Applera Corp. v. Micromass UK Ltd. (204 F. Supp. 2d 724, 2002)

OPINION

McKELVIE, District Judge.

This is a patent infringement case. Plaintiff Applera Corporation, formerly known as PE Corporation, is a Delaware corporation with its principal place of business in Norwalk, Connecticut. Plaintiff MDS Inc. is a Canadian corporation with its principal place of business in Toronto, Canada. Plaintiff Applied Biosys-tems/MDS Sciex, formerly known as Per-Hn-Elmer Sciex Instruments, is a Canadian partnership formed under the laws of Ontario and having a place of business there. Applera and MDS are general partners of Applied Biosystems/MDS Sciex. MDS is the owner of U.S. Patent No. 4,963,736 (the ’736 patent), entitled “Mass Spectrometer and Method and Improved Ion Transmission.” Applied Bios-ystems/MDS Sciex is the exclusive licensee of the ’736 patent. The plaintiffs will be collectively referred to as AB/Sciex.

Defendant Micromass UK Ltd. is a British corporation with its principal place of business in Manchester, United Kingdom. Micromass UK manufactures mass spectrometers, including the Quattro Ultima. Defendant Micromass, Inc. is a Massachusetts corporation with its principal place of business in Beverly, Massachusetts. Mi-cromass, Inc. distributes and sells mass spectrometers in the United States. The defendants will be collectively referred to as Micromass.

On February 18, 2000, AB/Sciex filed its complaint in this action alleging that Mi-cromass’s Quattro Ultima infringes one or more claims of the ’736 patent. Micro-mass denied infringement, and raised affirmative defenses and counterclaims. Mi-cromass’s counterclaims seek a declaratory judgment that the ’736 patent is invalid and unenforceable, and alleges that AB/ Sciex has filed this suit in an improper effort to maintain monopoly power in violation of section 2 of the Sherman Act, 15 U.S.C. § 2, or attempted or conspired to do so.

In July 2001, Micromass introduced a redesigned Quattro Ultima, in which part of its allegedly infringing structure, a hex-apole ro.d set, was replaced with a series of rings. AB/Sciex contends that the new Quattro Ultima, known as the “Ion Tunnel Quattro Ultima,” infringes the ’736 patent under the doctrine of equivalents. The original Quattro Ultima will be referred to as the “Hexapole Quattro Ultima.”

On October 22 and 23, 2001, the parties filed dispositive motions. AB/Sciex sought summary judgment on Micromass’s inequitable conduct defense and antitrust counterclaims. Micromass filed nine summary judgment motions. In five of those motions, Micromass contended that the Hexapole Quattro Ultima did not infringe the claims of the ’736 patent. Micromass also sought summary judgment that: (i) the asserted claims of the ’736 patent are invalid due to anticipation; (ii) the ’736 patent is invalid for indefiniteness; (iii) the ’736 patent is unenforceable due to inequitable conduct; and (iv) the Ion Tunnel Quattro Ultima does not infringe the claims of the ’736 patent under the doctrine of equivalents.

On December 13, 2001, the court held a hearing- in accordance with Markman v. Westview Instruments, Inc., 517 U.S. 370, 116 S.Ct. 1384, 134 L.Ed.2d 577 (1996), to construe the disputed claim terms. On February 6, 2002, the court issued a memorandum opinion construing the claim terms. Applera Corp. v. Micromass UK Ltd., 186 F.Supp.2d 487 (D.Del.2002). On February 7, 2002, the court held a hearing to address the summary judgment motions. At that hearing, Micromass moved for reconsideration of the court’s claim construction. It also presented its summary judgment motions. The court took the motions under advisement and permitted the case to go to trial. The court will present its decision on Micromass’s motion for reconsideration in this opinion.

The court bifurcated the trial into two components; a jury trial on the infringement and invalidity issues, and a bench trial on inequitable conduct and equitable estoppel. The. ten-day jury trial began March 4, 2002. At the conclusion of that trial on March 15, 2002, Micromass moved for judgment as a matter of law under Federal Rule of Civil Procedure 50(b). The court reserved judgment. The jury unanimously found that the Hexapole Quattro Ultima literally infringed the two independent claims of the ’736 patent, claims 1 and 14. The jury also found that: (i) the Ion Tunnel Quattro Ultima infringed claims 1 and 14 under the doctrine of equivalents; (ii) Micromass’s infringement was not willful; (iii) the claims of the ’736 patent were not invalid; and (iv) AB/ Sciex’s mass spectrometers were marked with notice' of the ’736 patent since February 1999. The jury awarded $47.5'million in damages, of which $41.3 million was attributable to the Hexapole Quattro Ulti-ma.

On April 3, 2002, the court conducted a one-day bench trial on -Micromass’s inequitable conduct and equitable estoppel defenses. On inequitable conduct, Micro-mass contends that the ’736 patent is unenforceable because AB/Sciex failed to present the PTO with material prior art, including an earlier European Patent application of one of the inventors of the ’736 patent. Micromass also contends that AB/Sciex did not present certain experimental evidence to the PTO during the prosecution of the patent, and made false and misleading statements describing the prior art to the PTO during reexamination. With respect to equitable estoppel, Micromass contends that before it developed the Quattro Ultima, AB/ Sciex led it to believe that the ’736 patent would not be asserted against it because it was invalid. The parties have submitted proposed findings of fact and conclusions of law on those topics.

On April 1, 2002, Micromass filed its post-trial motion for judgment as a matter of law or for a new trial. That motion addresses the jury’s verdicts on the validity of the asserted claims of ’736 patent, Micromass’s infringement, and damages. On validity, Micromass contends that it established at trial that the claims of the ’736 patent are anticipated by the European Patent application discussed earlier, and that its claims would have been obvious to one of skill in the art based on certain combinations of prior art references. With respect to infringement, Mi-cromass alleges that AB/Sciex failed to prove that its devices met two of the claim limitations of the asserted claims of the ’736 patent and that the Ion Tunnel Quattro Ultima was infringed those claims under the doctrine of equivalents. On damages, Micromass argues that AB/Sciex is not entitled to lost profits damages as a matter of law, and that the jury could not reasonably accept AB/Sciex’s reasonable royalty estimates.

AB/Sciex has also submitted two post-trial motions. It moved for summary judgment on Micromass’s antitrust counterclaims as inconsistent with the jury’s verdicts on invalidity and infringement. It also moved to alter or amend the judgment to add pre-judgment interest and enjoin further infringement.

This is the court’s decision on all of motions described above.

