Full opinion text
MEMORANDUM AND ORDER DOUGLAS P. WOODLOCK, District Judge. Plaintiff Cohesive Technologies produces equipment for high-performance liquid chromatography (“HPLC”), a process scientists use to separate a mixture of chemical compounds into its various components. Cohesive has patented an apparatus and method for increasing the efficiency of HPLC separations by increasing the rate of flow of solvent (the liquid in which the mixture of compounds is dissolved) through the HPLC column, which is filled with solid particles. These cases stem from allegations by Cohesive Technologies that the Oasis HPLC columns manufactured by Waters Corporation infringe Cohesive’s 5,772,874 ('874) and 5,919,368 ('368) patents. After a sixteen day trial, a jury in Civil Action No. 98-12308 returned a verdict for Cohesive finding infringement and upholding the validity of the '874 patent. Waters has moved for a judgment as a matter of law notwithstanding the verdict. Waters further alleges that the '874 patent is unenforceable because it was procured by inequitable conduct on the part of Cohesive. Cohesive claims that Waters’ infringement was willful and thus that the court should award enhanced damages and attorneys fees on that ground. After the jury verdict, I held a bench trial on the issues of inequitable conduct, willful infringement, and damages. Cohesive filed two separate civil cases related to Civil Action No. 98-12308. Civil Action No. 99-11528 alleges that the same columns that infringed the '874 patent infringe Cohesive’s '368 patent, a divisional patent of the '874 patent. In the second related case, Civil Action No. 01-12307, Cohesive seeks to enjoin Waters from making, selling, or using columns with 25-micron-sized particles, as opposed to the 30-micron-sized particles used in the accused columns in the other two cases. To provide a general context for discussion, a brief description of the technology at issue might be helpful. High-performance liquid chromatography is a form of column chromatography used frequently in biochemistry and analytical chemistry. HPLC is used to separate components of a mixture by using a variety of chemical interactions between the substance being analyzed and the chromatography column. An HPLC column generally consists of a “stationary phase” of solid particles, such as silica. Silica is a sand-like material to which other compounds adsorb with different affinities. Performing HPLC generally involves the elution of a mixture of compounds dissolved in a appropriate solvent, called the “mobile phase,” over the stationary phase. As the mobile phase elutes along the column, the compounds dissolved in the solvent react with and adsorb to the solid particles. Because of the variation in adsorption rates to the stationary phase, the components of the compounds elute at different times from the column, allowing the scientist to collect and analyze the separated components. The '874 patent, issued on June 30, 1998 to Cohesive’s Hubert Quinn and Joseph Takarewski, claims an apparatus for what Cohesive calls “turbulent flow” HPLC. The '368 patent issued on July 9, 1999 and claims the method for using the '874 apparatus. The invention purports to provide a means for efficient HPLC separations within shorter total elution or run times. Cohesive and Waters both sell hardware for liquid chromatography systems as well as the disposable or consumable columns that the systems require. In Part I of this memorandum, I will address the question of infringement and validity of the '874 patent raised by Waters’ renewed motion for judgment as a matter of law in Civil Action No. 98-12308. In Part II, I will address my factual findings and conclusions of law on the issues of inequitable conduct and willful infringement in that case. In Part III, I will address the questions of infringement and validity of the '368 patent raised by cross motions for summary judgment submitted by the parties in Civil Action No. 99-11528. In Part IV, I will address the question of summary judgment in Civil Action No. 01-12307, and, in Part V, I address the issue of damages. I. Post-Jury Trial Motions in Civil Action No. 98-12308 Regarding the '874 Patent In Civil Action No. 98-12308, Cohesive alleged that Waters infringed the '874 patent by its manufacture and sale of 1 x 50 mm Oasis HLB Extraction Columns containing Oasis 30 micron particles, and Waters counterclaimed that the '874 patent was invalid by reasons of anticipation and obviousness under 35 U.S.C. §§ 102 and 103. The issues of infringement and invalidity on the grounds of obviousness were tried to a jury, which returned a verdict in favor of Cohesive after I granted a directed verdict for Cohesive on the issue of anticipation under 35 U.S.C. § 102. Waters filed a motion for a judgment as a matter of law on the issue of infringement at the close of the evidence, which I denied. Waters has renewed its motion for judgment as a matter of law, adding a motion for a judgment as a matter of law on the issue of obviousness. A. Infringement 1. “Rigid” particles Waters first argues that Cohesive has failed to provide sufficient evidence to support a jury verdict that the particles in Waters’ columns are rigid. Every claim of the '874 patent requires that the accused device contain a chromatographic column with a bed of “rigid” particles. Thus, Waters’ OASIS HLB solid particles must be rigid in order for Waters to be liable for infringement. The jury found that Waters’ device did contain rigid particles. In the Markman hearing in this case, I construed “rigid” to mean “an object’s capacity to maintain substantially zero changes in density and volume under packing pressure of at least about 5000 psi and as a consequence substantially to resist plastic deformation under such pressure.” I further defined “plastic deformation” as “a change in the dimensions of an object under load that is not recovered when the load is removed.” I explicitly did not construe the term “rigid” to exclude polymeric particles (as opposed to the monomeric particles referred to in the preferred embodiment of the '874 patent). Quinn’s trial testimony provides sufficient evidence that the particles used in the accused Waters columns are rigid. He testified on direct examination that the Waters’ particles met the claim construction above: Q: If rigid means an object’s capacity to maintain substantially zero changes in density and volume, under packing pressure of at least about 5,000 psi, and as a consequence, substantially to resist plastic deformation under such pressure, do the Waters’ particles qualify as rigid? A: Yes, they do. Q: How do you know? A: Because one of the things that I did was to take the Waters particles and pack them at 5,000 psi, and the particles passed that test because they didn’t plug up or they didn’t crush or they didn’t collapse. Tr. Day 4 at 38. Cohesive’s attorney continued to examine Quinn on his observation that the particles did not collapse at 5,000 psi or greater: Q: Now, when you packed the particles at 5,000 psi, the Oasis particles, did they maintain substantially zero changes in density and volume under the packing pressure? A: Yes, they did. Q: And did they crush or deform or exhibit any plastic deformation under that packing pressure? A: No, they did not. Q: And when you packed them at 10,-000 psi, did they exhibit any crushing or collapsing or plastic deformation then? A: No, they did not ... There was no