scrutiny through peer review and publication." Id. at 1318. Alternatively, testifying experts may also show the validity of their theory by explaining "precisely how [the experts] went about reaching their conclusions and pointing to some objective source — a learned treatise, the policy statement of a professional association, a published article in a reputable scientific journal or the like — to show that they have followed the scientific method, as it is practiced by (at least) a recognized minority of scientists in their field." Id. at 1319.
The proponent of the expert testimony bears the burden of establishing the testimony's admissibility by a preponderance of the evidence. Butler v. Home Depot, Inc., 984 F.Supp. 1257, 1260 (N.D. Cal. 1997). The Court's gatekeeper role under Daubert requires the Court to make a "preliminary assessment of whether the reasoning or methodology underlying the testimony is scientifically valid and of whether that reasoning or methodology properly can be applied to the facts in issue." Daubert, 509 U.S. at 592-93.
ANALYSIS
In its Second Motion to Exclude Expert, SUMCO argues that the Second Declaration and Sixth Declaration of Mule'Stagno do not comply with the Court's March 1, 2004 Order and should therefore be excluded in their entirety. Specifically, SUMCO argues that MEMC has not effectively shown that Mule'Stagno has used scientifically reliable and recognized industry tests to establish whether the accused wafers have an "axially symmetric region," or whether "vacancies are the predominant intrinsic point defect," or whether the wafers have a region that is "substantially free of agglomerated vacancy intrinsic point defects" as defined by this Court.
The first flaw that SUMCO identifies in Mule'Stagno's Second Declaration is the fact that Mule'Stagno has not identified a single test that meets the Daubert standard of reliability and that supports his conclusion that the accused wafers contain the claim limitation of an axially symmetric region. As SUMCO points out, MEMC has the burden of proving that SUMCO's wafers, which are formed from a different process than prior art wafers, are axially symmetric. Instead of proving axial symmetry, Mule'Stagno asks the Court to assume that the accused wafers are axially symmetric based solely upon what he contends are similar "assumptions" made in prior art literature. However, not only do these articles concern wafers formed by a different process, but Mule'Stagno's own subjective "summary" of the articles is not sufficient to meet the standard set forth in the Court's March 1, 2004 Order.
For example, Mule'Stagno asserts that Shimura's discussion of striations in wafers "demonstrates his appreciation and acceptance of the axial symmetry of Czochralski silicon." See Powers Decl. at Ex. 3. Mule'Stagno also asserts that Zimmerman's Figure 1 "reflects an understanding that one-half or even one-quarter of a wafer may be tested," and that in Voronkov's work "[i]mplied is the fact that these regions would have appeared the same regardless of the precise position of the longitudinal cut." Id.
Similarly defective is Mule'Stagno's Sixth Declaration, in which he refers to two papers co-authored by Park and two SUMCO employees, Furuya and Harada. Although Mule'Stagno claims that these articles corroborate his view that all Czochralski wafers are axially symmetric, critically, his analysis fails to distinguish between silicon grown by prior art methods and silicon grown under the Park/Samsung method. Further, Mule'Stagno's summary of the articles is misleading. For example, while Mule'Stagno states that "[f]rom Figure 5, the authors identify an axially symmetric outer region of the silicon 20 mm wide," the term "axial symmetry" is, in fact, completely absent from the articles. As stated by the Ninth Circuit in Domingo v. T.K., 289 F.3d 600 (9th Cir. 2002), "[n]othing in either Daubert or the Federal Rules of Evidence requires the district court to admit opinion evidence that is connected to existing data only by the ipse dixit of the expert." Id. at 607. As such, Mule'Stagno's bare assertion that axial symmetry is regularly "assumed" and does not need to be proven is insufficient to defeat SUMCO's Second Motion to Exclude Expert.
The second flaw that SUMCO identifies in Mule'Stagno's declarations is the fact that he has not provided objective verification that it is scientifically valid to prove axial symmetry by testing only a sliver of the accused wafer. This Court previously expressed concern regarding Mule'Stagno's opinion that "it is routine practice in the industry to test a single strip of silicon . . . to characterize the entire wafer." Mule'Stagno has done nothing to cure this deficiency other than to supplement his Expert Report with more conclusory opinions and diagrams.
Further, as SUMCO points out, although Mule'Stagno has supplemented his Expert Report with references to several secondary sources, he still has not achieved what the Court has asked him to do. For example, Mule'Stagno has not shown that the method upon which he primarily relies, Bulk Microdefect Density (BMD) testing, is generally accepted by or espoused by a recognized minority of scientists in his field as a reliable test to prove what Mule'Stagno is attempting to prove. In fact, although Mule'Stagno refers to articles by Obermeier, Lee, and Sasaki, his references are so vague that they provide no basis for the Court to conclude: (1) that the method used by the authors is practiced by a recognized minority of scientists in their field; or (2) that he has followed the same methodology in his own testing. In fact, SUMCO makes clear that Mule'Stagno has not followed Obermeier's methodology. See Mot. at 9 (noting that Mule'Stagno used a different temperature for his first heat treatment than Obermeier, used a longer second heat cycle than Obermeier, and used a different etchant than Obermeier). Further, as SUMCO also points out, Mule'Stagno tested slivers of the accused wafers rather than the whole wafers that were tested by Obermeier, Lee and Sasaki, and provides no support or justification for this methodological difference. Additionally, Mule'Stagno's assertion that BMD testing is reliable is significantly undermined by the fact that he admits, in his Sixth Declaration, that the BMD testing he performed yielded inconsistent results when performed by different operators at different times. See Mathiowetz Decl. at Ex. 14 at p. 6.
