Elf Atochem North America, Inc. v. Libbey-Owens-Ford Co.

894 F.Supp. 844 (1995)

ELF ATOCHEM NORTH AMERICA, INC., Plaintiff,
v.
LIBBEY-OWENS-FORD COMPANY, INC., Defendant.

Civ. A. No. 94-670-RRM.

United States District Court, D. Delaware.

August 4, 1995.

Arthur G. Connolly, Jr., Connolly, Bove, Lodge & Hutz, Wilmington, DE, John C. Lowe, Richard B. Racine, Michael R. McGurk, James B. Monroe, Finnegan, Henderson, Farabow, Garrett & Dunner, Washington, DC, Stanley A. Marcus, Elf Atochem North America, Inc., Philadelphia, PA, for plaintiff.

Jack B. Blumenfeld, Morris, Nichols, Arsht & Tunnell, Wilmington, DE, J. Frank Osha, Thomas J. Macpeak, Sheldon I. Landsman, Kenneth J. Burchfield, Scott M. Daniels, Brett S. Sylvester, Sughrue, Mion, Zinn, Macpeak & Seas, Washington, DC, Alan L. Briggs, Squire, Sanders & Dempsey, Washington, DC, for defendant.

McKELVIE, District Judge.

This is a patent case. Elf Atochem North America, Inc. ("Atochem"), the owner of United States Patents 4,590,096 ("the '096 patent") and 4,696,837 ("the '837 patent"), directed towards methods for producing energy efficient glass, alleges Libbey-Owens-Ford Co., Inc. ("LOF") is willfully infringing certain claims of its patents. LOF denies liability and has counterclaimed for a declaratory judgment that the claims of the patents are invalid, not infringed, and unenforceable. This matter is scheduled for a ten-day jury trial beginning on December 4, 1995.

Atochem and LOF have filed cross-motions for a partial summary judgment on the issue of infringement of claims 1, 2, and 3 of the '096 patent. On July 17 and 18, 1995, the court held an evidentiary hearing to resolve the meaning of the disputed terms in claims 1, 2, and 3 of the '096 patent. This Opinion constitutes the court's findings on the meaning of the disputed terms in independent claim 1 of the '096 patent, as well as claims 2 and 3 which depend from claim 1.

FACTUAL BACKGROUND AND PROCEDURAL HISTORY

A. Procedural History of this Action

On December 15, 1994, Atochem filed a complaint in this court alleging LOF has infringed certain claims of the '096 and the '837 patents and demanding a jury trial. On February 8, 1995, LOF filed an answer and counterclaim denying liability and seeking by way of a counterclaim a declaratory judgment of invalidity, unenforceability, and non-infringement.

On April 25, 1995, Atochem filed a motion for a partial summary judgment of infringement of claims 1, 2, and 3 of the '096 patent. On June 1, 1995, LOF filed a cross-motion for a partial summary judgment of non-infringement of claims 1, 2, and 3 of the '096 patent.

B. The Technology and United States Patent 4,590,096

The '096 patent relates to a process to make energy efficient glass called Low-E or low-emissivity glass. Uncoated glass permits both visible light and heat waves to pass through it. However, by coating glass with certain chemicals the glass can be made to reflect the heat waves while still permitting the visible light waves to pass through. This allows one to see through the glass while the glass retains heat. When coated glass is used in a building it helps retain heat energy inside the building.

One process used to apply the thin coating to the glass is called chemical vapor deposition or "CVD." In a CVD process, a gaseous mixture of chemicals is brought into contact with a hot glass surface, causing the desired coating to be deposited on the surface of the glass by a chemical decomposition or reaction called "pyrolysis."

On May 20, 1986, the United States Patent and Trademark Office ("PTO") issued the '096 patent to George H. Linder from an application filed on December 28, 1984. He assigned his rights to that patent to M & T Chemicals Inc., the predecessor company to Atochem, who, in turn, has apparently assigned its rights to the '096 patent to Atochem.

The '096 patent is directed towards a method of producing energy efficient glass by forming fluorine-doped tin oxide coatings on that glass using a CVD process. The patent contains 10 claims. Claim 1 reads, in relevant part:

1. A process for producing a coated glass having a surface resistance of less than about 40 ohms/square at a thickness less than about 250 nm, an IR reflectivity of greater than about 70%, visible transmittance of greater than about 80% and solar transmittance of greater than about 70%, comprising utilizing a chemical vapor deposition process wherein a coating solution is introduced into a carrier gas stream, said solution comprising a substantially solvent free mixture of an organotin chloride and a reactive fluorine compound soluble in or miscible with said organotin chloride,

....

[T]he reactive flouring compound is an organic compound having at least one fluorine atom located alpha or beta to a functional group, wherein the functional group is selected from the group consisting of carboxylic acid, ester, aldehyde, anhydride, nitrile, ketone, ether, amine, acid halide, alcohol, halogen, amide and hydrogen; said mixture having a vaporization temperature below about 200° C. and a decomposition temperature substantially above the vaporization temperature of the mixture and below about 650° C.; and the gas stream contains sufficient water vapor such that the relative humidity of the gas stream at 18° C. is about 6% to about 100%.

Claims 2 and 3 read:

2. The process according to claim 1 wherein the organic fluoride dopant is trifluoroacetic acid, ethyl-4,4,4-trifluoroacetoacetate, perfluorobutyl iodide, trifluoroacetic anhydride, or ethyltrifluoroacetate.

3. The process according to claim 1 wherein the organotin compound is monobutyltin trichloride, methyltin trichloride isobutyltin trichloride, butyl dichlorotin acetate, butyldichlorotin dicetate, diisobutyltin dichloride, methyltin trichloride, demethyltin dichloride, dibutyltin dichloride di-t-butyltin dichloride or tin tetrachoride.

C. The Parties and the Market

Three United States companies, LOF, AFG and PPG, manufacture most of the low-E glass sold in the United States and produced using a CVD process. Atochem has granted licenses to AFG and PPG under its patents.

Atochem does not manufacture or sell coated glass. It grants licenses to other manufacturers who in turn manufacture coated glass using Atochem's chemicals for use in those licensed processes.

D. Motions for a Partial Summary Judgment

Both parties have filed motions for a summary judgment on infringement of claims 1, 2, and 3 of the '096 patent. In its motion, Atochem contends no genuine issues of material fact remain as the parties do not dispute the facts surrounding LOF's accused process and the meaning of the words in the claims is a question of law. According to Atochem, if this court adopts its interpretation of the claims it must find infringement.

In its motion, LOF also alleges no genuine issues of material fact remain as the parties do not dispute the facts surrounding its accused process and the meaning of the words in the claims is a question of law. According to LOF, if this court adopts its interpretation of the claims it must find no infringement.

Atochem and LOF dispute three portions of claim 1, the independent claim at issue. First, they dispute the meaning of the words "coating solution." Second, they dispute the meaning of the words "the gas stream contains sufficient water vapor such that the relative humidity of the gas stream at 18° C. is about 6% to about 100%." Finally, they dispute the meaning of the words "a coating solution is introduced into a carrier gas stream."

