Praxair v. ATMI (Fed. Cir. 2008)
Praxair’s patents cover pressurized storage containers for potentially hazardous gasses. The patents teach safety mechanisms to help prevent accidental rapid discharge. Of the three patents asserted against ATMI, the Delaware district court found two unenforceable due to inequitable conduct during prosecution. The asserted claims of the third patent were held invalid as indefinite under 35 U.S.C. §112¶2. On appeal, the Federal Circuit reversed-in-part.
Inequitable Conduct: Each patent applicant has a duty to act with “candor, good faith, and honesty” when prosecuting patent applications. Material breach of this duty can result in any resulting patents being held unenforceable due to inequitable conduct. Proof of inequitable conduct has been divided into two parts – requiring “clear and convincing evidence that the applicant (1) made an affirmative misrepresentation of material fact, failed to disclose material information, or submitted false material information, and (2) intended to deceive the [PTO].” Quoting Cargill v. Canbra Foods, Ltd., 476 F.3d 1359 (Fed. Cir. 2007).
Direct evidence is rarely available to prove intent to deceive the PTO when the patent applicant is accused of withholding material information. Rather, the intent element is typically inferred by the proof that the material withheld was highly material; that the applicant knew (or should have known) of the materiality; and that the applicant has no credible excuse for withholding information.
Highly Material Based on Statements in Prosecution: Looking at Praxair’s ‘115 patent, the court did not see the withheld prior art as highly material on its own. However, the reference became highly material based on Praxair’s statements during prosecution. Specifically, the reference became highly material when Praxair told the PTO that the prior art does not teach the elements found in the withheld art. Testimony of the inventors proved that they knew that those elements were, in fact, in the prior art.
Because those same prosecution statements were not submitted during prosecution of Praxair’s ‘609 patent, the Federal Circuit held that the withheld references could not be considered highly materiality.
|
“At the time the four statements discussed were made during the prosecution of the ‘115 patent, there had already been a notice of allowability indicating that all claims of the ‘609 patent would be issued. ATMI has not established, or even asserted, that the statements in the prosecution of the ‘115 patent somehow infected the prosecution of the ‘609 patent. Absent the four statements, the district court did not make any finding of intent with respect to the withholding of RFO art during the prosecution of the ‘609 patent.” |
Thus, the Federal Circuit affirmed the unenforceability holding as to the ‘115 patent, but reversed as to the ‘609 patent. On remand, the lower court will reconsider infringement of the ‘609 patent based on a claim construction error.
Indefinite Claims:
|
“Indefiniteness is a matter of claim construction, and the same principles that generally govern claim construction are applicable to determining whether allegedly indefinite claim language is subject to construction. Datamize, LLC v. Plumtree Software, Inc., 417 F.3d 1342, 1348 (Fed. Cir. 2005). Indefiniteness, like claim construction, is a question of law, and we review a district court’s entry of summary judgment on the issue of indefiniteness de novo.” |
Claims are generally only found indefinite if they are “insolubly ambiguous, and no narrowing construction can properly be adopted.” Here, the district court found the claim term “port body” of the third patent to be indefinite because it was “not described, labeled, or coherently discussed in the patent.” On appeal, however, the Federal Circuit reversed that finding – holding instead that the patent’s specification was adequate although not eloquent.
|
“Although the discussion of the port body in the ‘895 patent’s specification may not be a model of clarity, the specification adequately explains that the port body is a housing that sealingly engages the outlet of the cylinder and defines the fluid discharge path.” |
Extrinsic evidence: As with claim construction, the use of extrinsic evidence in determining definiteness is only secondary if applicable at all. In this case, the patentee’s expert was unable to point to any specific portion of the structure that could be considered the “port body.” On appeal, the Federal Circuit rejected that factual element as lacking impact on the legal question of indefiniteness.
|
“Even if we were to agree that Dr. Fronczak was unable to reach a single consistent construction of the port body, such extrinsic evidence would not prove the ‘895 patent invalid, since indefiniteness is a legal rather than a factual question.” |
Indefiniteness holding reversed.
