by Dennis Crouch
Although the result could have been guessed, the Federal Circuit has issued an important decision interpreting the scope of post-AIA 35 U.S.C. 102 — and the meaning of the “claimed invention.” Celanese Intl. Corp. v. Intl. Trade Comm’n, 22-01827 (Fed. Cir. August 12, 2024). In particular, the court affirmed the precedent of D.L. Auld — i.e., the on-sale bar continues to block patenting of an otherwise secret process when the patentee makes pre-filing sales of product made using that process. The ITC had invalidated Celanese’ artificial sweetener manufacturing process patent based upon these pre-filing sales. That judgment was thus affirmed on appeal.
Read the Decision: 22-1827.OPINION.8-12-2024_2365402
Background: Artificial sweetener Ace-K has been known for many years, Celanese invented and patented a new process for making the stuff and later filed a petition with the International Trade Commission (ITC) asking the agency to bar Anhui Jinhe Industrial and others from importing Ace-K made using their patented process.
The case includes a key undisputed fact: Celanese had been using its patented process in secret in Europe and selling Ace-K made by this process in the United States before September 21, 2015 – more than one year before the patents’ effective filing date. The question in the case was whether these pre-critical date sales trigger the on-sale bar under the AIA? The ITC’s Administrative Law Judge (ALJ) granted summary determination of no violation, finding the asserted claims invalid under the on-sale bar. Celanese appealed.
The statute in question – 35 U.S.C. 102(a)(1) – reads as follows:
A person shall be entitled to a patent unless—
(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Narrowing the focus for on-sale, the statute bars patenting if “the claimed invention was … on sale … before the [critical] date.” In Helsinn Healthcare S.A. v. Teva Pharmaceuticals USA, Inc., 586 U.S. 123 (2019), the Supreme Court addressed the AIA on sale bar, holding that sales trigger the bar even if not enabling to the public — affirming that pre-AIA precedent on this point had not been amended by changes in the statute. In Helsinn the patent covered the product being sold, not the process for making the patent as here in Celanese. That is seemingly an important distinction, and it is hard to explain under a strict textual analysis how a newly invented process is placed “on sale” based upon sales of the resulting product. However, pre-AIA precedent included parallel statutory language that was repeatedly held to bar such activities by the patentee.
Judge Reyna: The facts [in Helsinn] are different, but the legal issues pretty much the same, wouldn’t you say?
Deanne Maynard for Celanese: No, Your Honor. It’s different … in Helsinn, everybody agreed that the claimed invention itself was on sale. So that question was not before the Supreme Court. The question here is what is the meaning of ‘claimed invention’ in the AIA’s on-sale bar provision. The claimed invention itself must be on sale.
On appeal to the Federal Circuit, Celanese argued that the AIA fundamentally changed the on-sale bar’s scope. Their primary contention was that the AIA’s textual modifications to § 102 indicated congressional intent to exclude sales of products made by secret processes from the on-sale bar’s reach. Celanese pointed to several changes in the statutory language:
- The use of “claimed invention” instead of “invention” in § 102(a)(1).
- The addition of the catchall phrase “or otherwise available to the public.”
- Changes in related AIA provisions, such as §§ 102(b), 271(g), and 273(a).
Celanese argued that these changes, particularly the focus on the “claimed invention,” meant that only sales of the actual claimed process (not products made by it) could trigger the on-sale bar. They contended that the “otherwise available to the public” phrase implied a disclosure requirement for all prior art categories listed in § 102(a)(1), including “on sale.”
The Federal Circuit, in a unanimous panel decision authored by Judge Reyna, rejected Celanese’s arguments and affirmed the ITC’s decision. The court’s analysis began with a thorough review of pre-AIA on-sale bar jurisprudence, emphasizing the long-standing principle that sales of products made by a secret process could invalidate later process claims. The court cited seminal cases such as Pennock v. Dialogue, 27 U.S. 1 (1829), Metallizing Eng’g Co. v. Kenyon Bearing & Auto Parts Co., 153 F.2d 516 (2nd Cir. 1946), and D.L. Auld Co. v. Chroma Graphics Corp., 714 F.2d 1144 (Fed. Cir. 1983), to establish this precedent. Celanese had argued that these cases were not relevant because of the statutory change.
The Federal Circuit also relied heavily on the Supreme Court’s decision in Helsinn where the the Court wrote:
In light of this settled pre-AIA precedent on the meaning of “on sale,” we presume that when Congress reenacted the same language in the AIA, it adopted the earlier judicial construction of that phrase.