Mule'Stagno's descriptions of the other tests he has performed are similarly defective. For example, to support his Light Point Defect (LPD) and Atomic Force Microscopy (AFM) testing, Mule'Stagno's Second Declaration cites one article by Fujise as evidence of objective verification, but does not offer any evidence that the Fujise article or method is generally accepted by or espoused by a recognized minority of scientists in the field. And again, even if the Court had a basis to conclude that the Fujise protocol is generally accepted, it appears that Mule'Stagno did not follow that protocol. See Mot. at 7, 18 (noting that Fujise and Mule'Stagno used different instruments to perform the test). In addition, although Mule'Stagno states that "AFM can distinguish between pits in the [wafer] surface caused by polishing from pits in the surface caused by an agglomerated defect," the Fujise article does not discuss measuring agglomerated COP defects at all, and therefore does not support this conclusion.
Similarly, in support of his Flow Pattern Defects (FPD) testing, Mule'Stagno's Second Declaration cites an article by Yamagishi and three other articles that have cited to Yamagishi, but does not offer any evidence that the Yamagishi article or method is generally accepted by or espoused by a recognized minority of scientists in the field as a recognized test for revealing and identifying agglomerated vacancy point defects in silicon materials within the detection limit required by the '302 Patent. Moreover, there does not appear to be any support in the Yamagishi article or the other sources for Mule'Stagno's conclusion that the flow pattern activity he identifies in the accused wafers is due to "contamination" or "damage" rather than the presence of agglomerated vacancy defects. Further, the Yamagishi article and the other sources do not provide objective support for Mule'Stagno's methodology of scanning only one quarter wafer for regions of agglomerated defects. Similarly, in his discussion of Copper Decoration testing, Mule'Stagno cites articles by Kulkarni and Yamauchi, but does not aver that the protocols in the Kulkarni and Yamauchi articles are generally accepted by or espoused by a recognized minority of scientists in the field for visually detecting the presence of agglomerated intrinsic point defects.
Additionally, the Court does not have sufficient evidence upon which it can conclude that Mule'Stagno's Oxygen Induced Stacking Fault (OISF) testing is scientifically reliable. Although Mule'Stagno's Second Declaration cites articles by Harada and Dornberger, Mule'Stagno does not appear to have followed the methodologies employed by either Harada or Dornberger. Instead, it appears that he has mixed the two methods and has failed to provide any explanation as to why it was objectively valid to do so. Further, with respect to the Lifetime testing Mule'Stagno performed, Mule'Stagno's conclusory statements are so vague as to render the entire discussion of this methodology virtually meaningless. Indeed, in describing the relevancy of the test results, Mule'Stagno states only that "[t]he low oxygen levels contained in the accused wafers rendered these test result [ sic] less meaningful. While they are consistent with the conclusions I reached in my report, I did not rely on the lifetime test results to support those conclusions." Powers Decl. at Ex. 3 at p. 7. Based on this ambiguous remark, the Court has no basis to determine whether Lifetime testing is scientifically valid. Indeed, Mule'Stagno's quick dismissal of Lifetime testing appears to lead credence to SUMCO's suggestion that Lifetime testing does not support Mule'Stagno's ultimate conclusion.
Quite tellingly, MEMC's Opposition brief fails to respond to the vast majority of defects SUMCO identifies in its Motion, discounting them as "insignificant" because SUMCO fails to "show how any of these differences would in any way change the outcome of any of Dr. Mule'Stagno's tests." However, as MEMC concedes, since MEMC is the proponent of the expert testimony, it is MEMC — and not SUMCO — who bears the burden of establishing the expert's admissibility. Butler, 984 F.Supp. at 1260. MEMC has utterly failed to meet this burden. Accordingly, the Court GRANTS SUMCO's Second Motion to Exclude Expert.
MEMC also chooses to rehash arguments raised during claim construction and to debate, at length, whether SUMCO's definition of the term "axially symmetric" is proper. This argument is entirely irrelevant to the instant Motion, which focuses only on whether MEMC has sufficiently proven that Mule'Stagno's expert testimony is scientifically reliable and admissible under Daubert. Additionally, MEMC's attempt to rely on a previously undisclosed "expert," Dr. Werner Bergholz ("Bergholz"), is improper. Bergholz's declaration does not constitute the type of "objective source, such as a learned treatise, the policy statement of a professional association, a published article in a reputable scientific journal or the like" which this Court's Order required, and his testimony does not cite to any objective authoritative source of the type required by the Order. In addition, Bergholz does not describe any personal knowledge about the accused products or patent claim limitations, or explain why his experience with prior art silicon wafers applies to wafers produced by the Park/Samsung method. Further, it appears that Bergholz did not review or confirm the validity of Mule'Stagno's testing procedures of the accused wafers, or the resulting data, or the quality or accuracy of the conclusions reached by Mule'Stagno. Accordingly, Bergholz's testimony has been disregarded by the Court.
CONCLUSION
IT IS HEREBY ORDERED THAT Defendants' Motion to Exclude the Expert Report and Testimony of Plaintiff's Proposed Expert Luciano Mule'stagno [Docket No. 502] is GRANTED.
IT IS SO ORDERED.