Obviousness: Claim Construction (Q.of.Law) vs. Interpretation of the Prior Art (Q.of.Fact)

by Dennis Crouch

In re Aspen Aerogels, Inc. (Fed. Cir. 2020) (non-precedential)

The basic question in this ex parte appeal is whether the asserted references teach the layering of aerogels as required by the insulation materials claim pending before the USPTO.  The pending APN 14/446,663 was filed in 2014, but claims a chain of priority stretching back to a 2005 provisional patent application. The basic idea here is that aerogel is a great insulator, but is fragile, so the claim requires fiber-reinforcement.  The claim particularly requires two plys of fiber-reinforced aerogel — and that the fibers in the plys are interlaced with one-another.

Claim 1: A composite comprising

[a] at least one first ply of fiber-reinforced aerogel material adjacent to [b] at least one second ply of fiber-containing material,

wherein fibers from the at least one first ply of fiber-reinforced aerogel material are interlaced with fibers from the at least one second ply of fiber-containing material.

The examiner rejected the claim as obvious and that was affirmed by the PTAB. The Federal Circuit has now affirmed.

  • U.S. Patent App. No. 2002/0094426 (Stepanian) taught having the two layers of fiver-reinforced aerogel, but did not teach the interlacing of fibers.
  • Japanese Patent No. 2000-080549 (Sano) teaches interlacing of fibers of layers using needle-felting.

Aspen’s argument on appeal is that Stepanian does not actually teach plys as claimed but rather two masses of undifferentiated aerogel.  On appeal here, Aerogel presented this as a question of law involving the definition of the term claim term “ply.”  The Federal Circuit here rejected that approach — holding instead that the real question is whether the prior art discloses the plys.  The difference between these two approaches is very important on appeal: Claim construction is reviewed de novo on appeal without deference to the PTO determination; Content of the prior art is reviewed for substantial evidence with major deference going to the PTO.

[W]e determine the issue to be a factual question of whether the prior art teaches an undifferentiated mass of aerogel or teaches layers of fiber-reinforced aerogel as the Board determined. We review this issue for substantial evidence.

Slip Op.  Stepanian (which I have previously discussed in PatentlyO) clearly shows layers, and particularly defines Fig 3 (below) as disclosing “a 3-layer laminate.”  But, the issue is that Stepanian stacks the fibers and then pours in the areogel over the layers of fibers.  The PTAB made the factual conclusion that Stepanian disclosed “plys” of fiber-aerogel, even with the pour-over approach, and the appellate panel found substantial evidence for that conclusion.

Whether layers of fibrous material are filled with aerogel and then stacked or stacked and then filled with aerogel makes no difference if the result is the same as the claimed composite here.

Slip Op.

Note here, that Aerogel is also the owner of the Stepanian patent application (and issued patent). The problem though is that the Stepanian application was published back in 2002, more than 1-year before the priority date of the application at issue here.

Rejection affirmed.

If there really is a claim-construction issue here, on remand the patentee may want to amend its claims to make the proper construction exceedingly clear and then rejoin the rodeo.

= = = =

If we go back to the heart of this matter: Aergoel appealed on claim construction, the Federal Circuit says “the real issue is not claim construction but rather the content of the prior art” (paraphrase).  The court then goes on to refuse to decide the claim construction issue and instead decide the case on the alternative grounds.

Truthfully, this seems wrong. The appellate court should have decided the issue raised by the appellant (claim construction) and ended the case there.  I read appellant’s brief who particularly stated that no factual question is being appealed: “There is no dispute here about any relevant fact.”

= = = =

Poongs Muthukumaran, chief counsel for Aspen argued for his employer; Kakoli Caprihan argued for the PTO.

