Breakdown of In re Celgard

Breakdown of In re Celgard
(review of Dennis Crouch’s june 26, 2017 article:

quoted text of the article follows the writeup

Dennis Crouch is a regular contributor to, an excellent site for learning more about patents.

While I might be a patent attorney in Raleigh, NC, Patents are federal law, and so the Supreme Court’s holdings have massive effect upon the patent world. While I do primarily prosecution work, filing applications and communicating with the USPTO, the enforceability of these patents is primarily the work of patent attorneys which characterize themselves as patent litigators.

In re Celgard is a Supreme Court case which addresses the issue of the Federal District Court’s review of holdings of the USPTO’s Patent Trials and Appeals Board (PTAB). The PTAB is an administrative review of examiners’ holdings, and so the federal courts’ review is a transfer of the subject matter assessment, as a procedural administrative manner, into a judicial challenge as to whether or not the uspto, a governmental agency, has mishandled their obligations/ delegated rights under the laws giving rise to their administrative authority to issue patents.

The problem, of course, with such a transfer is that the USPTO is an agency populated by patent examiners, experts in the subject matter of inventors’ expertise, while the courts are legal scholars, which can only be expected to apply the laws uniformly across agencies. What makes the performance of these legal standards to the PTAB holdings, though, is that deciding the validity of a patent frequently reaches questions of what is known in the art, an assessment that requires the examiner, and PTAB, and then the courts, to all be in roughly the same ability to perceive what a hypothetical person, having “ordinary skill” in the relevant art, would have known how to practice, before the filing of the application upon which a patent was granted.

Simply put, the courts typically perceive their own lack of expertise, and, consistent with deferring to the greater expertise within the USPTO, tend to make assessments of PTAB holdings in a digital fashion. Of late, apparently, the federal courts seem to be rubber-stamping PTAB holdings. This might be a rote satisfaction of their duties to review decisions and hold whether the law has been followed, but several litigants feel that PTAB holdings touch on so many details, in coming to their decisions, that the courts are inappropriate to blanketedly approve, ESPECIALLY when the courts provide ZERO analys11is and discussion of how they came to their findings, and without itemizing how each sub-assessment should be approved, in order to approve the overall holdings.
In re Celgard challenges this behavior.

While the petitioner in the case raises several issues pertaining to how the courts are carrying out these “Rule 36” judgements, I believe that two issues  are most important to the notions of whether the courts are acting inappropriately:

“Whether the Federal Circuit’s issuance of Rule 36 judgments without opinions for the disposition of appeals from the Patent Office violates 35 U.S.C. § 144’s requirement that the Federal Circuit “shall issue” its “mandate and opinion” for such appeals?”


“Whether the Federal Circuit’s pervasive practice of issuing Rule 36 judgments without opinions to affirm more than 50% of appeals from the Patent Office has exceeded the bounds of reasonableness and is inconsistent with “principles of right and justice”?

HOWEVER, and this is my criticism of the article by Crouch, there is zero discussion of these matters. The precedent of significance, that the case could provide, is a change in the way that the courts are required to weigh-in on the appropriateness of the holdings of PTAB. The controversy underlying the conflict is merely a typical conflict as to whether or not prior art was appropriate to combine, in order to form the judgement to invalidate the claims of the application in question.

Having said all of that, I suppose I will now comment on the rest of the article, even though it veers off away from its most valuable purpose.

Celgard’s argument as to the validity of its patent is that the court (in failing to review PTAB’s logic), failed to hold the PTAB to the obligation to “consider the whole invention.” By “the whole invention,” Celgard is asking the courts consider how its specification supports its claims, in order to show that their meaning cannot have been met by the combination of references which the examiner applied. The invention of Celgard essentially prevents a formation of structures on the electrodes in a battery, by providing an impermeable layer between the chemicals that form the structures. The references which the examiner applied are combinable for teachings known in the art, without the benefit of the Celgard disclosure, which allow for a person in the art to have knowledge to have made a thing that can be described by the language of the claims in Celgard. Essentially, the claims were held invalid by the examiner because the examiner was able to point out that the claim language describes a thing that existed in the art, for their literal meaning, to whatever degree they can be interpreted, by merely reading them in the abstract, or perhaps in the context of the specification of Celgard, but only to the extent that it would be consulted in order to show that Celgard adequately discloses a thing that can be described and claimed as being a thing which could be defined by the specific language of those claims.

Celgard’s claim that the “whole invention” should have been considered is predicated upon the observation that the combination of the cited references, when combined, fail to provide the advantage and/ or function of Celgard’s invention, because their combination would be porous, and therefore the opposite of a device having the impermeability of Celgard’s device. The argument hinges upon the assertion of that the resulting device, built to a combination of the references, would essentially “break” the purpose of the invention of Celgard, and therefore Celgard is NOT taught by the references.