trace of the particles crushing or compressing or plugging up. There was no — I saw no evidence of that. Id. at 39-41. Quinn further testified that, “if the particles collapse, they will shut off all the flow channels, and you won’t be able to get any flow through the column.” Id. In other words, Quinn testified he was able to determine whether the particles collapsed under pressure by simply observing the flow of the packed column. I determined in a three-day Daubert hearing before trial that Quinn was a qualified expert and used scientifically reliable methods to generate his expert opinion. After post-trial review, I similarly conclude that Quinn’s testimony provides sufficient evidence for a reasonable fact finder to determine that the Waters particles are rigid within the meaning of the '874 patent. Waters does not dispute that Quinn’s method of observing the particles to determine rigidity is scientifically reliable. Waters argues instead that Cohesive provided no evidence of the first part of a two part definition of “rigid.” More particularly, Waters asserts that in order for Cohesive to prove that Waters infringed any claim of the '874 patent it must provide evidence that the particles (1) maintain substantially zero changes in density and volume under pressure of at least about 5000 psi, and (2) substantially resist plastic deformation at that pressure. It argues that Cohesive has provided evidence of (2) but not (1). A particle can compress under pressure, Waters asserts, but still seem as though it resists plastic deformation if it quickly regains its structure when pressure is released. Waters likens a particle that satisfies only element (2) of the definition to a Nerf ball, which would collapse under high pressure but immediately spring back to its original shape when normal pressure resumes. Although I agree with Waters that Cohesive must provide evidence of both factors in order to meet the claim construction of “rigid,” I find that Cohesive has done so. As stated above, Quinn testified that he observed no compression while applying 5,000 psi and 10,000 psi of pressure. Waters argues that the testimony is conclusory and thus not sufficient evidence of rigidity. But Quinn’s testimony was not merely conclusory. He stated that the proper test for detecting compression was to observe the particles for plugging or interruption in flow rate. He testified that he saw no evidence of compression during his experiment. Waters argues that its own expert’s testimony that the Waters particles collapse under pressure (Tr. Day 11 at 48-68) combined with testimony from expert Savaria that polymeric particles are generally more compressible than monomeric particles (Tr. Day 8 at 53) militate in favor of overturning the jury’s verdict on infringement. Regardless of any evidence that rebuts Quinn’s conclusion, however, Quinn’s testimony was sufficient to create a material dispute of fact for the jury to resolve. Thus, there is no basis for a judgment as a matter of law with respect to whether the particles in Waters’ accused device are rigid. 2. “Interstitial Volume Fraction of at least 45%” Waters claims that Cohesive’s calculation of interstitial volume fraction of 45% or more, which is a limitation of claims 1, 3, 7, 9, and 20 of the '874 patent, was not supported by reliable data. Effectively, Waters renews its Daubert challenge to Quinn’s testimony regarding his calculation of the interstitial volume of Waters particles. Interstitial volume fraction is the volume of the space between the particles in the column divided by the total volume of the column. The definition of interstitial volume does not include the volume of the space in the pores in the solid particles, but only the volume between each particle. Thus, the interstitial volume fraction is equal to the volume of the column minus the volume of the particle pores minus the volume of the skeletons of the particles, expressed as a percentage of the total volume of the column. In order to support a claim of infringement of the '874 patent, Cohesive must provide sufficient evidence that Waters’ column contains an interstitial volume of 45% or more. Cohesive has produced sufficient and reliable evidence that Waters’ column has an interstitial volume of 45% or greater. Cohesive’s Quinn measured the interstitial fraction of the Waters column by a method called size exclusion chromatography (“SEC”). See Tr. Day 4 at 87-88. (“I knew the interstitial fraction because I determined it by size exclusion.”) (“I wanted to use an independent method [to calculate the interstitial fraction] which was size exclusion chromatography.”) Size exclusion chromatography allows one to calculate interstitial volume directly by flowing variously sized molecules, dissolved in solvent, through the column and measuring the volume it takes to elute them (“elution volume”). The larger the molecule, the less likely it will fit into the pores of the particles and the more quickly the molecule will pass through the column. The smaller the molecule, the more likely it will adhere to the pores in the particles and the more volume it will take to elute it from the column. By determining the various elution volumes for different sized molecules in a class, one can deduce the pore volume, the interstitial volume, and thus the interstitial fraction as a percentage of the entire column. Quinn testified that he could not run the size exclusion chromatography on Waters’ accused column — a 50 x 1 mm column— because his equipment was not precise enough to measure the volume changes accurately in such a small column. See Tr. Day 4 at 66. Instead of using Waters’ 50 x 1 mm column, Quinn packed a larger 2.1 x 100 mm column himself with Waters’ Oasis 30 micron particles. He then used size exclusion chromatography to determine the interstitial volume fraction of the larger column. He determined, using this method, that the interstitial volume was about .46, or 46%. Waters claims that (1) Quinn failed to use the correct value for the density of Waters particles to determine the interstitial volume fraction, and (2) Quinn’s 2.1 x 100 mm column that he packed with Waters particles failed catastrophically and thus any resulting calculations were unreliable. Waters points to Quinn’s testimony under cross-examination regarding the density of Waters’ particles: Q: And if your measurements are correct, then according to your calculations, the density of the OASIS particles should be about 1.8? A: Yes, according to these measurements. Q: Do you know what the density of OASIS 30 micron HLB particles is? A: Yes. Q: What is it? A: It’s around 1.2. Q: 1.2. But your measurements — your SEC measurement that you’re relying on in this case show a density of 1.8, don’t they? A: Yes, they do. Q: So there’s something wrong with your numbers? A: Yes, there is. Q: And these are the numbers you are relying on in this suit? A: Mr. Scofield, I’m not relying on those numbers for the interstitial fraction. Tr. Day 5 at 100-01. Waters also refers to the cross-examination testimony regarding “a void” in the 2.1 x 100 mm column that Quinn packed: Q: So the column that you are relying upon for the SEC measurements and the empirical derivation of the Kozeny constant of 270 that you wish to apply in this case, that column, that one single column that you used, is a column that experiences a catastrophic problem and a void? A: Yes, after the SEC measurements were made, correct. Tr. Day 