33 thoughts on “Obviousness: Claim Construction (Q.of.Law) vs. Interpretation of the Prior Art (Q.of.Fact)

  1. 5

    “Truthfully, this seems wrong. ”

    Sure does. Still, from the standpoint of the wording of the decision which they were reviewing supposedly the PTAB made a factual finding of “plys” (I guess explicitly in the decision) being present in the reference. If that be the case, and it is found explicitly as fact that plys are present then you can’t dance to claim construction to save your argument/claims. You’ve got to directly address the, according to you, improper factual finding first, because otherwise the feds won’t review/change the factual finding (no de novo review) and you’re stuck with the wrong factual finding from the get go (you need to get the panel to word their decision more finely and not use the term plys in their decision as a factual finding and get them to find a stack of layers or whatever instead then get them to interpret that stack as “plys” explicitly through broad interpretation). So in a backwards round about way, the feds are correct here, though it appears the PTAB may have made a mistake in making their factual finding, and the court may should have reviewed that for being clearly erroneous for the applicant/patentee.

    1. 5.1

      Nope. There’s really no “dispute” as to what the reference discloses. That “fact” is not at issue. What is at issue is: does the reference disclose what is claimed? That issue is one of claim construction. Which is a question of law.

      1. 5.1.1

        In Markman, the Supreme Court held that judges, not juries, would evaluate and decide the meaning of the words used in patent claims. [Hence a question of law.] Judges were to look at the written description accompanying the patent claims as most relevant; the the application history; standard dictionaries, and lastly if necessary expert testimony from experts “skilled in the art” in issue.
        But “does the reference disclose what is claimed” is not the same as that.


          “The Federal Circuit here rejected that approach — holding instead that the real question is whether the prior art discloses the plys. ”

          How did the Federal Circuit propose to resolve that “real question” without resolving the question of law as to what the term “ply” means?


            A quick check indicates that “the real issue” is NOT even whether or not a “ply” means a “layer,” but a much more critical item is how any such ‘plies/layers’ interact one with another.

            I suppose the claim terms can be made into a nose of wax that the courts could bend and shape to fit their desired Ends…

      2. 5.1.2

        I went ahead and read the decision a bit last night, D’s writeup is slipshod reporting at best. They literally contested facts, claim construction, on and on and on according to the actual decision. Furthermore, the court explicitly reviewed the fact findings for substantial evidence and found that such was present. Tho u or I might disagree.

  2. 4

    @ Atari Man

    You were 2/3 correct: Dyk and Hughes were on the panel together with Schall. The opinion was written by Hughes.

    A scan of all three judges in Wikipedia reveals that none of them is recorded as having a degree in science and none of them specialised in IP law prior to appointment at the Federal Circuit.

    The subject-matter in this case appears extremely simple, but even so a scientific background might be helpful.

    One lesson for those of us who write patent specifications is that the correct addressee for the patent specifications that we write is not a fellow scientist or engineer but instead is a judge with an arts degree or in the case of a jury trial randomly selected jurors who mainly will not have a technical background. If that does not influence our drafting style both from the standpoint of the description and from the standpoint of what is claimed, then it should.

    1. 4.1

      I agree with Paul, with the caveat that writing applications (and claims) have multiple audiences (and besides the two that Paul mentions (judges and juries), two other important audiences include the patent office examiners and those business/technical folk who make licensing decisions.

    2. 4.2

      Nice point, Paul, about who is the addressee of the specifications and claims we write. The way I see it, there are two quite different addressees, equally important in their different ways, namely, the one you identify (being the one in fact) and the one in theory, the legal fictional one (being the PHOSITA).

      Until my curiosity was piqued by your post, Paul, I had not engaged with the issue here. It seems to me now though that this case provides a great teaching example. For one thing, how much difference (if any) does it make to the attributes of the layered fibre-reinforced product, which step comes first, gel-filling or layer-stacking. For another, setting aside obviousness for a moment, and assuming (and why not?) that when you inspect the finished product you cannot tell whether it was stacked first or filled first, why doesn’t Stepanian destroy novelty of the claim to that product?

      How about a method claim? Is there one?

      1. 4.2.1


        How exactly does a writer of an application have in mind the position of the legal fiction of the Person Having Ordinary Skill In The Art?

        While it certainly “sounds nice,” how is it actually done in practice?