Since the court didn’t weigh-in, I feel invited to give my assessment as to how that should have been handled. Tough luck, Celgard. Claim language is specifically for the purpose of defining the invention so broadly as to encompass ANYTHING which those sentences would describe, to the reasonable interpretation of any person having ordinary skill in the art. This is so that inventions are able to guard against nifty reconfiguration of the novel aspects of the claimed invention, to prevent non-literal subsequently-practied devices from skirting the narrow, specific, embodiments which are disclosed in the patent. The patent-rights-holder is entitled to guard against these variations that were not contemplated at the time of the invention, because the inventor is not supposed to have to conceive of every possible device which their claims describe. The inventor is only required to disclose enough to support the claims, the disclosure being enough to practice, to the extent that the claims attempt to secure protection, at least one “best” mode of the invention. A single embodiment that supports the claims.

Therefore, the claims should NOT be allowed to describe and be held out as an exclusive realm over things that the claims describe, even where such a “thing,” only exists as a part of the “know-how” of a cumulation of known practices. The applicant is only entitled to a scope of protection that describes nothing else that could have been practiced before the priority date of the application. That scope must fail to describe even things which are completely dissimilar to what is taught in the application.

As an example, consider:

I claim:

A tool having a working end with a force-application-tip, which terminates in a working vertex, characterized by a narrow and substantially linear edge, having a width, the force-application-tip abutting a first end of an elongate shaft having a predominant length, wherein the length of the elongate shaft is substantially transverse to the width of the linear edge.
The tool of claim 1, wherein the elongate shaft has a second end, the second end abutting a graspable handle, the handle constructed to stabilize the tool under application of force to the handle, such that the elongate shaft may translate force applied to the handle into force which drives the working end of the tool.
The tool of claim 2, wherein the elongate shaft has a torsional strength sufficient to translate torque applied to the handle to twist the working end, the linear edge having a left end and a right end and a center point between the left and right ends, and wherein twisting the working end means that, in operation, the linear edge applies a desired force perpendicular to the length of the linear edge, at the right end, and the linear edge applies a desired force perpendicular to the length of the linear edge, at the left end, and wherein the force at the right end and the force at the left end are applied in opposite directions to one another, with respect to a workpiece, but the same direction, with respect to a direction of rotation about the length of the shaft.
The application describes a screwdriver.

The examiner cites a chisel, and a method of causing a chisel to drive one edge into a workpiece, to make a narrow channel, by driving it while applying a twisting force.

Claims 1 and 2 describe both a screwdriver and a chisel, but, let’s say that the disclosure of the chisel fails to provide any knowledge that it could be twisted, whereas the screwdriver, as we all understand it, would almost certainly be described for a purpose to twist … screws… by applying a torque. DESPITE their clear differences, the inventor of the screwdriver is NOT entitled to protect his inventions using claims 1 and 2. It does NOT MATTER that his application teaches nothing about what a chisel is good for, nor does a chisel itself need to teach that, for example, its edge is only so thin as is necessary to engage the interior faces of the slot on the head of a screw. A chisel might actually be entirely inappropriate to be used this way, its sharpened edge being too narrow to fit a screw this way. But it doesn’t matter. Claims 1 and 2 describe such a device.

Claim 3 describes the torque-ability of the screwdriver. Because a chisel might seem to teach away from such torque-ability, as discussed in the previous paragraph, it would likely be unavailable to be considered an anticipation of claim 3. HOWEVER, the reference teaching making a channel, by twisting the chisel, while driving, provides a reason to twist a chisel, even though it does nothing to make the chisel useful for twisting in the way that the screwdriver may be twisted. The combination of the two references clearly still fail to meet “the whole invention” standard which Celgard’s argument would ask us to accept, but that doesn’t matter.

The two references are combinable for their own purpose: A reference showing how to use a chisel, and a chisel not incompatible for that chisel-ing purpose, may be combined in order to describe a cumulative knowledge that someone could use, in order to do some chiseling. That combination, presents a device suitable to teach an invention that is described by the language of claim 3. To illustrate the inappropriateness of the enforceability of such a patent, in a world where the chisel and twisting reference came first, consider that the screwdriver invention, if allowed protection on even claim 1, would be expected to be enforceable against ANY chisel, if the chisel was only first made or practiced after the time of the grant of the screw driver application.

The chisel is describable by the language of claim 1. If the screwdriver was first, then it falls within that broadest reasonable interpretation. And that is why the chisel coming first likewise anticipates the scope of claim 1.

So, good luck Celgard’s petitioner. At best, you’ll get your full explanation from the courts as to why you still don’t deserve a patent.


quoted text of the article, from
Celgard: Important Challenge to the Federal Circuit’s Pervasive No-Opinion Judgments

June 26, 2017 Dennis Crouch

By Dennis Crouch

In re Celgard (Supreme Court 2017) [2017-6-19 Celgard Cert Petition]

In what looks like a well-postured case, Celgard LLC has asked the Supreme Court to consider several questions stemming from the Patent Trial & Appeal Board (PTAB) decision that the claims of the company’s U.S. Patent No. 6,432,586 are invalid as obvious.  The patent claims a particular microporous membrane separator that adds a ceramic layer to effectively limit the growth of lithium dendrites – the major cause of lithium battery failures and overheating.  The approach created by Celgard has been widely copied (and further adapted) in the industry.