6 at 86 In 16-21. Neither of Waters’ arguments are persuasive. Quinn did not rely upon his (admittedly inaccurate) particle density calculation to determine the interstitial volume of the Waters column. He used an “independent” method of size exclusion chromatography which involved measuring only the various elution volumes of different sized molecules passed over the Waters particles. It is true that, in order to check his work, Quinn attempted to use his calculation for interstitial volume to calculate a constant in the Ergun equation. Quinn testified that one way to measure interstitial volume indirectly is by utilizing a relationship expressed in the Kozeny-Carman equation. The “pressure drop” of a column is related to the interstitial volume of the column by a set of equations, expressed first in Darcy’s law, Tr. Day 4 at 77-78, and then further refined by scientists Kozeny, Carman, and Blake. Id. at 78. The resulting equation that applies to “turbulent flow,” as opposed to laminar flow, chromatography is called the Ergun equation. Tr. Day 4 at 81-83. Quinn stated that the Ergun equation contains several parameters that one needs to determine empirically: the length of the column, the particle size, and the flow rate, for example. Tr. Day 4 at 80 In 3-7. In addition, one must also determine the value of a numerical constant, which Quinn called the Kozeny constant. Id. at 80 In 8-15. Quinn did not use the Ergun equations to calculate the interstitial volume. Rather, he used the interstitial volume that he measured empirically to derive the Kozeny constant. The purpose of doing this, he testified, was to double check his work. He used size-exclusion chromatography to calculate the interstitial volume fraction, plugged the interstitial volume fraction into the equation with the other parameters, such as pressure drop, and determined that the Kozeny constant was 270. This matched, according to Quinn, the well-known value of the Kozeny constant for his particular experiment. See Tr. Day 4 at 88-89. Quinn testified on cross-examination that he did not rely upon the density of the particles or the Kozeny constant to determine the interstitial volume fraction: The density measurements that you’re referring to does not have anything to do with the determination for interstitial fraction that I made in this size exclusion. Because the interstitial fraction was made directly by me. Okay. And it doesn’t depend on any value for the density. It is independent. It was determined directly. I simply included those numbers there for additional information, but those numbers had no bearing whatsoever on my determination of the interstitial volume. Tr. Day 5 at 101 In 20. See also Tr. at 88 In 19-23. (“I did a ratio and calculated the interstitial fraction without the use of any Kozeny constant because they drop out in this analysis.”). Waters contends that in back-calculating the Kozeny constant, Quinn used an erroneous measurement for the density of the particles. Whether Quinn used an erroneous measurement in calculating the constant is unclear from the testimony. But even if the Kozeny constant he calculated was inaccurate, Cohesive has still provided enough evidence that Waters’ interstitial fraction volume is greater than 45%. Quinn stated several times during cross-examination that he used an independent method of size-exclusion chromatography to achieve that result. Waters does not contend that size exclusion chromatography is an unreliable way of determining interstitial volume fraction. Thus, Quinn’s testimony is sufficient evidence that the interstitial volume fraction of the Waters column is greater than 45%. Waters’ second argument — that Quinn’s measurement of the interstitial volume is unreliable because the size-exclusion chromatography column failed — is also unavailing. Quinn testified that the failure of the 2.1 x 100 column occurred after he measured the interstitial fraction. Furthermore, the fact that the column was a different size than Waters’ accused column is likely immaterial. Interstitial volume is a percentage of total volume of the particles. There is no reason based upon this record to assume that the percentage volume of the space between particles depends upon the size of the column. Waters provides no evidence that the size-exclusion chromatography experiment conducted by Quinn was based upon unreliable experimentation or principles. The possibly erroneous calculation of the Kozeny constant notwithstanding, Cohesive has provided sufficient evidence to support a jury verdict that Waters’ accused device infringes the interstitial volume fraction limitation. 3. Interstitial channels with mean cross-section dimensions not less than 5 microns. Waters argues that Cohesive failed to prove infringement of claims 3, 16, 20 that require interstitial channels with mean cross-section dimensions not less than 5 microns. On direct examination, Quinn testified that “if you have particles that are greater than 30-micron and you have an interstitial fraction that is greater than 45%, then the spacing between the particles will not be less than 5-micron.” Tr. Day 4 at 101. Waters argues that because Cohesive did not provide adequate evidence of the interstitial fraction being greater than 45%, there is no evidence of cross-section dimensions of less than 5 microns. As detailed above in Section 1.A.2, because Cohesive produced enough evidence to support a jury verdict that Waters’ columns have an interstitial volume fraction of greater than 45%, Quinn’s testimony on direct is sufficient evidence to support a verdict of infringement regarding the size of the cross-section dimensions as well. B. Validity The jury found that Waters had not established by clear and convincing evidence that the '874 patent is invalid because the claimed invention was obvious. Waters asserts that the verdict should be overturned because it is not supported by substantial evidence. This contention stumbles at the threshold of the protocol for post-verdict review. Waters waived this argument by failing to move for a judgment as a matter of law with respect to obviousness of the '874 patent at the close of all the evidence before the case was submitted to the jury. Although Waters did move for judgment as a matter of law at the close of the evidence, it moved with regard only to the issue of infringement. As is well-established, “a motion for judgment as a matter of law made at the close of all the evidence preserves for review only those grounds specified at the time, and no others.” See Correa v. Hospital San Francisco, 69 F.3d 1184, 1196 (1st Cir.1995). In Correa, the defendant made a post-trial motion for a judgment notwithstanding the verdict on a legal theory it had failed to preserve at the close of the evidence. The court stated that: [A] renewed motion for judgment as a matter of law under Fed.R.Civ.P. 50(b) is bounded by the movants’ earlier Rule 50(a) motion. The movant cannot use such a motion as a vehicle to introduce a legal theory not distinctly articulated in its close-of-evidence motion for a directed verdict. Id. Waters failed to move for judgment on the question of obviousness at the close of all the evidence. Thus, it cannot pursue that theory on a renewed motion. Waters contends that because “[o]bvi-ousness is a question of law, and is thus freely reviewable on appeal,” Stiftung v. Renishaw PLC, 945 F.2d 1173, 1182 (Fed. Cir.1991), I should review their motion for