The evidence cited by both parties in support of their motions for a summary judgment included the patent, technical dictionaries, chemistry textbooks, statements by witnesses in depositions, and affidavits by technical expert witnesses. For example, Atochem contends a "coating solution" would be understood by one skilled in the art to include mixtures of gases, along with liquids and solids. In support of this position, Atochem cites to the affidavit of Dr. Karl E. Spear. In that affidavit, Dr. Spear states: "By any scientifically accepted definition, a `solution' can be composed of a gas dissolved in a gas." Dr. Spear then refers to a definition of "solution" that appears in Hawley's Condensed Chemical Dictionary (12th ed. 1993): "A uniformly dispersed mixture at the molecular or ionic level, of one or more substances ... in one or more other substances...." Dr. Spear explains that this dictionary is well-recognized in the field. He points to no language in the specification of the '096 patent which supports or contradicts his position.

In its cross-motion for a summary judgment (and answering brief to Atochem's motion), LOF contends the words "coating solution" as used in the '096 patent are limited to a mixture of two liquids and do not encompass a mixture of two gases. In support of its position, LOF cites to a sworn affidavit from Dr. Roy Gordon. Dr. Gordon states:

The language of Claim 1 itself makes it clear that the "solution" claimed in Claim 1 is a liquid. The express limitation that the solution is "substantially free of solvent" is meaningless in the gas phase, because in the gas phase, all of the mixed gases are freely miscible and the gases are therefore all "solvents" for each other.

Dr. Gordon then states: "The common and plain meaning of the term `solution' as a liquid is reinforced by the fact that every example of the '096 patent relates to a liquid combination of tin compound and dopant...."

In response to this affidavit, Atochem cites to an additional affidavit of Dr. Spear. In that sworn affidavit, he states:

A person of ordinary skill in the art would not read the language of Claim 1 of the '096 patent ... to indicate that the "solution" claimed is limited to a liquid, but rather that the solution claimed encompasses, at least, a liquid in a liquid, a solid in a liquid, or a gas in a gas, depending on which dopant and organotin are being used. The term "substantially free of solvent" has meaning in the gas phase to a person of ordinary skill in the art, because such person would know that gaseous components in the gaseous coating mixture stream are not typically referred to in the art as solvents. Consequently, such a person would know that the term "substantially solvent free" refers to the substantial absence of any solvent used with the dopant or organotin before injection into the gaseous stream.

Dr. Gordon further states: "`Solution,' as used in Claim 1 and as used commonly, is not limited to liquid solutions." Finally, he explains: "As a person of ordinary skill in the art, I know that the common and plain meaning of the term `solution' is as a gas, liquid, or solid dissolved in a solid; a gas or liquid or solid dissolved in a liquid; or a gas dissolved in a gas, depending upon the context."

Atochem also cites to a number of references in support of its position. For example, The Random House Dictionary of the English Language defines "solution," in part, as "a homogeneous, molecular mixture or two or more substances." Physical Chemistry defines "solution" as "a homogeneous mixture; that is, a solution is a one-phase system with more than one component. The phase may be solid, liquid, or gas."

In response to this information, LOF attaches another affidavit of Dr. Gordon. In that sworn affidavit, Dr. Gordon refers to a Compendium of Chemical Terminology published by International Union of Pure and Applied Chemistry which defines a "solution" as follows:

A homogeneous liquid or solid phase containing more than one substance, when for convenience one of the substances, which is called the solvent and may itself be a mixture, is treated differently from the other substances, which are called solutes.

In view of this evidence, LOF argues the more conventional definition of "solution" is limited to a mixture of liquids and excludes gaseous mixtures.

The foregoing is an example of the factual presentation made as to just one of the three portions of claim 1 of the '096 patent disputed by the parties in their cross-motions for a summary judgment. The paper record was similar with respect to the other two disputed portions of the claim.

E. This Court's Prior Practice on Summary Judgment in Patent Cases

This court has indicated it will grant a motion for a summary judgment only where no genuine issue of material fact remains for resolution, consistent with Federal Rule of Civil Procedure 56, the Federal Rules of Evidence, and Supreme Court precedent. Anderson v. Liberty Lobby, 477 U.S. 242, 248, 106 S.Ct. 2505, 2510, 91 L.Ed.2d 202 (1986). A dispute is not genuine where no reasonable jury could find for the non-moving party under the appropriate burden of persuasion. Id.

As one court has stated:

Summary judgment, with ever-lurking issues of fact, is always a treacherous shortcut and, in cases like these, too fragile a foundation for so heavy a load. Such relief is always discretionary, and in cases posing complex issues of fact and unsettled questions of law, sound judicial administration dictates that the court withhold judgment until the whole factual structure stands upon a solid foundation of a plenary trail where the proof can be fully developed, questions answered, issues clearly focused and facts definitively found.

In re Bloomfield S.S. Co., 298 F.Supp. 1239, 1242 (S.D.N.Y.1969), aff'd, 422 F.2d 728 (2d Cir.1970). That is, in some instances, this court is hesitant to rule as a matter of law on issues presented only with the cold paper record and possibly oral argument from counsel. Rather than schedule a summary judgment hearing with live testimony which usually results in increased time and expense for the parties and court, this court will deny the motion for a summary judgment and take the matter to trial where the record can be fully developed and enhanced by live testimony and exhibits.

The fact that a motion for a summary judgment asks this court to construe language in a claim of a patent would not change this court's approach. In such cases, this court would interpret a claim without the aid of a jury where the meaning of the claim was unambiguous.

However, where affidavits from expert witnesses cited in support or opposition to a summary judgment motion present equally plausible interpretations of disputed words in a claim, this court would deny summary judgment and allow a jury to resolve the meaning of the disputed terms based upon all of the evidence including the credibility and bias of the expert witnesses. In so doing, this court is cognizant of Judge Easterbrook's comment that "judges should not pretend that all nominally `legal' issues may be resolved without reference to facts" and that "[w]hat seems clear to a judge may read otherwise to [one skilled in the art]." In re Mahurkar Double Lumen Hemodialysis Catheter Patent Litig., 831 F.Supp. 1354, 1359 (N.D.Ill.1993) (Easterbrook, J., sitting by designation). And where it is clear that no reasonable jury could have rendered the decision that it did, this court consistent with the Federal Rules of Civil Procedure would grant judgment as a matter of law after the verdict.

Here, the court reviewed the papers submitted by the parties and determined that each party presented colorable contentions regarding the meaning of certain words in the claims of the '096 patent. In fact, the papers appear to demonstrate a genuine dispute of material fact over the proper interpretation to be accorded words such as "coating solution." Consequently, this court normally would have denied the motions for a summary judgment and permitted the parties to try the issues to a jury.