Dissenting in part, Judge Lourie saw no inequitable conduct with either patent. In particular, Judge Lourie noted that the evidence of intent to deceive the PTO was quite weak.
|
“While a smoking gun may not be needed to show an intent to deceive, more is needed than materiality of a reference. The district court did not find anything further here. In addition, the court did not engage in any balancing of materiality and intent as is required by our precedent.” |
Max,
As I recall, that 1992 experience was my last in dealing with European oppositions, at least directly. (I remember getting involved in the start of another European opposition, but that was as far as I got.) Haven’t had another opportunity since that time. Sad, I enjoyed Munich in 1992, would certainly like to go back there.
So, EG, did that happy experience, of terminating somebody else’s European patent, affect the way you prosecute your own cases through the EPO? Any concerns that somebody might oppose one of yours (after you’ve spent a small fortune translating it all over Europe)? Or would you perhaps relish getting an opposition, as a chance to further hone your already finely polished European B publication?
Max,
Interesting that you say it was harder to kill an issued EPO application 16 years ago. We didn’t have too much trouble killing the one that we did, but I also had a pretty good European associate at my side, too.
EG, back in 1992 it was hard to convince the EPO to kill one of its own new-born babies. So, well done, you and your EPO representative. These days, EPO Examiners and (particularly) Appeal Board Judges have fewer scruples. They better understand the mischief that a bad patent perpetrates on the innocent public. The Opponent only needs to get over the preponderance. I expect to see more EPO oppositions in the future, esp.from US companies.
I find myself agreeing with the Dissenting Opinion in this case.
“The only thing certain about oral proceedings at the EPO….is that there will be a surprise, somewhere along the line.”
Max,
So true. I attended an oral hearing in an opposition at the EPO in 1992. I was warned by our European associate to expect the unexpected. The other side provided upwards of a half a dozen claim amendments, only the last of which was accepted. And we still won.
The only thing certain about oral proceedings at the EPO….is that there will be a surprise, somewhere along the line.
I’m hopeful that we’ll have an after lunch decision like that on the day so I can go out and celebrate with a few Hefeweizens! It looks like a cut and dry case but you never know…
“The key word, first time caller, IMO is ‘uniform.'”
Not sure what you mean, ASE. As I said, I think the “uniform” limitation renders the claim indefinite since there is nothing in the spec to define the degree of uniformity required. However, in my sample claim, the question of uniformity is independent of the question of how light intensity is defined. Isn’t it? I’m not looking for engineering details in claims, but when the patentee puts them in there then they need to have real meaning. I simply want to know whether my product infringes the claim or not. If I can’t possibly tell, then it’s indefinite.
Well, good luck Frodo. It’s tough to win on that one, but oh so sweet when one does. Particularly to be cherished is to imagine afterwards, how the attorneys on the other side are going to explain it to the client that has just seen all patent rights in Europe disappear in a puff of smoke. How does it go again, down at the EPO, upon resumption, after lunch: “The oral proceedings are resumed. The Board of Appeal has come to a Decision. The patent is revoked. The written Decision will follow. The oral proceedings are closed”
Max, I’m with you on the last comment. I’ve got oral proceedings in the next couple of months, and we are hoping to take them out under Art 83 and 100(b). Fingers crossed for that one….
SCOTUS, another thought. One of the most potent killer attacks on an issued European patent is Art 83 and 100(b) “insufficiency of description” because, if you can make it good, it takes out the whole patent. Not like prior art attacks where patentee amends back to something tighter and stronger. Ask an old guy in Europe to tell you how adept everybody over here has become, over the last 30 years, in adapting to the lack of Art 84 as a ground of revocation, by squeezing their “indefiniteness” attacks in under Art 83. You might be surprised, how potent “insufficiency” can be as a weapon of total patent annihilation. Another reason why deliberate ambiguity might not be a good idea, perhaps (but I know nothing of this area) even exposing US patent attorneys to malpractice suits brought by angry clients deprived, purely and simply by deliberately ambiguous drafting, of all patent rights throughout 600 million inhabitant EPO-land. “Ambiguity” used to be a ground of revocation, in the UK, till 1978. It lives on, all over Europe, in Art 83 EPC. We don’t allow attacks on issued patents merely because the patent attorney (usually using English as a foreign language) couldn’t quite write a claim in English that was 100% “clear”. But carefully crafted ambiguity is another matter, and something that in Europe can come back and bite you. Some people say “In the end, sooner or later, you get served out in the way you serve others out”. Civil law jurisdictions have more freedom to get to that result.