judgment as a matter of law regarding obviousness even though they failed properly to preserve it. Waters cites Sjolund v. Musland for a similar proposition. See 847 F.2d 1573, 1576 (Fed.Cir.1988) (“[Wjhere a jury is asked to decide a question of law which depends on underlying factual determinations, the jury’s legal conclusion is subject to independent review by the court on appeal.”). These cases discuss the substantive standard for reviewing an obviousness determination on appeal. They are inapposite to whether Waters preserved its argument as is required by procedural law in this circuit. See Correa, 69 F.3d at 1196. Waters cites no case in which a district court judge independently reviewed a jury verdict of non-obviousness after the alleged infringer had waived the argument earlier in the proceedings. And Waters does not dispute that it failed to move for a judgment at the close of the evidence regarding obviousness. Waters is foreclosed from presenting its obviousness contention in a “renewed” motion to overturn the jury verdict. C. Conclusion Because Cohesive has provided sufficient evidence to support the jury verdict in Civil Action No. 98-12308, I will deny Waters’ renewed motion for judgment as a matter of law notwithstanding the jury verdict. II. Bench Trial: Findings and Conclusions Regarding Inequitable Conduct and Willful Infringement A. Inequitable Conduct Waters contends that Cohesive’s patent is unenforceable due to inequitable conduct because the applicants for the '874 patent omitted and misrepresented material facts in their application to the Patent and Trademark Office (PTO). In order to prevail, Waters must show by clear and convincing evidence that one of Cohesive’s applicants made a misrepresentation, that the misrepresentation was material, and that it was made with intent to deceive. See Honeywell Intern. Inc. v. Universal Avionics Systems Corp., 488 F.3d 982, 999 (Fed.Cir.2007). After a bench trial on this issue, I find and conclude that Waters has failed to produce clear and convincing evidence that Cohesive intended to deceive the patent examiners. The following constitutes my findings of fact and conclusions of law on this issue pursuant to Fed. R.Civ.P. 52. 1. Facts The following findings of fact are based upon the live testimony of Hubert Quinn and Robert Schiller, and further affidavit testimony of Hubert Quinn. a. Patent Prosecution History Summary Hubert Quinn and Joseph Takarewski are the named inventors on '874 and '368 patents, both of which are assigned to Cohesive. The application for both patents (the Quinn application) was based on one specification, and was filed originally on November 2, 1995. Claims 1-15 of the '874 invention include a limitation that requires “loading ... solute through [the column] ... at a reduced velocity greater than about 5,000.” Another limitation contained in Claim 1 of the '874 patent requires “an interstitial volume between particles of not less than about 45% of the total volume of said column.” ('874 patent Col. 20). Although none of the claims mention the term “turbulent flow,” Cohesive distinguished the claimed invention from the prior art in the patent application on the basis that the invention provided for turbulent, as opposed to laminar, flow. Cohesive described its invention as, “[a] method and apparatus for liquid chromatography at improved analytical and preparative speeds and quantities that involves flow of fluids through the column at flow rates sufficient to induce turbulent flow in those fluids.” Cohesive also referred to its invention in the patent application and specification as “turbulent flow chromatography.” By utilizing higher flow rates and pressure, the invention was designed to allow chromatographic separations to be performed with greater accuracy and speed than would laminar flow technology. In its application, Cohesive cited the Af-eyan '270 patent (U.S. Patent No. 5,019,-270) as an instance of prior art. That art describes a method of chromatography that provides higher speed chromatographic separations through the use of higher velocity flow. Cohesive also cited an article by Pretorius and Smuts entitled “Turbulent Flow Chromatography: a New Approach to Faster Analysis,” Analytical Chemistry, vol. 2, pp. 274-80 (1966) (“Pretorius reference”). In the original application, Cohesive stated that, “[t]he use of turbulent flow liquid chromatography would appear to be contraindicated by the teachings of Pretorius, et al., who cites at page 280, col. 2, that a separation in a tube with a diameter of 0.1 cm and a length of 2000 cm ... would apparently take 24 days.” Waters asserts that Cohesive’s statement misquoted the Pretorius reference, page 280 col. 2, which actually states that analysis time can be as short as 100 seconds under turbulent flow conditions, and says in the next sentence that a separation would take up to 24 days under laminar flow conditions. Cohesive does not deny that it misquoted the reference, but claims the misrepresentation was unintentional. The PTO rejected that original application as anticipated or rendered obvious by the Afeyan patent, but stated it was non-obvious with respect to the Pretorius reference. On June 11, 1996, Cohesive filed a continuation-in-part application (Quinn CIP application), which was ultimately approved. The Examiner initially rejected the CIP application on May 1, 1997, as anticipated by or rendered obvious by the Afeyan patent. In response to this action, Cohesive filed an Amendment B and Remarks dated July 24, 1997, arguing that the Afeyan patent did not anticipate or render obvious the Quinn invention. The examiner rejected all claims in this continuation-in-part of the original application on September 10, 1997, citing the Afeyan patent. Cohesive then filed another Amendment on December 31, 1997 (Amendment After Final), along with a declaration from Professor Georges Guiochon, an expert in chromatography. It is from the Amendment in Final that the Quinn patents were ultimately issued. b. The Guiochon Declaration The Guiochon declaration, which was included in the Amendment After Final, stated that the Quinn invention was novel and non-obvious: The apparatus and method were certainly not obvious to me at the time. I concluded then and am still of the same conclusion ... that the invention described in the '367 application was and is an extremely important, ingenious, and unobvious advance in the art of chromatography. ... I am not aware of any prior art that possesses the combination of elements of the present invention including the interstitial volume in excess of 45% combined with rigid, solid, porous particles. See Guiochon Deck, Def. Exhibit 9 at 4. Guiochon avoided the term “turbulent flow” when describing Quinn’s invention. His declaration stated that the “essence of the invention described in the '367 application as I view it is that the flow through the interstitial channels between the particles in the chromatographic column is at an abnormally high rate for such a column,” see Guiochon Decl. page 4, and that “a reduced velocity greater than about 5,000 for the liquid flow is believed to be sufficient to trigger the observed phenomenon.” Schiller admitted that in previous drafts of the Guiochon declaration, these two statements read “between the particles in the chromatographic column necessarily turbulent as opposed to laminar” and “... sufficient to induce turbulent flow," respectively (emphasis