I. FACTUAL AND PROCEDURAL BACKGROUND

The following facts are taken from the court’s earlier opinion, the ’736 patent and its prosecution history, and the evidence presented at both the jury and bench trials. For purposes of Micromass’s assertions of inequitable conduct and equitable estoppel, the following recitation comprises the court’s findings of fact.

A. Mass Spectrometry and the ’786 Patent

1. Background of the Technology

The ’736 patent discloses a technology used in mass spectrometers. Mass spectrometers are used to analyze the chemical composition of trace substances in a sample gas or liquid. They are useful in a number of endeavors, including pharmaceutical experimentation and testing food and drink for minimum quality standards. Mass spectrometers operate by applying an electrical charge to the molecules of the substance being analyzed, resulting in charged molecules known as ions. By applying an electrical field to the ions, the substance being analyzed can be separated into its constituent parts using the ratio of their molecular weight to the charge.

The ’736 patent-is directed to a typical type of mass spectrometer, referred to as a quadrupole mass spectrometer. Figure 1 of the' ’736 patent is an example of a quadrupole mass, spectrometer or two-stage mass spectrometer. It is comprised of four basic types of structures — an ionization chamber, an ion guide, a mass analyzer, and a detector. Beginning at the right of Figure 1, the ionization chamber (16) contains a duct (14) for inputting the trace substance. The substance is ionized by applying an electric charge with an electric discharge needle (18). The ionization chamber is typically at atmospheric pressure. The ions proceed next through a curtain gas chamber (24) and into the ion guide (30). The ion guide is a set of electrode rods (32) in a vacuum chamber created by pumping out the ambient gas (31). The quadrupole mass spectrometer has four electrode rods, spaced apart in three dimensions to create an interior space through which the ions are directed, although more than four rods can be used (six in a hexapole, eight in a octopole, etc.). The ion guide uses an alternating current (AC) to channel the ions through the central space. The altering positive and negative charges in adjacent rods forces the ions to oscillate between the rods while traveling down their length. This is known as “strong focusing.”

The focused ions, separated from the ambient gas, are directed through a small orifice (34) and into the mass filter chamber (38). The mass filter chamber is kept at an even lower pressure than the ion guide with a second vacuum pump (39). A second set of rods (40), known as the mass filter, applies both an AC voltage and a direct current (DC) voltage to select ions of a particular mass-charge ratio. Using a particular voltage in the mass filter separates the desired ions from the undesired, and permits the desired ions to continue on their path to a detector (48) that records their presence.

The basic structure of a quadrupole ion guide was well-known at the time of the application of the ’736 patent. It was disclosed in several papers, including the articles cited as prior art in the ’736 patent.

As plaintiffs expert witness, Dr. Christie Enke, explained, one of the challenges in constructing quadrupole ion guides was that improving ion transmission through the device required a very low pressure in the vacuum chambers, also called cells. Because low pressures were only achieved by using powerful vacuum pumps, the quadrupole mass spectrometers on the market at the time the ’736 patent issued were large and expensive.

The inventors of the ’736 patent, Donald J. Douglas and John B. French, were aware of this problem. They described a “classical equation” for ion transmission in which “ion signal intensity (ion current) transmitted through the cell decreases with increasing gas pressure in the cell.” ’736 Patent, Col. 1, In. 33-37. Douglas and French set out to solve this problem and it was their discovery that led to the ’736 patent.

2. The ’736 Patent

On November 15, 1989, Douglas and French filed United States Application Serial Number 07/437,047 with the United States Patent and Trademark Office (PTO). The application named Douglas and French as its inventors and MDS Health Group Limited, a predecessor entity of MDS, Inc., as assignee. The priority date was claimed from a Canadian patent application filed on December 12, 1988.

Douglas and French, and then- representatives, cited three prior art references to the PTO Examiner in their application. Those references were: (i) U.S. Patent Number 4,328,420 (the ’420 patent), for which French was the inventor; (ii) Dr. R. Smith, et al., “On-line Mass Spectrometric Detection for Capillary Zone Electrophoresis,” Anal. Chem., Vol. 59, p. 1230 (Apr. 15, 1987) (the “1987 Smith Paper”); and (iii) Dr. R. Smith, et al. “Capillary Zone Electrophoresis — Mass Spectrometry Using an Electrospray Ionization Interface,” Anal. Cham., Yol. 60, p. 436 (Mar. 1, 1988) (the “1988 Smith Paper”).

The Douglas and French application describes a typical quadrupole mass spectrometer, using both an ion guide and a mass filter. They claimed that the novelty of the application, however, was the description of particular parameters of pressure, electrode rod length, and voltage designed to improve the transmission of ions through the ion guide at a pressure higher than previously used and with “a large enhancement in ion signal.” '736 Patent, Abstract. According to the ’736 patent’s specification,

The inventors have now discovered that the classical equation describing ion signal intensity does not in fact describe the situation accurately when dynamic focusing is used in the interstage region and that when the gas pressure in the region of the ion optic elements is increased within certain limits and when the other operating conditions are appropriately established, ion transmission is markedly increased.

’736 Patent, Col. 1, In. 42-49. The gas pressure and other operating conditions of the ion guide were explained in the final two clauses of the two independént claims, claim 1 (a product claim) and 14 (a method claim). The penultimate clause of each claim, l(j) and 14(f), explains those conditions as the product of the pressure in the chamber and the length of the electrode rods (P x L), which must be “equal or greater than 2.25 x 10 "2 torr cm” and less than “that pressure at which an electrical breakdown will occur between the rod means of said first rod set.” The final clauses of claims l(k) and 14(h) describe a “means for maintaining” (claim 1) or “controlling” (claim 14) the kinetic energy of ions moving from the inlet to the first rod set “at a relatively low” level. The specification explains that this means for maintaining is a combination of the pressure in the ion guide vacuum chamber and a DC potential voltage applied to the inlet orifice that attracts ions into the ion guide. Together, these two claim limitations define the phenomenon of the invention, referred to as “dynamic focusing” or “collisional focusing.”

While the inventors admitted that the reason for “collisional focusing” was “not entirely understood,” they reasoned that higher pressures caused the ions to collide with more gas molecules as they approached and traversed the ion guide. These collisions reduce the energy of the ions. This loss of energy is called “cooling” or “thermalization.” The result of the effect is that “ions are being forced toward the center line of the system and that the mechanism which is causing the [ion transmission] enhancement is a kind of collisional focusing or damping effect that concentrates the ion flux closer to the central axis.” ’736 Patent, Col. 6, In. 64-68.