SCOTUS, you haven’t convinced me yet. Take the classic ambiguity “for”. Does it mean “suitable for” or “intended for”. When a Patent Office always takes the broadest meaning (as the EPO does, always, and sometimes to ridiculous extremes) then it’s “suitable for” and for the EPO a tennis ball would deprive of novelty a claim to “A ball for playing football”. “Applicant, you don’t like it.” says the EPO. “Tough. Amend out of it”. When it comes to infringement, the “for” claim will likely be construed in Duesseldorf as “intended for” but, in separate revocation proceedings in Munich, as “suitable for”. Result: not infringed and, anyway, invalid. Nil return on your deliberate ambiguity. Your US litigators might like deliberate ambiguity but my advice would be, don’t bank on it giving you an advantage, over here. It doesn’t have to be like it is, you know, if one has the will to change for the better. I’m not saying that ROW drafters are perfect, or saints, just that drafters, like traders, are rational and will do what gains them advantage, in their biggest single market, the USA. But the regulaters (Courts, PTO)can fight back. If they can be bothered to try.
Live by the Legal Fees, die by the Legal Fees
link to nytimes.com
I am not aware of the “really effed up” legal standard or the nuances of its application. I am aware however, that 112(2) relies on whether the scope of the claim, when read in light of the specification, would have been understandable to one of ordinary skill in the art as distinguished from, say, one with a “really effed up” understanding of electronic technology such as Mooney.
The key word, first time caller, IMO is “uniform.” Also, word to the wise, don’t look for engineering satisfaction from patent claims. The claims don’t have to provide specific engineering details unless they are necessary to distinguish over the art.
Max, there are at least two fundamental problems with Europe’s lack of indefiniteness as a defense. First, I know of many patent applicants that desire to draft ambiguity into their patent claims. They do this intentionally, knowing that it has good potential to lean in their favor during enforcement or licensing. U.S. applicants are not alone in injecting ambiguity though, as I have seem many cases originating from Europe and Japan with equally or even more ambiguity, written by “ROW” practitioners as you like to say. Second, the examiner should not have the last say of one’s patent rights, not just because some rarely get it right, but also because of the ones that usually do get it right, they are no less human than you or I and are prone to make mistakes and overlook an ambiguity from time to time. And notwithstanding the instances where an indefiniteness claim may be frivolous, I still think the better way is to allow the accused infringer to have recourse in the first place. Simply ignoring that there is an ambiguity ignores the intentions of applicants and the realities of examination.
fragile gels case should have turned 112 2nd as we know it on its head.
MM,
Not true. See Haliburton Energy Services v M-I (Fed. Cir. 2008)(“fragile gels” indefinite under 35 USC 112, 2nd paragraph)
“Even presuming that the intensity is measured at the surface of the board, and not a foot away (even that is not clear from the claim language), do I apply a sliding window that is exactly one centimeter by one centimeter to determine the intensity at each of several spots along the array? Or do I calculate an average intensity over an area that includes only the area occupied by the diodes, and then normalize to a per-sq.-cm. value? How do I handle irregularly shaped arrays of spaced-apart diodes?”
And in most of these indefiniteness cases the CAFC seems to say that if these questions can be “reasonably” answered by pouring through hundreds or thousands of pages of prosecution history and laboriously tweezing apart all of the “implicit” clues in the specification (including any cited art) then the claim is not indefinite.
Which is really effed up, if you ask me. The standard for finding a claim “definite” is so low as to be nearly non-existent. You almost have to recite a “pretend” word to get invalidated on indefiniteness grounds.
This discussion has certainly clarified some fascinating metaphysical aspects of maybe the most abstruse doctrine in patent law … first time caller, there was a recent case on test methods and definiteness, for a patent related to impedance matching … I think the patentee lost the infringement verdict exactly because the court couldn’t identify the “correct” method of testing for presence or absence of impedance-matching function in a given component, based on the claim language.