F. Markman v. Westview

On April 5, 1995, the Federal Circuit issued a decision in Markman v. Westview Instr., Inc., 52 F.3d 967, 979 (Fed.Cir.1995) (en banc), in which the court stated that "in a case tried to a jury, the court has the power and the obligation to construe as a matter of law the meaning of language used in the patent claim." As a rationale for that result, the court explained that competitors should be assured that a judge "trained in the law" will examine the patent and the prosecution history in a way to "arrive at the true and consistent scope of the patent owner's rights to be given legal effect," particularly in the light of the serious consequences attendant to a finding of infringement. Id. The court further indicated that it has "long been and continues to be a fundamental principle of American law that the construction of a written evidence is exclusively for the court." Id. at 978. Thus, the court concluded that the meaning of disputed technical words in patent claims is "strictly a question of law for the court," stating a clear policy preference to have disputes over technical words in a patent claim resolved by judges. Id. at 977.

The "obligation" created by the Federal Circuit to instruct the jury on the meaning of the words used by an inventor in a claim basically leaves a district court with three options. The court can attempt to resolve these disputes on the paper record. Second, the court can hold a trial to resolve the disputes. Finally, the court can wait until trial and attempt to resolve claim disputes the evening before the jury must be instructed.

In this case, the court determined that it required a trial to resolve the disputes over the claim language in the '096 patent. On June 16, 1995, the court entered an Order scheduling a two-day bench trial to resolve the meaning of the disputed terms in the claims and requiring the parties to submit a pre-trial order in accordance with the local rules of this court. On July 13, 1995, the court held a pre-trial conference in this matter and on July 17 and 18, 1995, the court held a two-day bench trial to establish a record for the resolution of the disputes over the meaning of the words in the claims of the '096 patent.

G. The Markman Trial

During the two-day trial, the court heard evidence on the three disputed portions of claim 1 of the '096 patent the parties identified in their briefing on the motions for a summary judgment. In support of their respective positions on the meaning of the disputed technical terms in the patent claim, both Atochem and LOF called technical expert witnesses to establish the level of skill in the relevant art and to offer their expert opinions on how one of ordinary skill in the art would understand the disputed words. LOF also called one of its scientists to testify on the '096 patent and LOF's accused CVD process.

1. The Experts and the Person of Ordinary Skill in the Art

Recognizing that words in patent claims are directed to people of ordinary skill in the relevant art, the parties first set out to establish both the art and the qualifications of the person of ordinary skill in that art. Atochem called Dr. Karl E. Spear, a Professor of Ceramic Science at Penn State University to testify at this trial. Dr. Spear has been associated with Penn State for 25 years. He holds a Bachelor of Science in Mathematics from Baker University in Kansas, and a Ph.D. in Chemistry from the University of Kansas. According to Dr. Spear, during the period from 1981 to the present, he has been a person of ordinary skill in the art relating to chemical vapor depositions and pyrolytic chemical processes, particularly relating to coatings on glass.

However, on cross-examination, Dr. Spear indicated that none of the publications listed on his curriculum vitae relate specifically to the area of fluorine doped tin oxide coatings. Dr. Spear explained that his only work with fluorine doped tin oxide coatings was a brief consulting engagement with M & T Chemicals (the predecessor to Atochem) in 1985. In fact, Dr. Spear testified that prior to 1985 and from 1986 until he was retained in this matter, he had done no work with respect to fluorine-doped tin oxide coatings.

LOF called Dr. Roy Gordon to offer technical expert testimony supportive of its interpretation of the words in the claims. Dr. Gordon is a Professor of Chemistry at Harvard University and the former Chairman of the Department of Chemistry at Harvard. He holds a Bachelor of Science in Chemistry and Physics from Harvard College, a Masters degree in Physics and a Ph.D. in Chemical Physics from Harvard University. Dr. Gordon has been a professor at Harvard since 1966. For the last 20 years, Dr. Gordon has focused his research on chemical vapor deposition of thin films, and particularly on fluorine doped tin oxide coatings. Dr. Gordon has extensive experience in experiments with fluorine-doped tin oxide coatings on glass and in fact, personally built a laboratory to study such reactions. Dr. Gordon holds approximately 12 United States patents relating to fluorine-doped tin oxides. In addition, Dr. Gordon has authored about ten articles in this field and received an R & D 100 Award in 1991 for his development of LOF's fluorine-doped tin oxide glass, known as Energy Advantage Glass.

As to the relevant art, Dr. Gordon testified that he considered the relevant art area to be fluorine-doped tin oxide coatings on glass. One of ordinary skill in this art would hold at least a masters degree in chemistry, physics or materials science and have five years of experience working in a laboratory producing fluorine-doped tin oxide coatings on glass. Dr. Gordon was aware of many such individuals. Dr. Gordon also indicated he considered himself one of extraordinary skill in the art. However, he explained that his opinions at trial were given with reference to how one of ordinary skill in the art would read and understand the terms in the patent.

On cross-examination, counsel for Atochem questioned Dr. Gordon about his potential bias in this action. In addition to possessing substantial expertise in the area of fluorine-doped tin oxide coatings on glass, Dr. Gordon explained that he also has a substantial and longstanding relationship with LOF. LOF is the exclusive licensee of his patents in this area. Last year, Dr. Gordon received more than $600,000 in royalties from LOF. Furthermore, the Energy Advantage Glass he developed is one of the products accused of infringing Atochem's patents in this action. For his services in connection with this suit, LOF is paying Dr. Gordon $250 per hour.

2. Evidence Regarding "Coating Solution"

The parties dispute the meaning of the words "coating solution" in claim 1 of the '096 patent. Atochem contends these words encompass a mixture of two gases. LOF contends these words would be understood by one skilled in the art to refer to a mixture of liquids and not to include a mixture of gases.

Atochem called Dr. Spear to testify on this issue. He testified that the word "solution" as understood by one skilled in the art would include a mixture of two gases. In fact, the term "solution" is so well known and common that the term will not appear in a glossary of terms and definitions for chemical vapor deposition being prepared by a task force within the International Union for Pure and Applied Chemistry ("IUPAC"). Dr. Spear, having read the specification and prosecution history of the '096 patent, found nothing that would suggest to one skilled in the art that the term "solution" was used in the patent to refer only to liquid solutions.

Claim 1 of the '096 patent indicates that "a coating solution is introduced into a carrier gas stream, said solution comprising a substantially solvent free mixture of an organotin chloride and a reactive fluorine compound soluble in or miscible with said organotin chloride." Dr. Spear explained that solutions are comprised of a solvent and a solute. The species that is present in the highest concentration is usually referred to as the solvent. The solute then is normally the species present in the lower concentration. As Dr. Spear interpreted this language in claim 1, he found no inconsistency between the word "solution" and "substantially solvent free" in the same passage. In his opinion, the words "substantially solvent free" referred to any solvent used to dissolve either the organotin chloride or the reactive fluorine compound. That being the case, these words would be satisfied so long as the final product being applied to the glass was "substantially" free of the solvent used to dissolve one of the precursors. According to Dr. Spear, based on information in the specification of the '096 patent, so long as the final product applied to the gas contains no more than 25% of a solvent used to dissolve one of the precursors it would satisfy this claim.