On May 8, 1990, the PTO Examiner issued a final Office Action allowing all 24 claims of the Douglas and French application. The PTO granted the application on October 16,1990 and the application issued as the ’736 patent.

B. Accusation of Infringement and the Reexamination History

In late 1996, Micromass scientist Dr. Patrick Turner and others published an article entitled “Interface Studies in the ICP-Mass Spectrometer” in Plasma Source Mass Spectrometry Developments and Applications (the “ICP Article”). The ICP Article described a device using a hexapole rod set and operating as both “a collision cell and for focusing ion beams in electrospray sources.” A collision cell is a mass spectrometer in which high pressure gas is used to collide ions with gas molecules and cause them to fragment. The article also discussed the “thermalising properties” of the device, which resulted in “improved resolution in an analyser.” The article did not give a detailed explanation of the structure of the ICP device, but simply suggested that it “thermalised[d] the ions produced in the ICP torch” and thus had a “low energy spread.”

Dr. William Davidson, AB/Sciex’s Vice President of Science and Technology, read the ICP Article and attended a trade conference at which the ICP device was discussed. He understood from the article and conference that the ICP device could be used both as a collision cell and to guide ions with increased focus. He also took note of the “thermalising” properties of the invention, which he understood to be another description of the phenomenon of collisional focusing. Concerned that the ICP device might infringe the ’736 patent, Dr. Davidson instructed AB/Sciex’s attorney to send a letter to Micromass, warning them of his concern. On July 10,1997, one of AB/Sciex’s outside counsel, Richard J. Parr, sent a letter to Micromass. The letter referenced the ICP article and stated that “[i]t appears to us that this device, if sold, will infringe the claims of the above-identified Canadian and U.S. patents

On April 16, 1997, Dr. David Yorke, Micromass’s Intellectual Property Manager, responded by letter to AB/Sciex’s correspondence. Dr. Yorke stated the ICP device “does not infringe any valid claim” of the ’736 patent and cited four references of prior art. Those references were:

(1) French, European Patent Application, Publication No. 0 023 826, February 11, 1981 (the “French application”);

(2) Boitnott et al., Optimization of Instrument Parameters for Collision Activated Decomposition (CAD) Experiments for a Finnigan Triple Stage Quadrupole GC/MS/MS/DS, 1981 Pittsburgh Conference On Analytical Chemistry and Applied Spectroscopy, Abstract No. 782 (the “Finnigan abstract”);

(3) Boitnott et al., Optimization of Instrument Parameters for Collision Activated Decomposition (CAD) Experiments for a Triple Stage Quadrupole (TSQ™ GC/MS/MS/DS), Finnigan Topic 8160 (the “Finnigan paper”); and

(4) Caldecourt et al., An Atmospheric-Pressure Ionization Mass Spectrometer/Mass Spectometer, International Journal of Mass Spectrometery and Ion Physics, Vol. 49, p. 233-251 (1983) (the “Caldecourt article”).

Dr. Yorke also stated that “Micromass does not propose to seek a license under [the ’736 patent] or its equivalents, at least in the foreseeable future.”

On April 24, 1997, Parr wrote back to Dr. Yorke and requested copies of the references cited in his letter. There was no further communication between the parties until this suit was filed.

On September 30, 1997, AB/Sciex filed a Reexamination Request for the ’736 patent with the PTO. AB/Sciex identified eight references of prior art as possibly raising substantial new questions of patentability, including the four references cited by Mi-cromass. AB/Sciex identified those four references as collision cell references that were “discussed during license negotiations with Micromass UK Limited.” The other four references AB/Sciex cited were:

(1)Schaaf et al., Trapped Ion Density Distribution in the Presence of He-Buffer Gas, Applied, Physics, Vol. 25, pp. 249-251 (1981) (the “Schaaf article”);

(2) Vedel et al., “Influence of space charge on the computed statistical properties of stored ions cooled by a buffer gas in a quadrupole rf trap,” 29 Physical Review, No. 4, pp. 2098-2101 (1984) (the “Vedel article”);

(3) Stafford et al., “Recent Improvements in and Analytical Applications of Advanced Ion Trap Technology,” International Journal of Mass Spectrometry and Ion Processes, Vol. 60, pp. 85-98 (1984) (the “Stafford Article”); and

(4) Stafford et al., European Patent Application, Publication No. 0 013 207, July 11,1984 (the “Stafford application”).

AB/Sciex described these four references as “ion trap” references.

1. Collision Cell References

With respect to the four references provided by Micromass, AB/Sciex described them as “tandem mass- spectrometers.” Tandem mass spectrometers, also' called “triple stage mass spectrometers” are comprised of two AC-DC mass filters separated by a collision cell containing an AC-only rod set. In the figure below, taken from the French application and French’s ’420 patent, the ions move from the first mass filter (4), through the collision cell (14), and then into the second mass filter (8).

The collision cell accepts the ions not filtered out by the first mass filter, then collides those ions into a gas at a high energy, causing them to fragment. These fragments are called “daughter ions,” and proceed into the second mass filter for further filtering.

AB/Sciex distinguished the tandem mass spectrometer references before the PTO Examiner by explaining that a collision cell operates differently from a two-stage quadruple mass spectrometer, such as the one described in the ’736 patent. It stated that

[a] collision cell, such as the one described in the French application, dissociates a parent ion into fragment ions by creating conditions whereby a high energy parent ion collides with a high pressure gas.... A mass spectrometer according to the invention, on the other hand, is intended to improve the transmission of ions through a cell. The mass spectrometer according to the invention uses an increased pressure to improve ion transmission and maintains “the kinetic energies of ions moving from said inlet orifice to said first rod set at a relatively low level” (claim 1). The French application would therefore teach away from the invention since it collides ions at high kinetic energies into a high pressure region to dissociate the ions into daughter ions, which is in contrast to the invention which uses low kinetic energy ions and an increased pressure to produce an improved transmission of ions entering the device.

Request for Reexamination at 18. AB/ Sciex made a similar distinction for each of the tandem references. Id. at 15-16 (Finnegan abstract), 18 (Finnegan paper), 21-22 (Caldecourt article).

2. Ion Trap References

The ion trap references, including the Schaff article, disclose a mass spectrometer in which AC-only electrodes generate electric fields that trap ions within them for some period of time. The electric field can be varied to eject ions of different mass to charge ratios from the trap for study. The Schaff article discusses using a light buffer gas such as helium to cool the ions in an ion trap. “Another way to reduce ion temperature is the addition of a small amount of a light buffer gas. Collision[s] between the ions and the gas tend to cool the ions.”