supplied). Schiller testified that he changed the sentences to omit any reference to the term “turbulent” because Guiochon “had reservations as to whether the flow was definable within the meaning of the word turbulence, which was up for a great many different interpretations.” See Bench Trial Tr. Day 5 at 57 In 7-9. In addition, Cohesive argued in the Amendment After Final, citing Guiochon, that the Afeyan patent taught methods using laminar flow, and distinguished the Quinn invention as teaching non-laminar, or turbulent flow. Cohesive relied on statements in the Guiochon declaration that, for example: “the intraparticle and interparticle flow described in Afeyan can only be considered laminar.” (Guiochon Decl. pg 5.); “Ramsay confirms the use of laminar flow ...” (Guiochon Decl. pg 6); “the [Nilsson] patent never teaches or suggests the use of any high value of reduced velocity nor of any turbulent flow characteristics of the particles.” (Guiochon Decl. pg 7). Guiochon’s submitted declaration, of course, never refers to the Quinn invention as turbulent flow chromatography, though he stated that the '367 method has “turbulent flow characteristics.”. See Guio-chon Decl. at 7. Although Cohesive’s patent application calls the Quinn invention “turbulent flow” technology, Cohesive failed to disclose that Guiochon had reservations about using the term “turbulent” to describe the flow Quinn created. Indeed, Guiochon told Cohesive’s patent attorney, Schiller, via letter prior to submitting his declaration that he did not think that the flow used in the Quinn invention was technically turbulent flow: It does not bother me to say that X or Y ‘does not teach the use of turbulence’ as long as it is true that they do not (that would be something else if Pretorius was a reference but he is not). Still, I do not believe that the flow is turbulent under the conditions stated in the patent. I still do not understand what is happening, but, as I understand it, you can patent something that you have done and can reproduce, whether you understand it or not, right? (Guiochon faxed letter to Schiller, 12/15/97, Def. Ex. 19.) Moreover, in his Declaration, Guiochon stated that he had previously been of the opinion that “turbulent flow in liquid chromatography would not work,” and that he was “very surprised” by the data that Quinn gave him regarding the invention. (Guiochon Decl. pg 3-4.) Guio-chon then edited his declaration to omit any reference to the Quinn invention as turbulent flow, preferring to call it “the Quinn effect.” Cohesive did not inform the patent examiner that Guiochon had made these edits. The Guiochon Declaration also included a calculation of “maximum reduced velocity” of 1,390 for the Afeyan patent — less than half the minimum value of the velocity claimed in the Quinn invention. In the calculation, Guiochon used what he stated was a “reasonable value of about 1 x 10 " for the diffusion coefficient.” Waters argues that if Guiochon had used the accurate diffusion coefficient, he would have calculated a maximum reduced velocity in the Afeyan patent of 139,000 to 1,390,000, a number well above the minimum 5,000 claimed in the Quinn invention. Guiochon’s declaration also included a statement that “[ajnother characteristic of the present invention that appears to me to be unique is the provision of an interstitial volume of 45%.” Guiochon Decl. at 5. Waters alleges that both Guiochon and Quinn knew that the Pretorius Patent Nos. 3,493,497 (the '497 patent) and 4,208,284 (the '284 patent) each disclosed chromatography columns with an interstitial volume greater than 45%, and thus that Guio-chon’s statement was a misrepresentation. Cohesive and Waters do not dispute the general facts of the prosecution history. The sole dispute is over whether Cohesive’s omissions and alleged misrepresentations were material, and, if so, whether there is clear and convincing evidence that Cohesive intended to deceive the patent examiner. 2. Discussion Waters asserts that Cohesive engaged in inequitable conduct by (1) stating that “turbulent flow” was contraindicated in the Pretorius and Smuts article, and then, in support, misquoting page 280, col. 2 of that reference by stating that a chromatographic separation would take 24 days rather than 100 seconds, (2) failing to disclose the opinion of Guiochon that the Quinn invention did not teach turbulent flow and that turbulent flow was not novel and had been shown in the prior art by Pretorius, (3) falsely stating that an interstitial volume of 45% was “unique” in the art, knowing that the Pretorius patents disclosed that element, and (4) inaccurately representing that the Afeyan patent taught a maximum reduced velocity of 1,390, a calculation which relied upon an erroneous particle diffusion coefficient. Patent applicants have a duty to act with good faith, honesty, and candor when submitting their applications to the Patent and Trademark Office. See Honeywell Intern. Inc. v. Universal Avionics Systems Corp., 488 F.3d 982, 999 (Fed.Cir.2007). If a patent applicant breaches this duty, an otherwise valid patent becomes unenforceable. See id.; J.P. Stevens & Co., Inc. v. Lex Tex Ltd., Inc., 747 F.2d 1553, 1559 (Fed.Cir.1984) (stating that inequitable conduct is broader than common law fraud and “includes failure to disclose material information, or submission of false material information, with an intent to mislead.”). The standard of proof for finding inequitable conduct is a high one; there must be clear and convincing evidence that Cohesive (1) failed to disclose material information or submitted false information in its dealings with the PTO, and (2) that it did so with the intent to deceive the patent examiner. J.P. Stevens & Co., Inc., 747 F.2d at 1559. Gross negligence does not suffice as intent to deceive, Kingsdown Med. Consultants, Ltd. v. Hollister, Inc., 863 F.2d 867, 876 (Fed.Cir.1988), and materiality for purposes of this determination means that there is a substantial likelihood that a patent examiner would have considered the omitted information important in deciding whether to allow the application to issue as a patent. See Molins PLC v. Textron, Inc., 48 F.3d 1172, 1179 (Fed.Cir.1995). The nondisclosure or misrepresentation to the patent examiner must meet threshold levels of both materiality and intent. Id. at 1178. Once the threshold levels of materiality and intent have been established, the trial court must weigh materiality and intent to determine whether the equities warrant a conclusion that inequitable conduct occurred. Id. The more material the information misrepresented or withheld by the applicant, the less evidence of intent will be required in order to find inequitable conduct. Honeywell Intern., Inc., 488 F.3d at 999 (citing N.V. Akzo v. E.I. DuPont de Nemours, 810 F.2d 1148, 1153 (Fed.Cir.1987)). If the court finds inequitable conduct, it must hold that the patent claims at issue are unenforceable. a. Misrepresentations regarding the Pretorius and Smuts references Cohesive stated in its patent application that the Pretorius, et al., reference “contraindicated turbulent flow in liquid chromatography.” Waters asserts that in support of this conclusion, Cohesive intentionally misquoted the Pretorius reference as concluding that a separation using turbulent flow in liquid chromatography would take 24 days, when the article actually says 100 seconds. Cohesive admittedly did misquote the reference. And the