Dr. Spear also testified on a European Patent Application offered by LOF as relevant to the issue of the proper interpretation of "coating solution." LOF offered EP0027403 ("the EP '403 application"), which according to its English translation was published on April 22, 1981, and describes a chemical vapor deposition process to make fluorine-doped tin oxide films on glass. In particular, Dr. Spear testified that the application did not teach one skilled in the art the process described in claim 1 of the '096 patent as he interpreted it. For example, Dr. Spear explained that one skilled in the art, upon studying the patent, would understand that the gas stream described and depicted in that application would have no humidity. While the application did not expressly indicate that the air brought into the system was cleaned and dried, and thus would lack all humidity, the air would nonetheless have been cleaned and dried because of the sensitive nature of CVD processes at that time. The absence of a description or diagram showing a dryer or filter in the EP '403 application did not surprise Dr. Spear. He explained that one skilled in the art would not include all obvious details in schematic drawings depicting processes.

In response, Dr. Gordon testified that a "coating solution" as those words are used in the '096 patent and understood by one skilled in the art would be a liquid mixture. In his opinion, those words would not describe a gaseous mixture. He based his opinion on other words in the claim, what he believed to be the more generally accepted definition of the term "solution," and information in the '096 patent specification.

Dr. Gordon agreed that the general definition of "solution" is any homogeneous mixture and includes mixtures of solids, liquids and gases. In fact, when shown several textbooks by counsel for Atochem he acknowledged that they included a mixture of gasses when describing a solution. However, he testified that the more common definition of a solution did not include a mixture of gases. Dr. Gordon also testified that the references attached to and cited in his affidavits submitted with LOF's motion for a summary judgment were consistent with his view of the more common understanding of the term solution.

Furthermore, he indicated this broad definition provided little assistance in understanding the scope of the term "solution" as used in claim 1 of the '096 patent. For example, Dr. Gordon believed the inventor did not envision solid solutions and in fact, rejected a prior art method of introducing solid particles into the carrier gas. In his opinion, in the field of chemical vapor deposition of fluorine doped tin oxide coatings, those of skill in the art most commonly understood "coating solution" to be a liquid solution.

Dr. Gordon testified that other words in claim 1 support his opinion. For example, claim 1 indicates that the "coating solution" is comprised of at least 2 components, an organotin chloride and a reactive fluorine compound. The claim then indicates that the second component or precursor (the reactive fluorine compound) must be "soluble in or miscible with" the organotin chloride. According to Dr. Gordon, had the inventor intended "solution" to encompass the combination of two gasses he would not have indicated that the gaseous reactive fluorine compound be "soluble in or miscible with" the gaseous organotin chloride. He reasoned that because all gasses are soluble in or miscible with each other the words have no meaning with respect to gasses.

Dr. Gordon also testified that the specification supports his opinion that a "coating solution" as used in the '096 patent is limited to liquid solutions. First, at Column 3, Line 55 of the '096 patent, in the section titled "Description of the Invention," the specification reads: "The compositions of this invention contain a liquid or low melting organotin compound...." Dr. Gordon believed the inventor was concerned with finding a liquid form of the tin source. Also, one skilled in the art would recognize that there are no organotin chlorides which exist as gases at room temperature.

At Column 4, Line 61 of the '096 patent, the specification indicates that where the two precursors are not mutually miscible one can add a third component in relatively small amounts to form a homogeneous mixture. This language, according to Dr. Gordon, has meaning only with reference to liquid solutions.

The patent also indicates at Column 4, Line 68, that "[b]oth the organotin halides and the fluoride compounds of this invention are ordinarily both liquids and miscible with one another." Furthermore, according to Dr. Gordon, one skilled in the art looking at the tables included in the patent and particularly at the rate of solution addition would recognize that at the rates listed, only liquid solutions would contain enough of the active ingredients to successfully coat the glass. If the solution were gaseous, not enough material would be deposited in the time period envisioned.

Finally, at Column 10, Line 16, the specification refers to the use of co-solvents to keep the coating solution stable at low temperatures. In some cases, a pair of liquids may be miscible with each other at room temperature but when cooled will separate into two layers. For example, if the coating solution is shipped in the winter it may arrive at the buyer in two layers. When it is then pumped out of the drums, depending upon where the hose is in the drum, one would get different materials rather than a solution. According to Dr. Gordon, a co-solvent can only be used to prevent the separation of two liquids. Where two gasses are used, apparently a co-solvent would have no effect on this so-called phase separation. That is, where two gases were cooled to the point of separation, a co-solvent could not prevent the immiscibility.

Dr. Gordon also looked to the Abstract portion of the specification of the '096 patent to further support his opinion on how one skilled in the art would understand the term "coating solution." The relevant portion of the Abstract of the '096 patent reads: "A chemical vapor deposition method for forming fluorine-doped tin oxide coatings uses a liquid coating composition which includes an organic fluorine dopant and an organotin compound." It then goes on to state: "A preferred liquid coating composition is monobutyltin trichloride and trifluoroacetic acid." While the claim uses the words "coating solution" and not "coating composition," they are both defined by claim 1 and the Abstract respectively as the combination of the organic fluorine dopant and an organotin chloride. Dr. Gordon could find no instance in the '096 patent where the term "solution" was used expressly or implicitly to encompass a gas-in-gas mixture. In fact, in each and every instance in which the term "solution" is used in the specification it refers either to a liquid or to a composition that functions in a way only a liquid solution could.

In addition to other words in claim 1, the specification, and knowledge of those skilled in the art, Dr. Gordon also testified about a European Patent application which in his opinion lent further support to his opinion that a "coating solution" must be a liquid/liquid mixture. Dr. Gordon testified that the EP '403 application teaches the invention in claim 1 of the '096 patent, as Atochem interprets the words in claim 1. For example, EP '403 shows air entering the system through a compressor. In Dr. Gordon's opinion the detail in the drawing and his knowledge of the technology would indicate that the air entering the compressor would be ambient air, even though the patent application is silent on this issue. The ambient air, in turn, would have a relative humidity of between 6% and 100% at 18° C.

Dr. Gordon testified that the EP '403 application related to the proper interpretation of "coating solution" because it discloses a process which introduces a gaseous coating solution into the carrier gas stream. According to Dr. Gordon, claim 1 of the '096 patent would describe the process disclosed in the EP '403 application if "coating solution" is interpreted to include a gas-in-gas mixture.

On cross-examination, counsel for Atochem questioned Dr. Gordon on the statement in the Description of the Invention, at Column 3, Line 58, "The dopants of this invention work equally as well if vaporized separately from the organotin." In particular, counsel for Atochem question Dr. Gordon as to why this phrase did not suggest that a "coating solution" was contemplated by the inventor as including a gas-in-gas. Counsel further inquired of Dr. Gordon why he had not referred to this statement when discussing references in the specification to gaseous solutions. Dr. Gordon responded that separately vaporizing the reactants was the most common way to practice a CVD process. So common, in fact, that in Dr. Gordon's opinion, it was disclosed but not claimed and not relevant to determining the meaning of "coating solution" in claim 1.

3. Evidence as to "the Gas Stream Contains Sufficient Water Vapor Such That the Relative Humidity of the Gas Stream at 18° C. is about 6% to About 100%"

In addition to "coating solution," the parties dispute the meaning of the words "the gas stream contains sufficient water vapor such that the relative humidity of the gas stream at 18° C. is about 6% to about 100%," which appear in claim 1 of the '096 patent. Atochem contends these words merely establish a lower limit for the amount of water which must be found in the gas stream. LOF contends that these words describe both a lower and upper limit on the amount of water in the gas stream.