        Keep in mind as well that while it may indeed be Best Practice to have as thorough of a prior art search done as the client can choose to afford (which, by the way, is one way to answer my question above), there is NO legal requirement for Clients to run a prior art search whatsoever, and no matter what an attorney may recommend, it is and remains the choice of the client to decide that level of inputs into the writing process.

        Complicating this of course, is the additional fact that WHEN WRITING, an attorney has no clue whatsoever what possible art an examiner MAY CHOOSE to use. And of course, by the time that an examiner DOES make that choice, the ability to ‘write’ has long lapsed (both of our respective sovereigns having ‘No New Matter’ rules).

        Certainly, choice of key words and the Art of story-telling may help dictate the choice of ‘key words’ that an examiner in the end may use (and over at IPWatchdog, they have run a number of stories — tied to a particular vendor that offers a service along that line — of how to ‘aim’ to have your application placed in a particular examining art unit), but is that what you have in mind with our comment?

        As to the particular facts of this case, (if I recall correctly, and somewhat NOT shown in the article here), is whether or not the “within layer” stacking orientation is important. One may well have a plurality of ‘layers,’ but if each of those layers have within themselves a mere random placement of fiber orientation, then the physical effect of multiple layers is simply not going to be the same as having layers of different ‘within’ orientations used as plies.


          Yes, patent applications need to be written for several different potential audiences. Those actually enforced must be read for lay juries, non-technical and non-patent-lawyer D.C. judges outside of Waco TX, yet quite possibly also read by IPR APJs with both technical and patent law educations. Even “the legal fiction of the Person Having Ordinary Skill In The Art” is an “audience.” Because technical terms used in the spec and claims are legally required to be interpreted by both the PTO and the courts as to what they mean to the POSITA as of the filing date, unless specifically defined differently in the spec.



            I hear what you are saying (echoing MaxDrei), but the post you are responding to is seeking more than just that echo.

            What does it ‘actually mean’ to write for an audience of that legal fiction known as Person Having Ordinary Skill In The Art (from a hands-on practical view and not just a platitude)?

            Keep in mind the points I presented: the choice of level of inputs (up to and including NO prior art search whatsoever) are the constraints facing the writing attorney.



              The point is that whoever else needs to understand the specification and figure out what the invention is, it must be understandable by Judge Reyna. Res ipsa loquitur, as we lawyers say.


                Anon again

                Incidentally, when drafting patent specifications for new filings I ALWAYS do personal searches through the USPTO online databases and especially through Esp@cenet which is invaluable. Keyword searches are quick and useful, and when you find useful references, Esp@cenet has a both past and forward citation search facility which greatly speeds the finding of relevant background material. You may not find everything, but you can find a lot, and very quickly. And doing it yourself is educational and helps the drafting process.

                1. Mr. Cole, I caution AGAINST your suggested approach – unless you have strict permission from the client for doing so.

                  This is NOT an issue of ‘best practices’ as much as it is understanding who has the final say on inputs to your writing.


            This is an invention in the field of fibre-resin composite materials, a field in which the complex physics and chemistry of surface-to-surface contact is key. Dennis quotes the court:

            “Whether layers of fibrous material are filled with aerogel and then stacked or stacked and then filled with aerogel makes no difference if the result is the same as the claimed composite here.”

            in which for me the key word is “if”. I mean, as a fact, is the result the same or is it different? My intuition tells me that if (as in the prior art) the fibrous layers are stacked before any gel is poured over the fibres they will be “interlaced” whereas if the layers are filled with gel before being stacked then the fibres will not be so “interlaced” (because they will be already wetted by the gel ie inside the gel volume of their respective layers and so after stacking will still be within what were distinct layers). Yet the claim is said to deliver fibere-interlacing, the prior art not. Huh?!?

            My thoughts are that if you want to avoid being enmeshed in endless litigation, the patent drafter should first understand the invention (its technical features and what effects they bring)
            before drafting and then, having understood it, explain it in a way a judge can grasp, even if that judge will need help to understand the terminology used by the skilled person and so, of necessity used throughout the drafted specification. After all, for sufficiency of description it is not necessary, is it, thank goodness, that the specification be effective to educate the judge, in the field of the invention, all the way up to the level of a skilled addressee?


              that the specification be effective to educate the judge, in the field of the invention, all the way up to the level of a skilled addressee?