During reexamination, AB/Sciex contended that the use of collisional cooling in ion traps would not have taught one of skill in the art to use the same damping effects to focus ions in an ion guide. AB/ Sciex also distinguished the ion trap references from the ’736 patent on the basis of the differing structures of the two types of mass spectrometers, including the absence of rod sets and an AC-DC mass filter.

3. The Reexamination

The PTO Examiner granted AB/Seiex’s request for reexamination on November 20, 1997. The Examiner cited three of the four collision cell references, including the French application, the Finnigan abstract and the Finnigan paper, and concluded that those references raised “a substantial likelihood that a reasonable examiner would consider these teachings important in deciding whether or not the claims are patentable.”

On February 3,1998, the PTO Examiner rejected the claims of the ’736 patent as obvious in light of the structure disclosed in the French application and the voltage parameters disclosed in the Finnigan abstract and paper. He stated, “it would have been obvious to a person having ordinary skill in the art to control the energies of the ions entering the French apparatus in accordance with the teaching of [the Finnigan abstract and paper] by providing DC voltage between the rods of the first quadrupole and the inlet wall.” Because it is relevant to Micromass’s anticipation and inequitable conduct contentions, the court will discuss the French application in some detail.

4. The French application and the %20 patent

The French European Patent Office application discloses a method for improving ion transmission in a tandem mass spectrometer. It is related to French’s United States ’420 patent and both depict Figure 1, shown above, as an embodiment of the invention. Both depict three rod sets through which ions travel. The first and third rod sets (10, 12) are mass filters utilizing AC-DC voltages and are located in sections (4 and 8) of a vacuum chamber (2). The second rod set (14) is a collision cell containing rods applying only AC voltages. A cross-section of the collision cell is shown in Figure 2. The collision cell has a target gas (29) input into the space between the rods (14) to dissociate the ions. That target gas is removed by a- refrigerating mechanism (38), ■ which acts as a pump to withdraw the target gas and lower the pressure. The AC-only rod set has two kinds of rods. Rods 14-1 are typical electrodes with solid centers. Rods 14-2 are called “open structure end extensions” and are placed on either side of the 14-1 rods. Open structure rod extensions are formed of thin stiff rods or wires “arranged in a curved configuration to simulate the shape of the outer portion of a normal quadrupole rod.” Figure 5 depicts the typical solid rods with the open structure rod extension.

The open structure rod extensions accomplish the purpose of the French application. As stated in the specification, in collision cells using solid AC-only rods, “relatively little gas can escape [between the rods], and therefore a substantial gap must be left between the ends of adjacent quadrupole sections,” to permit the collision gas to be evacuated from the chamber without increasing gas pressure in the mass filters. These gaps decrease ion signal intensity. By using open structure rod extensions through which gas can flow, “only a small proportion of the target gas entering the centre quadrupole section 6 will travel into the end sections 4, 8.” Because the gas can proceed through the open structure rod extensions (14-2), the mass filters (10, 12) can be located more closely to those AC-only rods (14). This “close coupling” of elements increases the ion transmission of the tandem mass spectrometer.

The specification of the French application explains that in a “typical” tandem mass spectrometer, the gas density in the “target region, i.e. in the space between rods 14-1,” is between 10 ~2 torr and 10 torr. It also describes the length of the rod extensions (14-2) as equal to the length of the solid rods (14-1), which is explained to be “(e.g. 4 inches).” Because 4 inches is approximately 10.16 centimeters, the P x L product of the French application is as high as 10.16 x 10 ~2 torr cm.

Though otherwise directed to a tandem mass spectrometer, the final paragraph of the specification of French application states that “it may also be used with only two [quadrupole] sections in a series, namely an AC-only section and an AC-DC section.” That paragraph goes on to state:

Such an arrangement is shown and described in the co-pending application of Sciex Inc., the description and drawings of which are hereby incorporated by reference into this application. In such system ions entering a vacuum chamber are guided into a conventional AC-DC quadrupole mass spectrometer by an AC-only section arranged in series with the conventional section, the rods of the AC-only section being of open construction to permit gas entering with the ions to flow through the rods and escape. The same phase and spacing relationships as described previously apply.

Claim 1 of the French application then goes on to claim a mass spectrometer in which an AC-DC rod set precedes the AC-only rod set, but the two rod sets being arranged “closely longitudinally” to one another. Further dependant claims add the collision cell structure to the AC-only rod set.

The “eo-pending application of Sciex Inc.” referred to in the French application is a similar application that resulted in French’s ’420 patent, which was cited by the inventors as prior art in their application for the ’736 patent. The ’420 patent also discloses a tandem mass spectrometer in which part of the AC-only rod set in the collision cell has an open structure and the rods are closely coupled to one another to improve ion transmission. However, claim 1 of the. ’420 patent reverses the two claimed rod sets, first an AC-only rod set with an open structure, and then an AC-DC mass filter. The tandem mass spectrometer embodiment, discussed throughout the specification, is only added by dependent claim 9, which adds a third rod set functioning as a mass filter before, in the path of ion travel, the collision cell. The specification of the ’420 patent is similar to that of the French application. However, the pressure in the target region of the collision cell is described as 10 torr to 10 “5 torr. Because the rods are the same length (4 inches), the highest P x L product disclosed is 10.16 x 10 3 torr cm, which is beneath the 2.25 x 10 ~2 torr cm limitation of the ’736 patent.

Dr. Enke, AB/Sciex’s technical expert, opined that the French application recites a gradient of pressures between the rods of the collision cell, such that the average pressure along the rods-could be calculated using either an exponential decay expression or a cosine squared distribution. Dr. Raymond March, Micromass’s expert, performed those calculations and concluded that the average pressure along the rods was 2.46 x 10 ~3 torr cm using exponential decay, resulting in a P x L ■ product of approximately 2.46 x 10 ~2 torr over the 10 cm rods. Based on this calculation, Dr. Enke admitted at trial that the French application’s collision cell, even assuming that it discloses a gradient of operating pressures, had a maximum which exceeded the 2.25 x 10 ~z torr cm claimed in the ’736 patent. ■

5. The Reexamination Concludes

On March 11, 1998, the PTO Examiner met with Geoffrey Sutcliffe, AB/Sciex’s attorney, and Dr. Douglas to discuss the patentability of the claims in the ’736 patent. In his interview summary, the PTO Examiner indicated that he reached agreement with AB/Sciex on all claims. He stated:

Applicant could remove French et al. as a reference by' establishing that the product of the pressure and length of the AC only quadrupole described in the reference when that quadrupole was used as a collision cell between two mass analyzing quadrupoles to fragment ions was not intended to be used when that quadrupole was used only as an ion guide and not to fragment the ions.