“is that claim indefinite in a lawsuit which concerns the 99 but not the 1?”
indefiniteness has nothing to do with the accused products. it’s either indefinite as a matter of law, because it doesn’t define ascertainable boundaries for infringement, or it’s not. Of course, like other invalidity issues, the fact that one district court fails to find it indefinite does not preclude others from doing so later. once it’s finally adjudicated as indefinite, it’s done.
If there are 100 potential infringing devices and 99 of them clearly infringe but 1 of them may or may not infringe because of claim language used, is that claim indefinite in a lawsuit which concerns the 99 but not the 1?
“The real indefiniteness problems are those where the claim recites some vague functional language, e.g., “durable,” “stable”, “flexible”, and the specification fails to define the term. Instead, the specification provides only examples where some test is applied to determine the durability/stability/flexibility/etc but a necessary minimum is not provided.”
That covers some of the indefiniteness problems that slip through. But, consider this (slightly edited) claim from US 6,371,637:
1. A flexible, low profile lighting system, comprising:
a flexible printed circuit board substrate adapted to support and electrically interconnect surface mount electronic components;
a plurality of surface mount light emitting diodes; and
wherein the plurality of light emitting diodes are surface mounted on the flexible printed circuit board substrate so as to define a conformably bendable lighting array, the array outputting a uniform light intensity of at least 2000 millicandles per square centimeter.
This claim does include one of those vague words, i.e., “uniform,” which is completely undefined in the specification. However, even if that word were omitted, the claim is still indefinite for a more subtle reason. The problem is that it is impossible to tell (and the specification does not clearly explain) how to calculate the light intensity metric to compare to the standard of 2000 millicandles per sq. cm. One illustrated embodiment has a single row of diodes that are each considerably less than a centimeter wide. Even presuming that the intensity is measured at the surface of the board, and not a foot away (even that is not clear from the claim language), do I apply a sliding window that is exactly one centimeter by one centimeter to determine the intensity at each of several spots along the array? Or do I calculate an average intensity over an area that includes only the area occupied by the diodes, and then normalize to a per-sq.-cm. value? How do I handle irregularly shaped arrays of spaced-apart diodes?
Although there are some configurations that probably infringe the above claim, it’s pretty easy to come up with a number of configurations for which it is impossible to tell whether it infringes the claim or not, despite the seemingly precise language of the claim. Since I can’t tell whether certain products infringe or not, the claim is indefinite, even if it clearly covers one or more definite embodiments. It’s not hard to see why this kind of problem slips by an examiner.
“The CAFC does go overboard in designating irrefutably factual issues as ‘a matter of law’ and therefore reviewable de novo. It’s a necessary evil in a system where the trial courts more often than not just don’t have a clue.”
Perhaps, but indefiniteness is not an “irrefutably factual issue.” Indefiniteness is inextricably bound up in claim construction. If claim construction is a matter of law (and the Supreme Court says it is) then indefiniteness is unquestionably also a matter of law. The CAFC is not arbitrarily designating it a matter of law so that they can review it de novo.
MD “The world gets progressively more transparent. We all can learn from each other. For “durable-type indefiniteness” the EPO is a particularly unforgiving environment. Game over.”
In the biotech/chem arts, the USPTO has been aware of the problem for years. As I alluded to upthread, it’s a problem with certain Examiners being less astute and rigorous than others. I would be surprised but not stunned to see the game played successfully by a biotech applicant in 2008. Of course, most of the CAFC opinions relate to patents that were prosecuted a few years back, at least.
The CAFC does go overboard in designating irrefutably factual issues as “a matter of law” and therefore reviewable de novo. It’s a necessary evil in a system where the trial courts more often than not just don’t have a clue.
Even with Moore and Prost, I’d rather take my chances with a CAFC panel than with most of these district courts. Not that district judges aren’t intelligent and excellent jurists, just that they are in over their heads in many complex IP punch-ups. I don’t know how they do as good as they do, which is not all that good.
The (costly) alternative to over-zealous de novo review is an IP courts system modeled on the bankruptcy system, for instance, with technically-trained judges.
Anonymous, agreed. But it’s not immutable and written in stone, as night follows day, is it, that the court has to waste its precious time with frivolous defences? Do you think (as I do) that England is a jurisdiction where, actually, parties in patent actions, in their own vital self-interest, are careful NOT to run frivolous defences?