Dr. Spear testified for Atochem on this issue. Dr. Spear explained that one skilled in the art would understand "sufficient water vapor" to mean that amount of water required to produce a coating with the properties specified in the claim. One skilled in the art would not understand "sufficient" to place a maximum limit on the amount of water that might be used in the system. The optimum amount of water for the process would be determined through experimentation. In fact, optimization is common in chemical vapor deposition processes because each system is unique.

Counsel for plaintiff indicated that the claim could be rewritten to say "sufficient water vapor such that the relative humidity of the water vapor is at least 6% at 18° C." Yet, according to Dr. Spear, one skilled in the art would not read these open-ended words to describe an infinite amount of water. At some point, the entire system would be comprised of water. However, Dr. Spear offered his opinion that this limitation could be satisfied by a system which had eight times as much water as needed to reach 100% humidity at 18° C., because the system was running at a higher temperature. Hotter air can absorb a greater amount of water. Therefore, were air containing 5 mole percent of water at a temperature above 18° C. to be cooled to 18° C., one would have air with a relative humidity of 100% and a puddle of water.

On cross-examination, Dr. Spear testified that the corresponding mole percent of water at 18° C. and 100% relative humidity is 2 mole percent. In other words, measured in moles (6.02 × 1023), 2 percent of the moles are water molecules at this temperature. In each example discussed in the patent specification, where the table indicated that the relative humidity at 18° C. was 100%, it corresponded to 2 mole percent of water and no more. According to Dr. Spear, the claim range sets the starting point for experimentation at the optimum amount of water needed to create a coating on glass which will produce glass with the properties set forth in the claim. That is, an amount of water corresponding to between 6 and 100% relative humidity at 18° C. would produce the desired properties. However, the words also describe greater amounts of water. While Dr. Spear agreed that the claim would not describe a system with an infinite amount of water, the exact upper limit on the amount of water would have to be determined through experimentation.

Dr. Gordon presented a different view of the meaning of the words "the gas stream contains sufficient water vapor such that the relative humidity of the gas stream at 18° C. is about 6% to about 100%." First, Dr. Gordon explained that relative humidity defines the water content of air at a certain temperature. At higher temperatures, air can hold more water vapor. Thus, the amount of water in air at 100° C. and 50% relative humidity is greater than the amount of water in air at 18° C. and 50% relative humidity.

To provide meaning to those words, Dr. Gordon explained how the inventor described the two methods of carrying out his invention at Column 5, starting at Line 26 of the '096 patent specification. He explained that both methods are very similar. In the first method, the inventor pulls air from the laboratory into the apparatus using a compressor. The air stream then divides into two measured gas streams. One of the gas streams passes through a dryer to remove all of the moisture from the air. That is, the gas stream coming out of the dryer will have zero percent relative humidity. The other stream passes through a humidifier or "water trap." This gas stream will have a relative humidity of about 100%. Furthermore, 18° C. equates to about 65° F., which approximates the temperature on a normal day in the laboratory. Therefore, the temperature of the water trap can be approximated to be 18° C. On those days when the temperature in the laboratory is a few degrees higher or lower, Dr. Gordon explained that a simple conversion chart can be used to normalize the relative humidity to 18° C. The wet and dry gas streams are then rejoined into a single gas stream.

With one gas stream at 0% relative humidity and the other gas stream at 100% relative humidity and the laboratory at 18° C., the operator, by using valves, can control the relative humidity of the gas stream which will then interact with the coating solution. To obtain a gas stream of 50% relative humidity at 18° C., the operator simply adjusts the valves governing the wet and dry gas streams such that the flow rates are equal. To obtain a gas stream of 0% relative humidity, the operator would simply shut off flow from the wet gas stream. Similarly, to obtain a gas stream at 100% relative humidity the operator would simply shut off the dry gas stream. The second method described by the inventor simply added a hygrometer after the wet and dry gas streams are joined to more accurately measure the relative humidity of the gas stream.

With that discussion from the specification, Dr. Gordon testified that in his opinion the minimum amount of moisture in the system as described by claim 1 of the '096 patent is 6% relative humidity at 18° C. The maximum amount of moisture in the system, which occurs when the operator turns off the dry gas stream, is 100% relative humidity at 18° C. According to Dr. Gordon, the inventor's discussion of the two methods of carrying out his invention indicate that only one source of introducing moisture into the system was contemplated, so that the words "to about 100%" have a definite meaning as an upper, maximum limit on the amount of moisture in the system. Furthermore, the inventor did not discuss the possibility of heating the water in the humidifier used to produce the wet gas stream. Therefore, "18° C." also has a definite meaning and limitation to Dr. Gordon. Thus, in Dr. Gordon's opinion, the words "gas stream contains sufficient water vapor such that the relative humidity of the gas stream at 18° C. is about 6% to about 100%" in claim 1 when read in light of the specification described a definite upper and lower limit on the total amount of moisture in the system, with the upper limit being equal to 2 mole percent of water.

4. Evidence as to the Words "Wherein a Coating Solution is Introduced Into a Carrier Gas Stream"

Third, the parties dispute the meaning of the words "wherein a coating solution is introduced into a carrier gas stream." In particular, Atochem urges that these words do not describe a system where the components of the coating solution must be introduced together as a solution into the carrier gas stream, that is, simultaneously. LOF, however, contends these words plainly require a "coating solution" to be "introduced" into the carrier gas stream. If the components of the solution are introduced separately, no solution is introduced.

Dr. Spear testified for Atochem on this issue. Dr. Spear explained that one skilled in the art would understand the words "A coating solution is introduced into a carrier gas stream" to mean that the coating solution is added to or combined with the carrier gas. Nothing in the language of the claim would alert a person of ordinary skill in the art that the elements of the coating solution had to be combined with each other prior to being combined with the gas stream and that the elements could not be joined in the gas stream at a connection and then proceed together so long as they are mixed prior to contacting the glass.

Dr. Gordon testified for LOF in support of its position. Dr. Gordon explained that the claim language clearly indicate one point of entry for the elements of the coating solution into the system. According to Dr. Gordon, nowhere in the claim does it indicate that the coating components are separately introduced or that the coating solution is broken up into separate parts. There exists a liquid coating solution, fully mixed, prior to introduction into the gas stream.

On cross-examination, counsel for Atochem questioned Dr. Gordon about language in the patent specification which describes separate introduction of the components of the coating solution. For example, Dr. Gordon was asked whether the language, "If a gaseous dopant or a separate liquid component is desired, it may be fed into the air mixture at a point before the first heated oil bath," which appears at Column 5, Line 40 of the '096 patent, describes the separate introduction of the constituents of the coating solution. According to Dr. Gordon, with this phrase, the inventor is describing a process well-known in the art and not claimed as one skilled in the art would understand the claims. Counsel for Atochem also asked Dr. Gordon about language appearing at Column 7, Line 4: "When the novel compositions are deposited by vapor deposition ("CVD"), Method A or Method B, the organotin dopant and solubilizer, if used, may be mixed at room temperature or added separately." Dr. Gordon again explained that the inventor was describing a well-known process not claimed. However, in each instance, Dr. Gordon agreed that the inventor was describing a process where the constituents of the coating solution were introduced separately into the carrier gas stream.