Pretorius reference does, indeed, state that high rates of turbulent flow in liquid chromatography could reduce separation time to 100 seconds (as opposed to 24 days for laminar flow.) The patent examiner would likely have relied on this fact in order to determine whether the Quinn invention, which touts high-speed separations, was anticipated by the Pretorius patent or otherwise rendered obvious by it. Thus, I find the misquote would have been relevant to the patent examiner. Notwithstanding the Pretorius results, however, I found as a matter of law in Civil Action No. 98-12308 that the reference did not anticipate the '874 patent. For its part, the jury supportedly found that the Pretorius reference did not render the Quinn invention obvious. Therefore I conclude the Pretorius misquote is not so material as clearly to render Cohesive’s patent anticipated or obvious as a matter of law. Even though the misstatement regarding the Pretorius result was relevant and arguably material, I find no evidence of Cohesive’s intent to deceive the patent examiner. The patent application included an extended discussion of the details of the prior art. The misstatement came from one paragraph in a long article that included both the timed separation results for laminar and turbulent flow in liquid chromatography, one right after another. The two results could reasonably have been transposed unintentionally. Without more evidence that Cohesive misrepresented the Pretorius results intentionally, I conclude that the misquote was a negligent error. b. Omission of Guiochon’s Opinion that the Quinn Invention Did Not Teach “Turbulent Flow” Waters asserts that Cohesive intentionally failed to disclose Guiochon’s opinion that the Quinn invention did not teach what he would call “turbulent” flow in order to deceive the patent examiner. Because I credit Cohesive’s testimony that it did not believe Guiochon’s definitional distinction was material to the ultimate conclusion of his declaration, I do not find clear and convincing evidence of Cohesive’s intent to deceive the patent examiner. Guiochon stated in the declaration that the Quinn invention was non-obvious, not anticipated by prior art, and “ingenious.” He stated that he was surprised by the efficiency of the separations that Quinn achieved with abnormally high-flow conditions, and described how the prior art had only appeared to teach “laminar flow” rather than flow with “turbulent characteristics.” It is true that Cohesive never disclosed to the patent examiner Guiochon’s opinion that the flow in Quinn’s invention was not truly “turbulent.” Cohesive does not dispute this fact. “Turbulent flow,” however, is not defined in the specification, nor is it mentioned in any of the actual claims of either the '864 or '368 patents. Schiller testified that it “never crossed his mind” that the patent examiner might reject the application based upon Guiochon’s reluctance to call the Quinn invention “turbulent flow.” Indeed, the patent examiner issued the patent despite the fact that Guiochon’s declaration does not characterize the Quinn invention as using turbulent rather than laminar flow. The essential conclusion of the Guiochon declaration was that he found the Quinn invention to be novel and non-obvious compared to laminar flow chromatography and the prior art, regardless of whether he would employ the technical label of “turbulent.” He stated in his deposition testimony: There are two things. There’s the Quinn effect and the interpretation of the Quinn effect. I consider the Quinn effect as real. I don’t agree with Mr. Quinn in the interpretation of the Quinn effect and I have a good one for myself. What I understand — it’s irrelevant to the existence of the Quinn effect. Guiochon Dep. at 46-47. Guiochon’s opinion that the flow was not definitionally “turbulent” was arguably relevant, in that a patent examiner could have considered it in determining whether the Quinn invention was anticipated by prior art. On the other hand, the patent specification makes it clear that the inventors merely theorized that turbulent flow was the cause of the increased efficiency and speed of chromatographic separations. See '874 Patent at 8 (“It is believed that under such conditions, turbulent flow of the mixture is induced ... ”). The essence of the invention was the enhanced performance of the chromatographic separation, and not the theory behind it. Thus, although I find Guiochon’s definitional reservations arguably relevant to a full discussion of invention, I do not find the patent examiner’s overall determination that the invention was novel would have changed had the drafting history of Guio-chon’s declaration on this point been disclosed. No doubt in hindsight it is apparent that it would have been better practice to have informed the patent examiner of Guio-chon’s opinion. Nonetheless, I cannot find by clear and convincing evidence that Cohesive’s omission constituted intentional deceit. There is a competing explanation for Cohesive’s omission, which is that it believed the Guiochon’s reluctance to label something as technically “turbulent” was a matter of mere semantics. While this competing explanation, is self-serving, it vitiates the effort to show intentional misrepresentation to the demanding clear and convincing standard. As a result, I find no inequitable conduct with regard to the material omission. For similar reasons, Cohesive’s failure to disclose Guiochon’s opinion that the Pre-torius reference taught turbulent flow has likewise not been shown to be intentional. Guiochon’s declaration stated that the Quinn invention was not anticipated or rendered obvious by prior art. Guiochon clearly knew of the Pretorius reference; he identified it in his faxed letter to Schiller. Waters claims that this faxed letter is evidence that Guiochon thought the Preto-rius reference taught turbulent flow in liquid chromatography. But it is at most unclear from the Guiochon fax whether he believed that the Pretorius reference taught turbulent flow in liquid chromatography. And even if Guiochon considered the Pretorius reference to teach so-called “turbulent flow,” in liquid chromatography, it is far short of establishing clear and convincing evidence of Cohesive’s deceptive intent. Cohesive reasonably believed that Guiochon did not consider turbulent flow definitionally the same as the flow created in the Quinn invention. If Guio-ehon did believe that the Pretorius reference taught turbulent flow, it is consistent with his conclusions that (1) the Quinn effect is not technically turbulent flow, and (2) the prior art does not anticipate or render obvious the Quinn invention. Moreover, the patent examiner had determined well before Cohesive filed its Amendment that the Pretorius reference, the title of which included the term “turbulent flow,” did not render the Quinn invention obvious. Thus, I find that (1) there is no showing by clear and convincing evidence of deceptive intent on the part of Cohesive by failing to disclose Guioehon’s opinion that the Pretorius reference taught turbulent flow, and (2) it is unlikely the disclosure would have had a material effect on the patent examiner’s disposition of the patent application. c. Alleged misstatement that 45% interstitial volume is “unique” Guiochon also stated in his declaration that an interstitial volume of 45% “appears to me to be unique.” Waters contends that this statement was false, and that both Quinn