DX 108 (Interview Summary Mar. 11, 1998).

Following the PTO Examiner meeting, on April 3, 1998, AB/Sciex submitted an Amendment to Non-Final Office Action, adding new dependent claims 25-30 to the patent. The Amendment also addressed the prior rejection of claims 1-24 discussed at the March 11 interview and enclosed a Declaration from Dr. French supporting the patentability of those claims. The Amendment stated' that the Declaration “establishes] that the French application does not suggest that the claimed first chamber have [sic] the specified product of pressure and length and furthermore provide[s] additional reasons that the claims are in condition for allowance.”

The French Declaration further distinguished AC-only rod sets in a collision cell from the same rod sets used in an • ion guide. Paragraph 9 of the Declaration made the same point that Dr. Douglas had made in his meeting with the PTO Examiner. “The French application does not suggest that the recited pressure range and rod length may be used in a quadrupole section which acts as an ion guide and which is not intended to fragment the ions.”

■ In Paragraph 6 of his Declaration, Dr. French also recited the pressure gradient theory that would later be introduced at trial by Dr. Enke. In describing the French application, he stated:

The gas pressure in the second quadrupole 6 is higher than pressures in quadrupole sections 4 and 8 .... The density distribution of the target gas varies across a length of the second quadrupole 6 with the pressure peaking near a center of the quadrupole section 6 and falling off at either end of the quadrupole section 6. Page 7 of the French application recites a pressure range of 10 ~2 torr and the ends of the quadrupole section 6 would be at or below about 10 “4 torr.

French Declaration ¶ 6.

On June 15, 1998, the PTO Examiner issued a final Office Action confirming that claims 1-24 were patentable, but rejecting claims 25-30 as indefinite. With respect to claims 1-24, the Examiner stated that the “declaration of Dr. French ... establishes that the apparatus disclosed in the French application does not operate with a product of pressure and rod length greater than or equal to 2.25 x 10 ~2 torr cm in a chamber containing a rod set operated with only AC voltages applied.”

On August 12, 1998, AB/Sciex filed a Response After Final Office Action, in which it argued that claims 25-20 are patentable. AB/Sciex also continued to contend that claims 1-24 were not suggested by the French application and recited four reasons: “(1) the French application teaches away from the invention by suggesting that pressure be reduced in the first chamber, (2) the French application does not suggest the product of pressure and rod length in the first chamber, (3) the French application does not suggest collisional focusing, (4) the French application does not suggest improving the transmission of ions entering the first chamber.”

Following further correspondence relating to the new claims, on May 25,1999, the PTO issued a Reexamination Certificate confirming the patentability of claims 1-24 and allowing claims 25-30.

C. The Jury Trial

1. AB/Sciex’s Infringement Presentation

AB/Sciex’s first witness at trial was its expert, Dr. Enke. Dr. Enke explained what a mass spectrometer was and the purposes for which it was used. He explained the basic mechanics of ion motion, including how ions are guided by electric fields and the loss of energy caused by collisions with gas. Dr. Enke explained the traditional problems with low pressure ion guides. He also described the invention in the ’736 patent and that it was novel because it taught operating ion guides at a relatively high pressure in a manner that achieved increased ion transmission. Dr. Enke also explained the “collisional focusing” employed by the invention. He illustrated his testimony using a diagram of the API 3000, an AB/Sciex product that utilizes the ’736 patent.

Dr. Enke also testified about Micro-mass’s Quattro Ultima mass spectrometer. The Hexapole Quattro Ultima is a tandem mass spectrometer comprised of a quadrupole mass filter, a hexapole collision cell, and another quadrupole mass filter. These are all shown in the figure below within chamber 4. Preceding this structure are two hexapole ion bridges, contained in chambers 2 and 3, that function as ion guides. Ions are created with a Z spray in an atmospheric pressure chamber, are then directed into an empty chamber (chamber 1), and then enter the two hexapole ion bridges (chambers 2 and 3). The ions then proceed into chamber 4, where they encounter the tandem (or three-stage) mass spectrometer structure — a mass filter, collision cell, and another mass filter. Finally, the remaining ions are detected at the end of the process.

Hexapole Hexapole Ion Bridge Pre-Filter Post-Filter Collision Cell Photomultiplier Phosphor Focus Ring Conversion Dynode Z Spray H H H Quadrupole M Quadrupole Ion Source M M -BB-- -BB-

Dr. Enke testified that the two hexapole ion bridges were ion guides within the claim limitations of the ’736 patent, and that they directed ions into the mass filter that followed. He explained that by practicing the ’736 patent, the Hexapole Quat-tro Ultima achieved improved transmission of ions. In support of that conclusion, he discussed computer simulations that he ran using parameters taken from the Quattro Ultima. He also presented Micro-mass documents showing that even though the size of the orifice before the mass filter was reduced in size, a large percentage of ions continued into the mass filter. Based on this testimony, he opined that the Hex-apole Quattro Ultima practices collisional focusing and contains all the limitations of claims 1 and 14 of the ’736 patent.

In July 2001, Micromass replaced the hexapole ion bridges with “ion tunnels.” The ion tunnels are comprised of 84 separate electrode rings, shaped like washers, with a hollow center. The rings are arranged sequentially so that ions proceed in a straight line down their center. The voltages in adjacent rings is alternated and the ions oscillate through their middle. Dr. Enke testified that the ion tunnel was the equivalent of the “first rod set” of the ’736 patent because in accomplishes the same function, in the same way, with the same result. That is, it guides ions through to the mass filter and accomplishes collisional focusing. He noted one distinction, however. While the ion rods only produce electric fields in two dimensions as the ions traverse their length, each ring creates an electric field in a third dimension, called the “axial” direction. The axial direction is the direction of ion travel, both towards and away from the inlet and interchamber orifices. Dr. Enke testified that this axial electric field has a negligible effect because the number of rings cancels out any effect on ion travel.

On cross-examination, Dr. Enke was questioned on his understanding of several of the parameters of Quattro Ultima. He also explained differences between the ’736 patent and several articles of prior art, including collision cell references such as the Caldecourt article and a tandem mass spectrometer named the TAGA 6000 that was manufactured by AB/Sciex. He also distinguished ion traps from the ’736 patent.