5. Evidence as to LOF's Accused CVD Process

In addition to presenting testimony and documents regarding the scope and meaning of the claim terms and a European patent application, LOF presented testimony and documents regarding its accused CVD process. LOF called both Dr. Gordon and Dr. Michel Soubeyrand, an LOF employee, to testify on this issue.

In his testimony, Dr. Gordon explained the process used by LOF to make its low-E glass. In general, LOF uses a CVD process to produce fluorine-doped tin oxide coatings on flat glass. LOF starts with dimethyltin dichloride, a tin compound, present in a storage vessel. Liquid tin compound is then brought through a flow meter into a thin film evaporator, where it is joined by a gas stream of dry helium. The tin compound/helium gas stream then joins a nitrogen/oxygen gas stream. This four-component gas stream then passes through another evaporator, located about 200 feet from the point where the nitrogen/oxygen mixture was added, where it joins a water and trifluoroacetic acid ("TFA") gas stream, heated to about 400° F. This six-component stream passes through a static mixer, which thoroughly mixes all of the elements.

As a result of his interpretation of claim 1 of the '096 patent, Dr. Gordon explained that the LOF process did not infringe. First, the components of the coating solution are not simultaneously introduced into the carrier gas stream. Second, the LOF process does not have a wet carrier gas into which the components of a coating solution are introduced. Water is not added in the LOF system until the TFA is added. Moreover, the amount of water added with the TFA is substantially more than 100% relative humidity at 18° C. Dr. Soubeyrand provided significantly greater detail on the LOF process, including LOF's development of that process.

DISCUSSION

I. The Impact of Markman on Trial of Patent Cases

The Federal Circuit's decision in Markman will undoubtedly change the face of patent litigation as it clearly did in this case. A number of procedural issues flow from this decision.

In the normal course of litigation, courts should endeavor to manage cases in such a way that all claims and issues in a civil action are presented for resolution in one trial. Lis v. Robert Packer Hosp., 579 F.2d 819 (3d Cir.), cert. denied, 439 U.S. 955, 99 S.Ct. 354, 58 L.Ed.2d 346 (1978). Moreover, in accordance with Congress's mandate in the Civil Justice Reform Act, this court schedules cases for trial within one year from the filing date of the complaint and generally schedules trials to last no more than two weeks.

The court's experience with this relatively rigid time frame shows that it promotes more efficient resolution of civil matters. For example, faced with firm and certain trial dates parties are often encouraged to settle rather than proceed to trial. Furthermore, and perhaps more importantly, this time frame assures parties that by a date certain they will have a final resolution of issues facing and affecting them as a result of litigation. Fixed trial time often leads parties to try only the core issues of their case, without wasting the court's and jury's time with issues upon which the party may have little chance of success.

In spite of this general practice, separate trials or staged resolution of issues may be worthwhile in certain circumstances. See Johns Hopkins Univ. v. CellPro, 160 F.R.D. 30, 33 (D.Del.1995). As this court noted, Federal Rule of Civil Procedure 42 provides that a court may order separate trials on claims or issues in an action when it is in the interests of efficient judicial administration. Id. In CellPro, this court addressed claims for willful infringement and the resulting mischief which has become a consistent problem in patent cases. Id. at 33-35. This court refused to order separate trials of liability, damages, and willful infringement because CellPro failed to satisfy the requirements of Rule 42. Id. at 35-37.

In Markman, the Federal Circuit stated, in no uncertain terms, that it would have the final say as to the meaning of words in a claim of a patent, according no deference to decisions by the various United States District Court Judges. 52 F.3d at 979. That is, in spite of a trial judge's ruling on the meaning of disputed words in a claim, should a three-judge panel of the Federal Circuit disagree, the entire case could be remanded for retrial on different claims.

As evidenced by this case and others pending in this court, in view of Markman, parties will now routinely move for the early resolution of the claim construction issue either under Federal Rule of Civil Procedure 56 or 12(b)(6). In a bench trial, the court can delay resolution of the claim construction issue until all of the evidence has been presented. However, in a jury trial, delaying resolution of this issue until trial may raise serious practical problems of how to adequately and fairly rule on these often difficult and vitally important issues at the close of the evidence while a jury waits. Moreover, in jury cases, it may be more efficient to put the case in a posture to have the Federal Circuit review the claim interpretation issue before trying the case to a jury, in order to avoid wasting two weeks or more of a citizen's time because the court erroneously instructed the jury on the meaning of a claim term. However, this approach could add significant time and expense to the ultimate resolution of the litigation.

Sensing the importance of claim construction on the outcome of patent cases, parties will likely seek ways to promptly bring the issue before the Federal Circuit. In cases where parties dispute facts surrounding the accused product, once the court resolves the meaning of claim terms the parties will seek an immediate interlocutory appeal to avoid the possibility of dual trials should the Federal Circuit reverse the trial court's claim construction on an appeal from a jury verdict. Where the parties do not dispute the facts surrounding the accused product, they may stipulate to the remaining issues on infringement and appeal to the Federal Circuit from a final order on that issue. In either situation, a case could be appealed to the Federal Circuit only months after the complaint is filed. This would be an unusual procedure for a district court to consistently face in patent cases.

Finally, it remains to be seen what the impact of the court's new role as arbiter of the meaning of disputed words in the claims of patents will have on a party's right to a jury trial on validity issues. For example, it is unclear how the issue of indefiniteness will be presented to a jury where this court instructs the jury on the meaning of vague words that commonly appear in patent claims, such as "substantially."

II. Claims and the Meaning of Technical Terms

Claims are the numbered paragraphs appearing at the end of the patent specification which contain words used by the inventor to describe his or her invention. Those words set the outer boundaries of the patentee's exclusive right to prevent others from making, using, or selling the invention described by those claims during the life of the patent. In re Vamco Mach. and Tool, Inc., 752 F.2d 1564, 1577 n. 5 (Fed.Cir.1985).

The patent laws focus the requirements of the information that must appear in each patent and its claims on those skilled in the art. 35 U.S.C. § 112. For example, the first paragraph of 35 U.S.C. § 112 tells us that a patent must teach one of ordinary skill in the art how to make and use the invention. The second paragraph of § 112 tells us that a claim must particularly point out and distinctly claim the subject matter which the applicants regards as his invention, the so called definiteness requirement. From that requirement, we know that those skilled in the art should be able to read a patent and its claim and understand what can be done without the patent owner's permission. Beachcombers v. Wildewood Creative Prods., Inc., 31 F.3d 1154, 1158 (Fed.Cir.1994).

Having obtained a patent in an ex parte proceeding between a technically trained patent lawyer or agent and a technically trained patent examiner employed by the United States Patent and Trademark Office ("PTO"), the owner of a patent may seek to preclude others from making or using what is described in the claims of that patent. 35 U.S.C. § 271. In fact, a person may be liable for infringement whether or not that person intended to infringe a patent. See Intel Corp. v. United States ITC, 946 F.2d 821, 832 (Fed.Cir.1991).