              Again – a nice sounding platitude, but entirely unworkable in the real world.

              Are you even aware of the range of understanding of judges to whom you may need to understand their level of understanding in order to be able to write as you suggest? Compound that with the notion that you are indeed writing for future judges.

              What types of budgets do your clients give you to write these ‘masterpieces?’

              (on top of that, do you also write with any awareness of patent profanity, notably applicant admitted prior art?)


                You seem to have read the sentence so fast that you overlooked the presence in it of the word “not”. Otherwise, I don’t understand what you are trying to say.

                What I’m saying is that the addressee that the drafter needs to have in mind is not only the judge but also the PHOSITA , whereby the drafted specification is as lengthy as it needs to be (both to get the Exr/Board/Judge onside and to enable the skilled reader) but not one word longer than that.

                That way, the cost of drafting is minimised and the specification is compact. The rest is, as you say, patent profanity.

                1. so fast that you overlooked the presence in it of the word “not”.

                  That I did – mea culpa.

                  (we actually appear to be MORE in agreement than disagreement – but I am still trying to ascertain what this ‘writing to the PHOSITA means outside of being a nice sounding platitude)

  3. 3

    The lesson here is that if you have a product improvement, file an application on the improvement before either the product or any patent or publication disclosure of it is public if you can.

    1. 3.1

      Well, yes, Paul. Since time immemorial, the watchword in any First to File jurisdiction is “Every time you manage to find a way to take a further step forward, however small, make sure you tell the Patent Office about it, before you tell anybody else.”

      1. 3.1.1

        Not quite Max – the point is to tell the further step forward to the patent office before the prior step is public [and thus potential prior art] if you can. [Worse, some people even think that a U.S. CIP application avoids this problem.]


          OK, Paul, but infected as I am with FtF thinking, it is axiomatic to me that one tells the PTO without any delay, lest some other filer nips in before you do. But if you don’t have your step forward fit to file before the 18 month PTO A publication of your original application you can’t file your follow up applications before the original filing publishes.

          Lest anon jump in and castigate me for not mentioning the option to opt out of 18 month A publication, I mention it.

          More interesting is the difference between jurisdictions on the issue whether inventor’s own earlier but not yet A-published filings count as prior art and, if so, whether also for obviousness, or just for novelty.


            I chuckle – even as you attempt to paint my post with the key US attribute as some type of negative thing (neither that item, nor my ‘castigating’ are negatives).

            Those who have less than World Domination on their mind should always be cognizant of the trade-offs involved with sacrificing the US offered path of non-publication (especially when one considers that publication is always a later option to take once a non-publication path has been started down upon).

            I also chuckle as you yourself appear to feed a past discussion point of ours (even as you never really fully join that prior discussion) on the mechanism of the legal fictional person called Person Having Ordinary Skill In The Art, and the consistent treatment of that person across both legal concepts of anticipation and obviousness.

  4. 2

    Who were the judges who ignored the appellant’s point of appeal and decided based on a point not raised and not in dispute? Oh, right, Dennis thinks that CAFC judges are fungible so there’s no need to tell us who was on the panel.

    I’ll wager Judge Newman was not among the three. Nor were Judges O’Malley or Moore.

    But there’s a good chance that at least one of Prost, Dyk and Hughes were on the panel.


            Thanks Night Writer,

            Perhaps I could be a bit more precise. How about:

            Ben, your post only looks bad FOR YOU in that it is obvious that you miss the actual point of what Atari Man provides and makes it look like you are just trying to muddy the water and obstruct the point being presented.

            When you post like this, you FEED the impression that certain others have of you (for example, Night Writer) and you appear to be nothing more than a t00l for those seeking to denigrate patent rights.

  5. 1

    “Truthfully, this seems wrong. The appellate court should have decided the issue raised by the appellant (claim construction) and ended the case there. ”

    CAFC shrugged.

    Atlas would be proud.

Comments are closed.