Helsinn. Although the word “invention” was changed to “claimed invention” within 102(a)(1), the court identified this as “no more than a clerical refinement of terminology for the same meaning in substance.” The court noted that pre-AIA precedent had often used “claimed invention” interchangeably with “invention” in on-sale bar contexts. As for the “otherwise available to the public” phrase added to 102(a)(1), the Federal Circuit echoed the Supreme Court’s that this catchall category was “simply not enough of a change” to conclude that Congress intended to alter the settled meaning of “on sale.”
The Federal Circuit also dismissed Celanese’s arguments based on other statutory provisions such as §§ 271(g) and 273(a) — both of which show that Congress understands how to particularly legislate on the distinction between products and processes, and, in the case of the 273 prior-user defense, the role of pre-patenting commercial use. The Federal Circuit appears to have generally rejected these arguments because they are tied to infringement statutes rather than those focused on patentability. “Patentability (or validity) and infringement are distinct issues concerning different actors and actions, governed by different frameworks with different rationales.” Celanese as well as an amicus brief cited legislative history indicating Congress intended to eliminate the forfeiture doctrine that previously barred patents on secret processes commercialized through product sales. The Federal Circuit largely dismissed this legislative history.
One argument that I find interesting relates to Harmonization. If you recall, one driving force behind the AIA was to harmonize the US patent system with that of the rest of the world., and the AIA legislation particularly states that promoting this harmonization is a goal of the legislation in order to “promote greater international uniformity and certainty in the procedures used for securing the exclusive rights of inventors to their discoveries.” Unlike typical non-statutory legislative history, this ‘sense of congress’ was adopted as part of the legislation and should be applied as an interpretative element. The amicus brief filed by the National Association of Manufacturers notes that the patent laws in other countries generally do not bar patents on secret processes commercialized through product sales — and that fact should weigh in favor of Celanese’ position. The Federal Circuit, however, did not address this issue.
For those in the daily grind, this decision is not a surprise, but it underscores the importance of prompt patent filing, especially when the patentee is concurrently engaged in commercial activities involving the invention. Take note, that the court appears to also maintain the W.L. Gore precedent that differentiates between the effect of an inventor’s own commercialization of an invention and commercialization by a third party.
= = =
Judge Reyna authored the unanimous opinion joined by Judges Mayer and Cunningham. Deanne Maynard of Morrison Foerster argued for the patentee-appellant Celanese and was also represented on the brief by Seth Lloyd, Brian Matsui, and Aaron Fountain. Benjamin Richards argued for the ITC. Nicole Saharsky of Mayer Brown argued for the accused infringer as intervenor and was also represented on the brief by Clark Bakewell, Gary Hnath, Bryan Nese, Minh Nguyen-Dang, and Scott McMurry. Brian Pandya of Duane Morris LLP represented amicus curiae National Association of Manufacturers.
It is notable here that Maynard and Saharsky are both prominent Supreme Court counsel — suggesting that the parties foresee this battle rising up to the next level.
Is there any argument to be made that the product sold in the past is different from the “product” resulting from the claimed process for making the product (i.e., “new and improved!”)?
Perhaps, but if the item previously was an actual trade secret, how is anyone going to know that something actually IS “new and improved?”
Congress expressed the desire to streamline and make sections 102 and 103 to be “race-oriented,” and as simple as “what is the date? With Y/N direct from that question.
This type of “preserving the old” is difficult to square with the stated reasons that were expressed with the AIA (and a primary reason why the first views from the Patent Office were as they were.
I thought prior art had to be publicly available. Guess not.
With the erosion of patent rights in the US, innovators are turning to trade secrets.
But if one relies on trade secrets, this decision is irreversible (no patent) after 1 year.
Checkmate.
Re: “But if one relies on trade secrets [for commercial product protection], this decision is irreversible (no patent) after 1 year. Checkmate.” Well put.
Although, in my experience, it was unknown to very many patent attorneys that this was always the trade-off since the Metallizing Engineering Doctrine’s CAFC adoption, and not AIA eliminated as the PTO and some others had argued, as I had warned in the 2011 article on this blog cited at 4.1 below.
Certainly, as the Giants DID sleep, AND as Congress did plainly state their intentions with AIA to eliminate “other considerations” that had made pre-AIA 102 and 103 rather unwieldy, it is entirely reasonable to have postulated that with the semi-race, the error of past Giants was being corrected.
It does appear that we will (yet again) need to await Congress to speak up with a “and this time We mean it” [can couple with eligibility fixes]
> I thought prior art had to be publicly available. Guess not.