To enforce its patent, the owner will then file a complaint in federal district court. In that complaint, the patentee will set forth the basis for jurisdiction and for his claim for infringement consistent with the Federal Rules of Civil Procedure. Not surprisingly this complaint will also almost always contain a claim for willful infringement, treble damages, compensatory damages and seek a trial by jury. In response, the accused infringer will deny liability and counterclaim for a declaratory judgment that the claims of the patent are invalid. Also, the accused infringer will almost always contend that the patent is unenforceable. The patent owner will then file an answer to the counterclaim.

To prove infringement of a claim in a patent, a patent owner must persuade a jury by a preponderance of the evidence only that the claim describes the accused product. However, before the jury can compare the claim to the accused product, case law instructs us that the claim, directed to people of ordinary skill in the art, must properly be interpreted. Read Corp. v. Portec, Inc., 970 F.2d 816, 821 (Fed.Cir.1992).

If the meaning of words used in a claim to describe an invention to the audience of people skilled in the art are not in dispute then claim construction is a fairly simple process. In practice, however, parties rarely agree as to the meaning of the claim terms. In most cases, the parties will look first to the accused product and describe it. Then, the parties will look at the words used in the claim to see where an interpretation of a word or group of words does or does not describe the accused product or process. As to these words, the patent owner will propose a meaning that precisely describes the accused product or process. The accused infringer will do just the opposite.

It is this process, central to many patent cases, that generates the disputes over the meaning of claim terms which this court must now resolve as a matter of law in every patent infringement action. Not surprisingly, resolution of the claim interpretation issue often resolves the infringement issue, as it will in this case.

III. Canons of Claim Construction

The construction of words appearing in a patent claim is merely a way of elaborating on "the normally terse language" of the claims, in order to understand and explain, but not to change the scope of the claims. Scripps Clinic & Research Found. v. Genentech, Inc., 927 F.2d 1565, 1580 (Fed. Cir.1991). A number of factors may be considered in construing or interpreting words that appear in claim, including other words in the claim, other claims in the patent, the specification, the prosecution history, and expert testimony and other evidence outside the patent. SmithKline Diagnostics, Inc. v. Helena Labs. Corp., 859 F.2d 878, 882 (Fed. Cir.1988).

An inventor is free to define words used in a claim. However, where the inventor chooses to redefine words he or she must clearly indicate the new definition in the patent or prosecution history. Beachcombers, 31 F.3d at 1158. That is, where the inventor fails to clearly indicate he or she has adopted an uncommon or new definition to a word used in a claim, the common meaning to one of ordinary skill in the art controls.

We also know that examples set forth in the specification of a patent do not necessarily limit the scope of the claim. Transmatic, Inc. v. Gulton Indus., Inc., 53 F.3d 1270, 1277 (Fed.Cir.1995). However, an inventor may limit the permissible interpretation of a word in a claim through arguments or amendments made during the course of the prosecution of the patent. Southwall Tech., Inc. v. Cardinal IG Co., 54 F.3d 1570, 1576 (Fed.Cir.1995). Presumably, in such a case, upon reviewing the file history, one skilled in the art would understand that the inventor had adopted a narrower construction of a term.

Expert testimony and other evidence outside the confines of the patent and prosecution history may also be used to understand the meaning of words used in a claim. SmithKline, 859 F.2d at 882. Such evidence may clarify the meaning of a term, which is unfamiliar to a judge trained in the law, to those of ordinary skill in the art. However, this extrinsic evidence may not be used to vary or contradict words used in a claim. Markman, 52 F.3d at 981.

IV. Construction of Claim 1 of the '096 Patent

A. Use of EP '403 to Construe the Claims

LOF contends the meaning of "coating solution" is further illuminated by the fact that if the court construes this term to cover a gas-in-gas mixture it would be anticipated by EP '403. LOF's expert testified that he understood the EP '403 patent to depict and describe each and everything described in claim 1 of the '096 patent.

Prior decisions from the Federal Circuit indicate that where a term is susceptible to two reasonable interpretations, the court can look to prior art. See Rawlplug Co. v. Illinois Tool Works, Inc., 11 F.3d 1036, 1041 (Fed.Cir.1993); Texas Instruments, Inc. v. United States ITC, 871 F.2d 1054, 1065 (Fed.Cir.1989). The court should then adopt that meaning which would preserve the validity of the claim. Texas Instruments, 871 F.2d at 1065.

Atochem contends Markman and its progeny do not mention prior art as a claim construction tool and thus a court is no longer permitted to resort to the prior art to construe a claim except as a dictionary or to the extent arguments were made to distinguish prior art in the prosecution history. See Markman, 52 F.3d at 979-81; Transmatic, 53 F.3d at 1277. More importantly, Atochem contends looking to EP '403 in this case would interfere with its right to a jury trial on anticipation.

A claim is invalid for anticipation where a single reference in the prior art discloses all of the elements and limitations of the claim. Scripps Clinic, 927 F.2d at 1576. That is, the words of the claim describe what is disclosed in the prior document. However, "there must be no difference between the claimed invention and the reference disclosure, as viewed by a person of ordinary skill in the art." Id. In order to understand how one skilled in the art would view a particular reference, the court may require expert testimony. However, all aspects of the claimed invention must be disclosed in the reference and expert testimony may not be used to fill in the gaps. Id. Clearly, what a reference discloses to one skilled in the art is a classic question of fact. Id.

Here the parties dispute the teaching of EP '403. The effect of the patent on each and every limitation in claim 1 of the '096 patent is not clear on its face. In fact, Dr. Spear testified that EP '403 discloses the use of dry air in the carrier gas stream. Dr. Gordon testified that EP '403 takes in ambient air which would necessarily contain moisture in the relevant ranges. This dispute appears to raise a genuine issue of material fact as to the extent of the disclosure of EP '403. Consequently, to preserve Atochem's right to a jury trial on the issue of anticipation, this court will not look to the EP '403 patent in construing the words in claim 1 of the '096 patent.

B. The Proper Interpretation of the Words "Coating Solution"

Atochem contends a "coating solution" includes mixtures of liquids, solids or gases. LOF contends a "coating solution" is limited to a mixture of liquids.

It appears from the testimony offered by Dr. Spear and from the references submitted with the briefing and offered at the hearing by Atochem that the word "solution" is most accurately defined as "[a] uniformly dispersed mixture at the molecular or ionic level, of one or more substances (the solute) in one or more other substances (the solvent)." See D.I. 63, Ex. 4 (Hawley's Condensed Chemical Dictionary). It may well be that the most common solutions are mixtures of liquids. However, this does not mean that the inventor intended to limit the definition of the word "solution" to mixtures of liquids or that one skilled in the art would understand that the term was limited to a mixture of liquids.

The specification does not clearly indicate the inventor intended to limit "coating solution" to liquid solutions. On the one hand, the Abstract reads:

A chemical vapor deposition method for forming fluorine-doped tin oxide coatings uses a liquid coating composition which includes an organic fluorine dopant and an organotin compound.... A preferred liquid coating composition is monobutyltin trichloride and trifluoroacetic acid.