AB/Sciex also presented the testimony of Dr. Donald Douglas, an AB/Sciex scientist and one of the • inventors of the ’736 patent. Dr. Douglas explained that his invention was achieved while trying to reduce the size of the vacuum pumps used in AB/Sciex’s earlier TAGA 6000. He also testified concerning why his invention was novel compared to the prior art. AB/Sciex then presented ■ the testimony of Gerald Mossinghoff, a former Commissioner of Patents, to explain the procedures for prosecuting patents and seeking reexamination before the PTO.

AB/Sciex also called Dr. William Davidson, a Vice President of Science and Technology at AB/Sciex. Dr. Davidson described AB/Sciex’s TAGA 6000 product and explained that it had many shortcomings, including a large, refrigerated eryo-pump that was expensive and unreliable. The search for a smaller device without a eryopump resulted in the discovery of collisional focusing. Collisional focusing was incorporated in several new mass spectrometers that lacked a eryopump and were thus were considerably smaller than the TAGA 6000, but had the same sensitivity. AB/Sciex also developed the API 3000, which it introduced in 1998. The API 3000 contains the technology of the ’736 patent and has a high sensitivity. Dr. Davidson explained that there is a high demand for the API 3000 and that AB/ Sciex has sold over one thousand of them. In the market for high sensitivity mass spectrometers, AB/Sciex’s only real competitor is Micromass, although Dr. Davidson testified that another company named Finnigan had recently produced a high sensitivity mass spectrometer.

Laura Lauman, an Executive Vice President of AB/Sciex, testified that she supervised the marketing efforts for the API 3000. She explained that there was strong demand for the product and that it was more competitive than the Micromass product at the time, the Quattro LC. However, after the Quattro Ultima was introduced, AB/Sciex began offering larger discounts on the API 3000 to remain competitive. Throughout the manufacture and marketing of the API 3000, Lauman explained, there was the capability of selling more API 3000 devices than AB/Sciex sold. AB/Sciex also called Joseph Anacle-to, Director of the LCMS product line of mass spectrometers for AB/Sciex. Ana-cleto testified that AB/Sciex had the capability to manufacture more ABI 3000 devices from 1998 through the present.

Finally, AB/Sciex’s damages expert, Dr. Marion Stewart, testified to both the profits allegedly lost by AB/Sciex from sales of the Quattro Ultima and a reasonable royalty for the ’736 patent. Dr. Stewart discussed the demand for the API 3000’s sensitivity, its sales history, and the price erosion caused by the Quattro Ultima. He also concluded that the Quattro Ultima was the only competitive alternative to the API 3000 in terms of sensitivity for most of the period from 1998 until the trial. He took into account the recently introduced Finnigan product and attributed to it some of the sales that might have been made by AB/Sciex if the Quattro Ultima did not exist. He opined that AB/Sciex’s total lost profits were $52.3 million. If only the Hexapole Quattro Ultima infringed, lost profits were $45.6 million.

Dr. Stewart also testified that a reasonable royalty for the ’736 patent would be $225,000 per machine. He acknowledged that this was a high number, but explained that it was justified because AB/Sciex was unlikely to license the ’736 patent to its competitors. The total reasonable royalties calculated by Dr. Stewart amounted to $45.5 million. If the Ion Tunnel Quattro Ultima did not infringe, the total reasonable royalty for just the Hexapole Quattro Ultima was calculated to be $38.9 million. If Dr. Stewart assumed a lower measure of Micromass’s incremental profits, then he calculated the total reasonable royalties to be $34.1 million.

On cross-examination, Dr. Stewart was examined on the assumptions underlying his lost profits analysis, including whether there was demand for the sensitivity of the API 3000.

In conclusion, AB/Sciex presented deposition testimony of several Micromass witnesses, including Micromass’s technical director, Dr. Robert Bateman, another technical director, Dr. Kevin Giles, the Quattro Ultima Project Manager, Dr. Stuart Jarvis, the Managing Director of Micromass, Norman Lynaugh, and the Financial Director of Micromass, Paul Robinson. Them testimony related to the demand for sensitivity, and the development of the Hexapole and Ion Tunnel Quattro Ultimas. AB/Sciex also presented the video deposition of Micromass’s Intellectual Property Manager, Dr. David Yorke, in which Dr. Yorke explained the correspondence between the parties relating to the ICP device in 1997.

2. Micromass’s Noninfringement and Invalidity Presentations

Micromass’s first witness was Micro-mass Technical Director Dr. Robert Bateman. Dr. Bateman testified that Mi-cromass developed the Hexapole Quattro Ultima because it wished to improve upon the sensitivity of its prior device, the Quattro LC, and respond to AB/ Seiex’s API 3000. Dr. Bateman explained the changes he made to the Quattro LC, including increasing orifice sizes near the ion source, which increased the pressure in the subsequent chambers. This change to the ion source increased the sensitivity two and a half to three times. Dr. Bateman also made improvements to the detector, resulting in an increase in sensitivity of a factor of two. He also put small AC-only rods, called “stubbies,” on either side of the mass filters to improve ion transmission, added an additional vacuum chamber near the ion source, and added an additional ion guide structure. In the resulting Quattro Ultima, the pressure in the ion guide was typically around 1.5 to 2 torr, but Dr. Bateman operated the device at pressures as high as 10 torr. On cross-examination, Dr. Bateman admitted that the Hexapole Quattro Ultima operated at above 2.25 x 10 “2 torr cm. He also explained a later change to the device in which the size of one of the interchamber orifices had been decreased from 2.5 to 1.2 mm, a reduction of 77% of orifice area, but stated that 80% of ions continued to pass through the orifice.

Dr. Bateman also testified about the development of the Ion Tunnel Quattro Ulti-ma. Micromass developed that product after meetings with researchers doing similar work. They incorporated ion tunnels into the Quattro Ultima and found a 100% increase in ion signal intensity. Dr. Bate-man explained that ion tunnel had 84 rings that alternated in current. The rings created electrical fields in three dimensions; one dimension (z) in the axis of ion travel and two dimensions (x and y) perpendicular to that axis. Because the rings, unlike rod sets, created an axial electric field, there was a space exactly halfway between any two rings at which the AC voltage was zero. Thus, the ions would oscillate in the axial field, and, if they traveled too slowly, would become trapped at that halfway point of zero voltage.