and Guiochon were aware that several prior art references, specifically the Pretorius patents, disclosed an interstitial volume of greater than 45%. Waters also claims that this information was highly material. Guiochon did state in his declaration that a 45% interstitial volume fraction was unique to the Quinn invention, but he made the statement to distinguish it from the Afeyan technology: Another characteristic of the present invention that appears to me to be unique is the provision of an interstitial column of 45% claimed in claims 1 and 36 of the '367 application, and which appears to contribute to the capability of the invention to provide the requisite high reduced velocity of 5000 or more. Afeyan et. al. teach using a column with a ‘void fraction of 0.35" (Col. 15, lines 31-32) which is far below the claimed interstitial volume of 45%. Guiochon Decl. at 5. That paragraph is the final paragraph of a section which discusses only the Afeyan references. There is no evidence that Guiochon was trying to distinguish the Pretorius references on this ground. Defendant does not claim that Afeyan teaches an interstitial volume of greater than 45%. Furthermore, the patent examiner had already determined that the Quinn invention was novel over the Pretorius patents in an earlier application; the Amendment with which Guio-chon’s declaration was filed was a response to, among other things, the novelty of the Quinn invention over the Afeyan patent. Thus, I do not find clear and convincing evidence that Guiochon, and through him Cohesive, made a material representation in this paragraph, or, if he did, that he did so intentionally. d. Misrepresentations Relating to the Maximum Reduced Velocity Disclosed in the Afeyan Patent Waters claims that Guiochon also miscalculated the “maximum reduced velocity” disclosed in the Afeyan '270 patent to be 1,390, using what his declaration called a “reasonable value of about 1 x 10 ~ for the diffusion coefficient.” Guiochon Decl. at 5. Waters contends that this diffusion coefficient was not a proper coefficient to use given the large biomolecules disclosed in the Afeyan patent, which have diffusion coefficients orders of magnitude smaller, in the range of 1 x 10 “ or 1 x 10 ~. Guiochon stated in his declaration that the Afeyan patent characterized a “bed velocity” of 5,000 cm/hr to be “very high.” He then stated that bed velocity, which is a linear velocity, must be scaled by the diffusion coefficient and particle size to determine what the corresponding reduced velocity is. According to Waters, his resulting calculation of the corresponding reduced velocity was understated by a factor of 100 to 1,000. There is no clear and convincing evidence that the Guiochon declaration contained a miscalculation of the reduced velocity that corresponds to a bed velocity of 5,000 cm/hr. Guiochon did indeed cite a diffusion coefficient of 1 x 10 ~ in his declaration. See Guiochon Decl. at 5. There is no evidence, however, of what the units of his coefficient were or which mathematical formula Guiochon used to determine the reduced velocity. The only evidence of the precise formula by which to translate linear bed velocity into reduced velocity is provided by the Defendant in demonstrative exhibit, a worksheet with hand-written calculations hypothesizing Guiochon’s purported reduced velocity determination. Def. Ex. 10. Although the exhibit provided by Defendant supports the hypothesis that Guiochon used the same equation in his calculations, I do not find this unauthenticated scratch sheet to be clear evidence that Guiochon in fact relied upon that equation. The testimonial evidence offered by Defendant also does not provide clear and convincing evidence that Guiochon made a mistake. Quinn testified in his deposition that, although he believed large biomolec-ules to have a diffusion coefficient of 10 ~7 to 10 ~8 cm2/sec (Quinn Dep. at 159), he could not confirm whether Guiochon’s calculation was inaccurate. He stated: I am not going to correct Professor Gui-ochon’s calculation, and I’m going to tell you why. In his calculation, I don’t know what assumption that he used for, A, the bed velocity, for the porosity of the column, et cetera. There are a number of other factors that go into the estimation of linear velocity, so therefore I’m not even going to — I’m not going to recalculate these numbers, because I may not use the same assumptions he did. Quinn Dep. at 162. I find no demonstration by clear and convincing evidence that Guiochon’s declaration contained a misrepresentation as to the reduced velocity associated with a linear velocity of 5,000 cm/hr. Even if Guiochon’s declaration did contain an inaccurate calculation of the reduced velocity that corresponds to a linear velocity of 5,000 cm/hr, it is not likely that this calculation would have been material to the examiner’s novelty determination. Guiochon stated independently of his calculation of the reduced velocity that, “I have found no teaching or suggestion whatever in Afeyan et. al., of the term ‘turbulence’ or a variant thereof, and liquid flow rates that would result in a reduced flow rate greater than 5,000, and/or means for producing the same.” Guiochon Decl. at 5. Furthermore, the Defendant does not indicate where the Afeyan patent discloses a method for chromatography at a reduced velocity of 5,000. The Afeyan reference does refer to a “very high” bed velocity of 5,000, but only to demonstrate the flaws of the old technology. The Afeyan reference states only that 5,000 cm/hr is a bed velocity that would lead to severe “premature solute breakthrough” in conventional chromatography. See Afeyan Patent at column 15, line 42. This statement does not support the converse conclusion, that is, that the new Afeyan technology teaches a method to solve this issue at a velocity of 5,000 cm/hr. The Defendant has not clearly demonstrated where the Afeyan patent discloses a method for successful chromatography at a bed velocity of 5,000 cm/hr. Thus, even if a bed velocity of 5,000 is associated with a reduced velocity of 139,-000, as the Defendant claims, one cannot conclude as a result that the Afeyan patent anticipates the Quinn invention. Guiochon stated that he found nothing in the Afeyan patent that taught a method for chromatography at a reduced velocity of 5,000. The Defendant does not provide evidence to rebut this conclusion, and thus I do not find that Guiochon’s potentially erroneous estimation of the corresponding reduced velocity to be a material misrepresentation. Even assuming that Guiochon’s declaration contained a misrepresentation as to the reduced velocity associated with a bed velocity of 5,000 and that the miscalculation would have been material to the patent examiner, there is no evidence that the mistake constituted more than negligent error. Quinn testified that “it is common in the chromatography literature to select 10 “5 cm2/see as a typical value,” citing several scientific references. See Quinn Decl. of 3/6/02 at ¶ 16. Waters claims that Guiochon should have known this diffusion coefficient was an improper value based upon the Afeyan patent’s disclosure of the use of large biomolecules. It provides no evidence, however, that Guiochon intentionally chose the wrong coefficient. In the absence of evidence to the contrary, I find it reasonable that if Guiochon chose a “typical” but ultimately incorrect diffusion coefficient in his analysis, he did so by mistake. Thus, I do not find clear and convincing evidence that Cohesive engaged in inequitable conduct on this ground. 