Malcolm, a question. Do you see any sign of Examiners in the USPTO, having newly observed how EPO examiners deal harshly with “durable” type indefiniteness, then borrow those techniques, in their examination process at the USPTO. The world gets progressively more transparent. We all can learn from each other. For “durable-type indefiniteness” the EPO is a particularly unforgiving environment. Game over.
Defendants raise frivolous defenses all of the time, but that’s just the nature of defenses and I don’t think it’s unique to the U.S.
Fair enough, that there anonymous, but does it ever happen, anywhere, any time, that a Defendant raises a defence that doesn’t quite reach the level you assess as “credible”. Doesn’t any well-heeled Defendant caught infringing (and I have in mind here the deep pocket corporate professional infringer of patents such as those that might be held by JAOI and his ilk) “clutch at straws” or include “everything but the kitchen sink” in multitudinous defences, of various degrees of credibility, aimed at exhausting as early as possible the patent owner’s fighting fund, before we get to the serious question: infringed Y/N?
I didn’t read this case, but “port body” seems like a term that should be construeable after taking into account the interpretation of one of ordinary skill.
The real indefiniteness problems are those where the claim recites some vague functional language, e.g., “durable,” “stable”, “flexible”, and the specification fails to define the term. Instead, the specification provides only examples where some test is applied to determine the durability/stability/flexibility/etc but a necessary minimum is not provided.
What happens during prosecution of such cases is that the applicant invokes the functional language only where needed to avoid the art. In some instances the Examiner will tell the applicant that the functional term is not being construed as a limitation (a proper position) … yet does not require the applicant to remove the term from the claim. If the applicant is lucky or persistent, the Examiner may forget the earlier position he/she took or may be replaced by a new Examiner and voila! the claim now contains a term that can not be PROPERLY construed.
The CAFC has the last laugh in these dumbxss cases though because it sifts through all the instrinsic and extrinsic evidence and says that one construction does in fact “win” over the others. Thus, the CAFC will argue, the claim is subject to construction and is not indefinite. I think everyone understands that this is bogus, including the CAFC, but there we are.
MaxDrei writes:—–
I’m more bothered about the ever-present opportunity to defend lengthily, expensively and spuriously, against well-founded assertions of infringement, by asserting that the asserted patent is tainted by a degree of “indefiniteness”.
—–
Maybe those with more experience will disagree, but I think that it’s a very rare case where a defendant can raise a credible indefiniteness defense. I understand the complaints for some quarters that defendants are able to get further with inequitable conduct defenses than they should, but I’m not aware of any similar problems with indefiniteness.
It seems bizarre to me that indefiniteness should ever be a ground for complete invalidity of a patent, when the claim language has been examined for compliance prior to grant. Certainly, it can be a serious issue in litigation, but I’d have thought that if a claim is so indefinite as to defy any sensible construction then it cannot be infringed. In that case, why not let the patentee go back to the PTO for a re-issue to cure the indefiniteness, but estop him from suing again for the same alleged infringement.
You might say that this will encourage patentees to seek the vaguest claim language they can get past the PTO, but that’s part of the game anyway, is it not?
If a party is threatened with a patent that seems to have wilfully obscure claims, could they seek a court order to force the patentee to get a clarifying re-issue? I guess that opens a whole can of worms.
Or suppose the court rules the patent to be unenforceable as granted, due to indefiniteness, and leaves it for the patentee to cure it by re-issue if they wish.
Slightly off topic (ok very off topic), but does anyone know of any decent or half decent IP Auction websites/companies?
Thanks Ed. How did we ever manage before Wikipedia came along? (although I believe it was actually Patrick O’Brien that put me on to “Hear him, hear him”.
Anonymous, I’m not so much occupied by how many times, in absolute, the CAFC strikes down a patent as indefinite. I’m more bothered about the ever-present opportunity to defend lengthily, expensively and spuriously, against well-founded assertions of infringement, by asserting that the asserted patent is tainted by a degree of “indefiniteness”. I’m from out of town. Does it happen like that, in real life? As to the “worst excesses of indefinite claim drafting” is the USPTO really so powerless that the courts need to have that power? Course, in the EPO, virtually all Exrs are working in EFL (English as a foreign language) so have difficulty understanding any even half-fancy lawyer’s claim. In the EPO, you get nowhere unless you keep your claim language deadly simple. Simple enough even for a non-specialist district court judge operating in a non-English European mainland jurisdiction to understand.