In order to avoid royalty payments and supply agreements with Celgard, a host of competitors filed at least six inter partes review (IPR) petitions against the ‘586 patent.[1] Although the bulk of the challenges have been rejected or found without merit by the PTO, the PTAB ultimately agreed with one of the several grounds presented by LG and ruled claims 1-6 and 11 obvious over a combination of two Japanese published patent applications in a parts-and-pieces analysis.[2] As it does with most appeals from the PTO, the Federal Circuit affirmed the PTO judgment without opinion. The petition explains:

The [Federal Circuit] decision was labeled as a “Notice of Entry of Judgment Without Opinion” and stated that “[t]he judgment of the court in your case was entered today pursuant to Rule 36. This Court affirmed the judgment or decision that was appealed. None of the relief sought in the appeal was granted. No opinion accompanied the judgment.”

The Federal Circuit also denied Celgard’s subsequent petition for rehearing and rehearing en banc.

With this basic setup, Celgard has petitioned the Supreme Court for writ of certiorari – asking four questions.

First, following identically from the pending case of Oil States, petitioner asks:

Whether inter partes review—an adversarial process used by the Patent and Trademark Office (“Patent Office”) to analyze the validity of existing patents—violates the Constitution by extinguishing private property rights through a non-Article III forum without a jury?

This question should at least serve as a place-holder pending outcome of Oil States.

The second question stems from my recent article on the Federal Circuit continued issuance of judgments without opinion in appeals of PTO decisions.  Petitioner asks:

Whether the Federal Circuit’s issuance of Rule 36 judgments without opinions for the disposition of appeals from the Patent Office violates 35 U.S.C. § 144’s requirement that the Federal Circuit “shall issue” its “mandate and opinion” for such appeals?

In addition, petitioner argues that the Federal Circuit’s continued and pervasive use of Rule 36 judgments without opinion is a violation of the principles of justice and should be rebuked by the Supreme Court:

Whether the Federal Circuit’s pervasive practice of issuing Rule 36 judgments without opinions to affirm more than 50% of appeals from the Patent Office has exceeded the bounds of reasonableness and is inconsistent with “principles of right and justice”?

Something that makes this a helpful case is the fact that Celgard also has a very strong argument on the substantive patent law merits (nonobviousness) and there are no serious troll or eligibility undertones.  In its final petition question, Celgard focuses on the obviousness question asking:

Whether the Patent Office’s consistent practice of failing to consider the claimed invention “as a whole” and failing to consider whether the combination of elements would lead to “anticipated success” in an obviousness determination conflicts with 35 U.S.C. § 103 and this Court’s precedent in KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 421 (2007)?

The obviousness analysis by the PTO is what I see as a somewhat typical pieces-and-parts analysis – finding that most of the elements of the claims were taught by one of the prior art references and then finding another reference that taught the remaining limitations.  Of course, this approach of using the invention as a blueprint for framing the prior art has the problem of likely hindsight bias – and is particularly troublesome where the invention is not simply a combination elements but instead is a new structure. Here, for instance, the claim requires, inter alia, “a ceramic composite layer . . . adapted to at least block dendrite growth. . .”  One reference purported to teach a non-ceramic layer used to block dendrite growth (Tobishima) and the second taught a ceramic layer – albeit one that that would not block dendrite growth because of large open holes (Tojo).  The PTO determined that it would be appropriate to combine the two references in a way that form a ceramic layer that blocks dendrite growth. In response, the petition argues:

Given the different problems addressed by Tobishima and Tojo (and their respective, different solutions), a threshold question for obviousness should have been whether it would have been obvious to combine the references to achieve a two-layer separator that blocks dendrite growth. Had this question been considered, which would have considered the claimed invention as a whole, the Patent Office would have found that the combination would destroy the very purpose of the references and would result in a separator with “open holes” that cannot block dendrite growth. Instead of considering the claimed invention “as a whole” and its unique ability to block dendrite growth through a ceramic composite layer, the Patent Office instead looked to whether the references contained all the elements of the claims. This analysis is not proper.

Although not the primary focus of the obviousness argument, the secondary factors of nonobviousness are interesting here.  The PTO gave them no weight even though LG apparently copied the invention and then touted the safety benefits of the ceramic layer in its advertising.

As in the Cuozzo case, the original IPR petitioner has dropped out of the case after a settlement, but the PTO is defending its own judgment.

= = = = =

[1] See IPR2015-01511 (Ube Maxell); IPR2014-00679 (SK Innovation); IPR2014-00680 (SK Innovation); IPR2014-00692 (LG Chem); IPR2014-00524 (Mitsubishi Plastics); IPR2013-00637 (Sumimoto Chemical).  See also, Celgard, LLC v. LG Chem, Ltd. et al 3:14-cv-00043 (W.D.N.C.); Celgard, LLC v. SK Innovation Co., Ltd. 3:13-cv-00254 (W.D.N.C.); and Celgard, LLC v. Sumitomo Chemical Co., Ltd. 3:13-cv-00122 (W.D.N.C).

[2] References: Unexamined Japanese Patent Application No. H5-190208 (hereinafter, “Tobishima”) and unexamined Japanese Patent Application No. H11-80395 (hereinafter, “Tojo”).