3. Conclusion Because Waters has failed to meet its burden of establishing inequitable conduct by clear and convincing evidence, judgment will enter for Cohesive on this issue. B. Willful Infringement Cohesive alleged that Waters’ infringement was willful and in knowing disregard of the '874 patent, and consequently it should be awarded enhanced damages on that ground. Under 35 U.S.C. § 284, “the court may increase damages up to three times the amount found or assessed.” While the statute is silent as to when enhanced damages are appropriate, the Federal Circuit en banc held earlier this month that “proof of willful infringement permitting enhanced damages requires at least a showing of objective recklessness.” See In re Seagate Technology, LLC, 497 F.3d 1360, 1371 (Fed.Cir.2007) (en banc). Recognizing that “the term [reckless] is not self-defining,” the Federal Circuit has determined that: to establish willful infringement, a pat-entee must show by clear and convincing evidence that the infringer acted despite an objectively high likelihood that its actions constituted infringement of a valid patent.... The state of mind of the accused infringer is not relevant to this objective inquiry. If this threshold objective standard is satisfied, the patentee must also demonstrate that this objectively-defined risk (determined by the record developed in the infringement proceeding) was either known or so obvious that it should have been known to the accused infringer. Id. (citation omitted). The patentee bears the burden of persuasion and must prove willful infringement by clear and convincing evidence. See Comark Comm. Inc. v. Harris Corp., 156 F.3d 1182, 1190 (Fed.Cir.1998). Willful infringement is not established by the simple fact of infringement, even where the accused has knowledge of the patents. Id. In determining whether infringement was willful, a court should consider: (1) whether there was a bona fide disagreement regarding patent invalidity or infringement, (2) whether the infringer solicited or followed the advice of counsel, (3) whether there was continued infringement after notice of probable infringement was received, (4) whether there was a degree of similarity between the patented and accused devices, (5) whether the infringer took efforts to avoid infringement, and (6) whether the infringer was indemnified against infringement costs. Because I find that Waters obtained an opinion of counsel in good faith, and because there was a bona fide dispute over whether the Oasis polymeric particles infringed Cohesive’s patents, Cohesive has not demonstrated that Waters’ infringement was willful. The following findings are based upon the affidavit testimony of Eduardo Bouvier, Alice Carroll, Michael Early, Anthony Janiuk, Ian King, Patrick McDonald, Uwe Neue, and Robert Plumb, the deposition testimony of Devette Russo, Ian King, and Michael Early, and the live testimony of Anthony Janiuk and Michael Early. 1. Facts Cohesive published its application for the '874 patent in May 1997. A little more than a week after the patent was filed, Dr. Patrick McDonald, a Water’s employee, forwarded the patent application to Waters’ in-house counsel, Anthony Janiuk. Waters then obtained a copy of the Cohesive prototype column that Cohesive had supplied to Glaxo-Wellcome for testing. Waters began manufacturing and producing its columns in February 1998, four months prior to the issuance date of the Cohesive patent. Dr. Bouvier, a scientist employed by Waters, compared the Cohesive prototype to the Oasis column in order to determine whether Cohesive’s patents would preclude Waters from selling an HPLC column filled with 30 micron Oasis particles. Bou-vier completed experiments and presented his results in May 1997 focusing on the particle size, interstitial fraction, and reduced velocity of Oasis columns. He used size-exclusion chromatography to determine the interstitial volume and pore-size distribution of the Oasis 30-micron particles. Bouvier, Janiuk, scientist Uwe Neue, and Ian King, Waters’ marketing manager for the Chromatography Division, held a meeting with scientists and in-house counsel on August 19, 1998 to review the Cohesive patent application and discuss whether Waters’ columns infringed. Bouvier presented the results of his experiments, which demonstrated, among other things, that the Oasis particles collapsed when subjected to a pressure of up to 5,000 psi. Bouvier’s experiments also showed the particles resisted post-compression deformation at pressures of 7,000 psi and above. After the August 25 meeting, Janiuk drafted an opinion that concluded that the Waters columns were not infringing. In his opinion, Janiuk stated that “Oasis particles are unmistakably in the nature of a polystyrene-type composition” and that therefore the Oasis HLB columns “do not have rigid particles as such term ‘rigid’ is defined in the patent.” Janiuk did not confer with outside counsel. He wrote an opinion memorandum dated September 1, 1998, summarizing the scientific evidence communicated at the August meeting, con-eluding non-infringement, either literally or under the doctrine of equivalents. Cohesive argues that Waters opinion ignored evidence that articles resist plastic deformation at pressures up to 7,000 psi. See Tr. Transcript Day 7 at 33-36. 2. Discussion I do not find clear and convincing evidence that Waters’ infringement by manufacture and sale of the Oasis HPLC columns was undertaken willfully. First, Waters did not copy the column it obtained from Cohesive through Glaxo-Well-come. Cohesive’s particles are silica particles 50 micron in diameter, while Waters’ particles are polymeric particles of 30-micron size. Second, Waters engaged in sufficient due diligence in determining whether its product would infringe to convince me that its effort to avoid infringement was in good faith. I find that it manufactured the accused column only after satisfying its obligation to ensure there was not a high likelihood, considered both objectively and subjectively, that its actions would constitute infringement. Waters obtained a copy of Cohesive’s September 1997 patent application, conducted experiments, and held a meeting among its in-house counsel and scientists on August 25, 1998, prior to the sale and manufacture of the infringing columns. The scientists presented their results and unanimously concluded that the Oasis polymeric particles were not rigid. In particular, Dr. Bouvier conducted tests, the evidence of which was admitted at trial in this case, showing that the Oasis particles were compressible at pressures well below 5,000 psi, that the interstitial volume of the column was 40%, and that the flow in the Waters columns was not “turbulent.” . Cohesive contends that Waters’ production of the infringing columns after learning of Cohesive’s patented technology in August 1998 was willful and knowing because Dr. Bouvier ignored evidence that the Oasis particles “resist plastic deformation,” as defined in my claim construction relating to the term “rigid.” Waters does not dispute that Oasis particles resist plastic deformation. Waters claims, however, that Bouvier reasonably believed the particles not to be rigid because they demonstrated substantial collapse during compression at pressures 5,000 psi and greater. I adopted a functional definition of rigid from the descr