Micromass then introduced the video deposition of Dr. Bruce Thomson, one of AB/Sciex’s chief research scientists, Laura Lauman, Micromass’s Executive Vice President in charge of marketing, and Dr. Scott Tanner, another AB/Sciex research scientist. Dr. Thomson’s testimony concerned the investigation he conducted on the Hexapole Quattro Ultima at a customer’s site in Nebraska to determine if the Quattro Ultima infringed. Lauman’s testimony concerned her understanding of Dr. Thomson’s results and what effect his discoveries concerning the Quattro Ultima, including the change to the ion source and increase in orifice size, would have on AB/ Sciex’s marketing. Dr. Tanner testified about ion traps and the fact that it was well known before the ’736 patent that an input of gas into an ion trap would cause collisional cooling. Dr. Tanner also testified about the differences between electrodes that create electric fields in three dimensions, such as an ion trap, and, those that do so in two dimensions, such as an ion guide rod set.

The next witness was Dr. Raymond March, Micromass’s expert witness. Dr. March explained why, in his opinion, the ’736 patent was invalid. Dr. March explained that the 1987 Smith article disclosed the structure of a two-stage mass spectrometer and that collisional cooling in ion traps was well-known prior to the ’736 patent. He also discussed an article he wrote concerning the operation of gas in ion traps and a collision effect on ions, R.F. Bonner & R.E. March, “The Effects of Charge Exchange Collisions On The Motion Of Ions In Three-Dimensional Quadrupole Electric Fields. Part II. Program Improvements And Fundamental Results,” 25 International Journal of Mass Spectrometry and Ion Physics 411 (1977) (the “1977 March article”). Dr. March also discussed a patent held by Dr. Enke, United States Patent No. 4,234,791 (the ’791 patent), in which Dr. Enke disclosed collision cell technology and referenced an earlier article by Dr. March concerning ion traps to support some of his conclusions. Dr. March also explained some of the collision cell prior art, including the Caldecourt article, AB/Sciex’s TAGA 6000, the French application, which used P x L products higher than the limit of the ’736 patent. Finally, Dr. March discussed an article by L. Hanley and S.L. Anderson, in which an ion guide is used and collisional cooling is observed. L. Hanley & S.L. Anderson, “Metal Cluster Ion Chemistry,” Proceedings of Quebec Symposium on Optical and Optoelectronic Applied Sciences and Engineering — Sub-symposium on Laser Applications in Chemistry, June 2-6, 1986 (the “Hanley and Anderson article”).

Dr. March also testified that the stacked rings of the Ion Tunnel were substantially different from rod sets because they created a three-dimensional electric field, like an ion trap.

Micromass then presented the deposition testimony of one of AB/Sciex’s principal research scientists, Dr. Covey. Dr. Covey explained that collision cells and ion guides are somewhat alike. Both use AC-only rod sets to guide ions through a space, but the collision cell also uses a high pressure air stream to dissociate ions into fragments.

Micromass also introduced the deposition testimony of Dr. French, one of the inventors of the ’736 patent. Dr. French stated that the structure of a two-stage ion guide was well-known at the time of the patent. He also discussed the prior art references he did not cite to the PTO, including the Caldecourt article and his French application. When presented with the French application, he admitted that although he prepared a Declaration during reexamination of the- ’736 patent concerning the French application, he did not remember seeing that application before preparation for his deposition in this-matter.

Micromass next introduced the deposition testimony of Dr. Davidson of AB/ Sciex. Dr. Davidson discussed AB/Sciex’s TAGA 6000, including its operating parameters and P x L product.

The next witness was Dr. David Yorke, Micromass’s Intellectual Property Manager. He testified about his correspondence with attorneys for AB/Sciex in 1997, his impression that the ’736 patent was made invalid by various prior art references concerning collision cells, and his belief that the patent would not be enforced.

Micromass next presented the deposition testimony of Dr. Allen, a research scientist with AB/Sciex. Dr. Allen’s testimony concerned the products initially manufactured by AB/Sciex containing the technology of the ’736 patent, including the API 300 and API 365. He explained that those devices realized only small improvements in ion sensitivity over their predecessor, the API III, which did not use the teachings of the ’736 patent. The API 300 and 365 had smaller pumps than the API III and were described by Dr. Allen as very different from their predecessor. Further development resulted in the API 3000 and API 4000, which each improved the sensitivities of their predecessors by factors of five to ten.

Micromass’s final witness was its damages expert, Raymond Sims. Sims explained that, if Micromass were found liable, an award of lost profits was not warranted because there was no demand for collisional focusing, there were non-infringing alternatives to the ’736 patent, and AB/Sciex was not likely to be able to make all the sales claimed by Dr. Stewart. Instead, he opined that a royalty of $6,000 per infringing device was reasonable, although significantly higher than other licensing patents in the industry. With a $6,000 royalty, the damages payable to AB/Sciex would be $1,242,000 in total, and $990,000 if only the Hexapole Quattro Ultima infringed.

3. AB/Sciex’s Rebuttal Presentation on Validity ■■

Following Micromass’s presentation of its defenses, including the affirmative defense that the ’736 patent was invalid, AB/ Sciex presented rebuttal testimony on the invalidity defense. AB/Sciex’s first witness was Dr. Enke. He addressed the pri- or art references that were discussed in the testimony of Dr. March, including the 1987 and 1988 Smith articles, the TAGA 6000, the French application, the Caldec-ourt article and his ’791 patent. Dr. Enke distinguished collision cell references, explaining that they were not directed to the problem the inventors sought to solve— guiding ions through an AC-only rod set. Rather, collision cells operated at high pressure to cause dissociation. He also distinguished the Hanley and Anderson article as related to reducing the spread of kinetic energies in metal ions, and not guiding ions into a mass filter. Dr. Enke then addressed ion trap references, including Dr. March’s 1977 article. He explained that while those articles discussed collisional cooling, they did not suggest using collisional focusing to improve the transmission of ions in an ion guide. Finally, Dr. Enke disputed Micromass’s contention that an ion tunnel was actually a series of ions guides. Dr. Enke explained that the purpose of ion tunnels is to guide ions through to a mass filter, and thus they do not trap ions between rings.

Micromass presented brief rebuttal deposition testimony from Dr. Neil Reed and Dr. Davidson of AB/Sciex. Those witnesses testified about the workings of the TAGA 6000.

4. The Jury’s Verdict

The jury returned a verdict that the claims of ’736 patent were not invalid. The jury found that Hexapole Quattro Ul-tima literally infringed the asserted claims of the ’736 patent and that the Ion Tunnel Quattro Ultima infringed those claims under the doctrine of equivalents. The jury awarded $47.5 million in damages to AB/ Sciex, $41.3 million of which was attributable to the Hexapole Quattro Ultima.

D. The Bench Trial

Following the jury trial, the court conducted a one-day bench trial on inequitable conduct and equitable estoppel. Micro-mass supplemented the evidentiary record on both subjects.'