Wikipedia scmhkipedia, in hobbit speak it’s here here!
For what it’s worth, Wikipedia believes it’s “hear! hear!”
link to en.wikipedia.org
I’ll add, though, that Max probably makes a fair point about the fact that indefiniteness is a legal question reviewed de novo by the CAFC. If you think that CAFC claim construction is inconsistent and panel-dependent, then any claims that are *arguably* indefinite aren’t safe until they’ve been reviewed by the CAFC. And that’s something that detracts from legal certainty.
Max, indefiniteness is a high bar to clear to invalidate a patent. The district court invalidated the patent for indefiniteness, and the CAFC reversed. The CAFC doesn’t require that a patent is 100% clear and 100% devoid of ambiguity. Rather, it requires that potential licensees and infringers be able to have some reasonable idea about what the patent covers. To go along with your statement about “maximizing legal certainty,” it’s not only patentees who prefer legal certainty. Indefiniteness doesn’t have a lot of teeth as a ground for invalidation, but it has enough to rein in the worst abuses of unclear claim drafting. If the US were to get rid of indefiniteness, it would need to concurrently improve the examination process at the PTO.
I don’t know of any egregious cases of patents being invalidated in the US for indefiniteness where they shouldn’t have been.
OTOH, I don’t know of any owners of *really* ambiguous European patents who are extracting unjustified royalties because their competitors don’t know enough to predict how litigation might play out.
Maybe it’s better in Europe and maybe it’s not. I don’t think we have the empirical data to decide. In principle, though, I don’t see anything wrong with indefiniteness as a ground for invalidation or with the way the CAFC has implemented it.
If the PTO won’t accept 112s as a grounds for re-exam (in my limited experience, that argument’s a loser), then why should a district court judge with a history/poli-sci undergrad major be permitted to determine what would or would not be understood by an engineer with 10 years experience in the art to which the patent relates?
The government’s own expert, aka the examiner, could understand the specification, so why did the court have so much trouble? I agree with those above in that indefiniteness should not normally be grounds for invalidity of an issued patent.
It’s I’m right with ya Maxie!
Point of order Mr Frodo. Is it “Here, Here” as in (I’m right with you on that” or is it “Hear, Hear” as in “It’s going in one ear, Max, and NOT going straight out the other” or in “Hear him, hear him”? Or maybe it’s both. I’ve always wondered. Anyway, thanks Frodo. On this blog, all expressions of support are refreshing, and welcome. And about this time of day, it’s morning in America, at least, on the east coast. Wonder who’s got any appetite, today of all days, for blogging the subject of “lack of clarity”.
Always wondered why it seems that the PTO is able to pass on indefiniteness in the first instance and not put the issue to bed for good. It’s not as if there’s something not in front of the PTO that stops them from being able to determine the issues, as can be in cases of invalidity, inequitable conduct, etc. Seems like the examiner, at least, was able to and did know what was claimed, whether it was enabled, etc.
Here here Max, I agree.
Just to note: since the 1949 UK Patents Act (and probably before then too, I’ll have to check) it has not been possible to revoke a patent on lack of clarity. It’s a ridiculous ground to have a patent revoked on.
Readers, since nobody else wants to go first, I will, with one of my “Europe is better” mailings.
Many readers want investment in innovation, and suppose that this is more likely if the degree of legal certainty can be lifted up. There never was yet, in the history of issued patents, any one that was 100% clear and 100% devoid of ambiguity. So, allowing indefiniteness as a ground of invalidity, and fixing it that indefiniteness is a matter of law, to be decided de novo at the CAFC, contradicts the public policy objective of maximising legal certainty. Europe in 1973 had the great advantage that it could write its pan-European patent statute on the shoulders of the patent statutes of other jurisdictions, learning from their mistakes. It decided that “lack of clarity” should NOT be a ground of revocation. Was that a good move? Is it something the US Congress might consider, next year, after reflecting on EPO caselaw over the last 30 years? There might be something of the “Law of Unintended Consequences” in all of this. Readers?
Comments are closed.