The Description of the Invention also includes the following passage: "The compositions of this invention contain a liquid or low melting organotin compound, a fluorine dopant which is either soluble in the organotin compound or capable of being solubilized, and an optional fluoride solubilizer." These passages might suggest the inventor intended only liquid coating solutions.

However, in the Description of the Invention, the inventor indicates: "Both the organotin halides and the fluoride compounds of this invention are ordinarily both liquids and miscible with one another." (emphasis added). This section also states: "The organotin compounds of this invention are normally liquid or low melting solid compounds." The specification, rather than modify the definition of "solution," tends to confirm the fact that many solutions are more commonly liquid mixtures. However, this language cannot reasonably be interpreted to suggest the inventor "clearly set forth" a different definition from the conventional one. See, 31 F.3d at 1158.

Other words appearing in claim 1 do not suggest "coating solution" is limited to a mixture of liquids. The use of the words "soluble in or miscible with said organotin chloride" in claim 1 do not limit "coating solution" to a mixture of liquids as LOF advocates. As Dr. Spear testified, gases can be soluble and miscible with one another. Apparently, by definition, all gases are soluble in or miscible with each other. It would appear that rather than using those words in claim 1 to limit the definition of "coating solution," the inventor used those words to limit the class of liquids that could be employed as the coating solution.

Finally, the prosecution history of the patent, submitted with the briefing but not mentioned at the hearing, provides no assistance in resolving this issue. Consequently, the court finds a "coating solution" in claim 1 of the '096 patent means any uniformly dispersed mixture, at the molecular or ionic level, of one or more substances in one or more other substances, where the substances are further defined by the claim.

C. The Proper Interpretation of the Words "Sufficient Water Vapor Such That the Relative Humidity of the Gas Stream at 18° C. is about 6% to about 100%"

In addition to their disagreement over the meaning of "coating solution," the parties dispute the meaning of the words "sufficient water vapor such that the relative humidity of the gas stream at 18° C. is about 6% to about 100%" that appear in claim 1 of the '096 patent. In particular, Atochem contends those words only set a lower boundary, focusing on the word "sufficient." LOF contends those words set both a lower and upper limit for the permissible water content of the gas stream.

Claim 1 states "the gas stream contains sufficient water vapor such that the relative humidity of the gas stream at 18° C. is about 6% to about 100%." These words clearly show that the inventor contemplated a range of moisture content in his CVD process.

In addition, Dr. Gordon's testimony regarding the more detailed description of the invention in the specification of the '096 patent confirms that these words clearly express a limited range of moisture values. As Dr. Gordon indicated the specification describes the invention as bringing ambient air into the system, splitting the air into a dry gas stream and a wet gas stream, and then bringing them together again as the carrier gas stream in such a way that the operator could regulate the moisture content. However, the process described in the specification limits the range of permissible values for that moisture content from 0% to 100% relative humidity at 18° C., which is apparently a value representing the usual ambient air temperature. With that apparatus, the inventor then experimented with different moisture contents as reflected by the tables in the specification of the '096 patent. According to Dr. Gordon, reading the language used in claim 1 in the light of the specification, one skilled in the art would interpret the words at issue to express an upper and lower limit on the permissible moisture content of the system.

Moreover, Atochem's position does not appear reasonable in light of the language chosen by the inventor. Atochem and Dr. Spear suggest that the words at issue here describe any system that can be brought to 18° C. and have a relative humidity of 6% to 100%. In particular, for systems that operate at much higher temperatures, so long as the system contains enough water that when the system is cooled to 18° C. there is between 6% and 100% relative humidity, the claim limitation is satisfied. Any excess moisture would simply condense out and form a puddle. In effect, Atochem contends the words in the claim could be rewritten to read: "sufficient water vapor such that the relative humidity of the gas stream at 18° C. is at least 6%."

However, the inventor chose to express his invention, in part, by using the words "the gas stream contains sufficient water vapor such that the relative humidity of the gas stream at 18° C. is about 6% to about 100%." Had the claim read "sufficient water vapor such that the relative humidity of the gas stream at 18° C. is about 6% to about 80%," Atochem's argument clearly would fail. With these words, one could not have questioned that the claim set a lower and upper limit.

The fact that the inventor used the words "to about 100%" fortuitously gives Atochem the ability to argue the words described higher amounts of water. As explained during the trial, air at 100° C., for example, can hold significantly more water vapor than air at 18° C. Therefore, if air at 100° C. and 80% relative humidity were cooled to 18° C. one would get air at 100% relative humidity and a puddle of water. Atochem contends the words of the claim describe this situation. However, the inventor's expression of his invention using the words "to about 100%" appears to clearly suggest he intended an upper limit on the moisture content and in fact, Dr. Gordon testified that one skilled in the art would interpret the claim that way.

The specification confirms this conclusion. First, Dr. Spear testified that each time 100% humidity was expressed as the moisture content in the tables, the system had 2 mole percent of water, which corresponds to 100% relative humidity and 18° C., and no more. Moreover, Dr. Gordon explained the inventor's description of his invention and how moisture was brought into the gas stream. This description of the invention and how the gas stream was humidified confirms that the inventor contemplated an upper and lower limit on the moisture content of the gas stream. Consequently, the court finds "the gas stream contains sufficient water vapor such that the relative humidity of the gas stream at 18° C. is about 6% to about 100%" sets both a lower and upper limit for the moisture content of the gas stream, where the upper limit equals a relative humidity of about 100% at 18° C.

D. The Proper Interpretation of the Words "A Coating Solution is Introduced Into a Carrier Gas Stream"

The final dispute over the meaning of words in claim 1 of the `096 patent concerns the words "a coating solution is introduced into a carrier gas stream." Atochem contends the phrase permits separate introduction of the components or constituents of the "coating solution" so long as the components eventually end up combined in the gas stream. LOF contends the words require that the components of the "coating solution" be mixed first and then introduced into the carrier gas.

The words of the claim describe the invention in part as requiring that "a coating solution" be "introduced into a carrier gas stream." The claim does not further define how the coating solution is introduced. The claim only requires that the coating solution be introduced in some way.

In describing Method A in the Description of the Invention, the inventor explains that "[i]f a gaseous dopant or a separate liquid component is desired it may be fed into the air mixture at a point before the first heated oil bath." The description of Method A indicates that this point before the first heated oil bath is also before the organotin is introduced. The specification goes on to state: "When the novel compositions are deposited by chemical vapor deposition ("CVD") Method A or B, the organotin, dopant and solubilizer, if used, may be mixed at room temperature or added separately." (emphasis added). Thus, the specification clearly indicates that the inventor contemplated the possibility of bringing the components of the coating solution into the gas stream at different points along the process so long as they are mixed at some point. Consequently, the court finds "a coating solution is introduced into the carrier gas stream" permits the components of the "coating solution" to be introduced separately into the carrier gas stream.

CONCLUSION

Having construed the disputed words in the claims in the manner set forth above, the court will enter an Order establishing the meaning of these terms.