1. Inequitable 'Conduct

Micromass contends that AB/Sciex misled the PTO during prosecution of the ’736 patent by failing to cite the French application as prior art and failing to introduce the results of experiments done by Dr. Douglas at P x L products below 2.25 x 10 ~2 torr cm. Micromass also contends that AB/Sciex misled the PTO during reexamination by failing to disclose that it had accused Micromass’s ICP device of infringing and by mischaracterizing the French application. Most of those topics were covered during the jury trial, but Micro-mass presented additional evidence to the court regarding Dr. Douglas’s low-pressure experiments.

During the jury trial, Dr. Douglas testified that he arrived at the P x L parameter of the ’736 patent because he did several experiments and found “interesting increases in ion signal consistently above a pressure of about 1.5 millitorr” using a rod length of 15 cm. “So 15 cm times 1.5 millitorr gives 2.25 x 10 ~2 torr cm.” He also recounted conducting at least one experiment at a pressure lower than 1.5 mil-litorr in which he recorded an increased ion signal, but described that result as “rather small.”

At the bench trial, Micromass presented deposition testimony from Dr. Douglas about these experiments conducted below the 2.25 x 10 ~2 torr cm threshold. In that testimony, Dr. Douglas recounts an increase in ion signal of 4.41 times when he increased pressure from 5.3 x 10 ^ torr to 1.27 x 10-3 torr. The 1.27 x 10-3 torr figure, assuming a 15 cm rod set, results in a P x L product of 1.8 x 10 ”2 torr cm. Micromass also submitted Dr. Douglas’s Laboratory Notebook No. 42, used between September 29, 1988 and October 28, 1988, which confirmed these results.

Micromass questioned Dr. Douglas about a 1992 article he wrote with Dr. French entitled “Collisional Focusing Effects in Radio Frequency Quadruples,” published in the Journal of the American Society for Mass Spectrometry. Dr. Douglas explained that Figure 5 of the article depicts experiments he did showing ion signal enhancement between 5.0 x 10 torr and 1.27 x 10 ~3 torr. Dr. Douglas could not remember, however, whether the data for those experiments discussed in the article resulted from tests he performed in 1988, before the ’736 patent application was filed, or whether those experiments were conducted later.

2. Equitable Estoppel

Micromass contends that AB/Sciex’s correspondence with it in 1997 led Micromass reasonably to believe that AB/Sciex agreed the claims of the ’736 patent were invalid and would not be enforced. Dr. Yorke testified during the bench trial to supplement the evidence he had already given concerning his interaction with AB/Sciex in 1997. He testified that after he responded to AB/Sciex’s January 10 letter with his own letter on April 16, 1997, he concluded that AB/Sciex agreed with him that “the patent was not valid” and that the ICP device “did not infringe any valid claim of the patent.” He then had no contact with AB/Sciex for three years, until the February 2000 filing of this suit. At some point during those three years, Dr. Yorke instructed the Micromass development team, including Dr. Bateman, “that the patent can be ignored.” He testified that he would not have so instructed the development team had AB/Sciex indicated to him that it believed the ’786 patent to be valid. Development of the Quattro Ultima was a $1.5 million project. Dr. Yorke testified that rather than risk losing this investment, the performance levels of the Quat-tro Ultima could have been achieved in “innumerable” other ways, including keeping the pressure low and improving the detector. Dr. Yorke testified that after Micromass learned that the ’736 patent would be asserted, it developed the Ion Tunnel Quattro Ultima, which it believed would not infringe.

On cross-examination, Dr. Yorke testified that he did not confirm whether AB/ Sciex continued to make fee payments to the PTO on the ’736 patent after 1997, although such information is publicly available. AB/Sciex also introduced deposition testimony from Dr. Yorke, in which he is asked to explain what he meant by his statement in the April 16, 1997 letter that the ICP instrument “does not infringe any valid claim of these patents.” AB/Sciex attempted to show that, by this statement, Dr. Yorke was ambiguous about whether he was stating that the ICP device did not infringe, or whether the claims of the ’736 patent were invalid. Dr. Yorke testified that he believed both.

II. MICROMASS’S MOTION FOR RECONSIDERATION OF THE . COURTS CLAIM CONSTRUCTION OPINION

Following the issuance of this court’s claim construction opinion, see Applera Corp. v. Micromass UK Ltd., 186 F.Supp.2d 487 (D.Del.2002), Micromass moved for reconsideration of the court’s construction of “first” and “second,” as those terms are used throughout claims 1 and 14. First and second are used in ’736 patent to designate the claim elements “first vacuum chamber” (30), “second vacuum chamber” (38), “first rod set” (32), “second rod set” (40), “first space” (between 32), and “second space” (between 40).

During claim construction, AB/Sciex argued that the use of first and second had no significance other than to identify and differentiate the different structures from one another. Micromass took the opposite position — that the “first vacuum chamber” must be the very first chamber, in the path of ion travel, with less than atmospheric pressure, and the “second vacuum chamber” must be the very next vacuum chamber. Because the Quattro Ultima had additional, empty vacuum chambers and used more than one ion guide, adoption of Mi-cromass’s claim construction would have foreclosed literal infringement.

While Micromass described its position as consistent with the plain meaning of “first” and “second,” the court rejected this assertion and noted that “first” and “second” are defined only by the context in which they are used. Furthermore, Mi-cromass’s construction would have excluded one of the preferred embodiments of the invention — -Figure 12, which depicted an additional, empty vacuum chamber that preceded “first vacuum chamber” (30) and “first rod set” (32).

Micromass also purported to find support for its construction of “first” and “second” in the prosecution history. During reexamination, AB/Sciex distinguished tandem mass spectrometers, such as that in the French application, on several different bases. One of the distinctions drawn by AB/Sciex was structural. It argued that tandem mass spectrometers were different from the claimed invention because the “first” rod set in the “first” vacuum chamber of a tandem mass spectrometer was an AC-DC mass filter, and not an AC-only rod set as required by the claims of the ,’736 patent. Likewise the “second” rod set and “second” vacuum chamber of a tandem mass spectrometer was an AC-only collision cell, whereas the claims of the ’736 patent require an AC-DC mass filter.

Based on this distinction, the court concluded that AB/Sciex was using “first” and “second” as more than mere identifiers of separate elements, but as descriptions of location in the path of ion travel. It therefore declined to adopt AB/Sciex’s claim construction of these terms in full. However, the court concluded that “first” and “second” only required that, of the two claimed elements, the “first rod set” in the “first vacuum chamber” (the ion guide) must precede the “second rod set” in the “second vacuum c