I think you’re confusing the “on sale” and “public use” defenses.
That “checkmate” surprise IS going to be a rude awakening for those “affiliated” with the AIA influence and I do expect this realization will prompt action for Congress.
That lovely “Prior User Right” new submarine right is not going to be enough for those powers that be that €an $peak volumes to Congress and hold their attention.
Re: Celanese Intl. Corp. v. Intl. Trade Comm’n, 22-01827 (Fed. Cir. August 12, 2024). Particularly pleasing to see this Fed. Cir. decision in view of the comments discussion on the recent blog. Also, proving that the PTO was wrong in its commentary in its initial AIA enablement rules in saying that the AIA had overruled the personal forfeiture doctrine of Metallizing Eng. Co. v. Kenyon Bearing & Auto Parts Co., 153 F.2d 516, 520 (2nd Cir. 1946) which the Fed. Cir. had expressly adopted. The court cites Metallizing Engineering in affirming the precedent of D.L. Auld that blocks patenting more than a year later of a secret process when the patentee makes pre-filing sales of a product made by that process that does not disclose that process. Unfortunately, the Court chose to affirm the ITC decision on the grounds that “found Celanese’s asserted patent claims invalid under the on-sale bar, 35 U.S.C. § 102(a), because Celanese sold products made using the patented process more than one year before the effective filing dates of the asserted patents.” As Dennis notes, this requires an unnecessary contortion of the § 102(a) “on sale” bar statute. Selling an invention (e.g., a patent) does not trigger the bar, and selling products not containing the invention logically does not either. Which is precisely why the highly respected Judge Learned Hand developed the Metallizing Engineering Doctrine. To prevent one from having trade secret protection for many years for an invention used commercially to make and sell products, and to then tack on another whole term of patent protection only after the secret is out. A non-statutory personal estoppel that is not prior art and only applies to the would-be IPL serial protections attempter. Dennis says “Take note, that the court appears to also maintain the W.L. Gore precedent that differentiates between the effect of an inventor’s own commercialization of an invention and commercialization by a third party.” Note that there is no such differentiation for third parties selling products that DO contain or disclose the invention – that is prior art against anyone.
See my prior case law discussion of this issue in “The Ambiguity in Section 102(a)(1) of the Leahy-Smith America Invents Act” 2011 Patently-O Patent Law Review 29.”
Nice comment, Paul. Can you clarify this bit below?
“Selling an invention (e.g., a patent) does not trigger the bar”
I assume that selling a non-patent description of the invention at issue also does not trigger the on-sale bar. I’m just not sure why either point is relevant to the case you’re discussing …
Re ““Selling an invention (e.g., selling a patent) does not trigger the [on sale] bar.” [There is a CAFC case on that] I was merely using that to illustrate the semantic or other confusion as to the 102 statutory “on sale” patenting bar as between inventions, products made by inventions, and products containing an invention.
Someone else below is apparently confusing the issue here with the U.S. AIA one year grace period. An inventor with a trade secret invention that has not been for more than one year; disclosed, publicly used AND [this decision] not Commerically used for more than one year, can, of course, validly file for a U.S. patent. Othewise, not. That happens all the time.
Emphasis added, but this too shows that the AIA is indeed different (with the semi-race mode):
“… and to then tack on another whole term of patent protection only after the secret is out.”
If indeed, the secret is out, then no one gets a patent.
Instead, and even if merely others are ‘getting close,’ the actual patent bargain is still being met regardless of any other protections under different parts of the Constitution.
This seems right. I appreciate the argument that sale of the product-made-by-TS-process should have been distinguishable from the TS-process itself. But then I think that there may be services which are in essence only the TS-process which may then later be patented, extending the overall IP protection for that process beyond what law and policy intends.
(Am I thinking of all those financial trading patents that were all coincidentally “invented” following State Street Bank, which just might have been trade secrets at the NYSE trading firms for decades, and then needed CBM as a specific post-grant tool to invalidate the ones that were actually enforced where CBM had a sunset interestingly timed to be about 20 years after SSB? Perhaps.)
The now-sunseted AIA Covered Business Methods (CBM) post-grant PTO proceeding was indeed triggered by those obtaining patents on software systems some banks had been secretly using for years. But the Metallizing Engineering Doctrine and this case only applies to the trade secret invention inventors, not others independently inventing and patenting the same idea. Another section of the AIA expanded prior secret commercial users protection.
Yes – the sua sponte Prior Users Rights. A true submarine as such would only ever surface after another had committed to the Patent process, and obtained what they were to believe would be exclusive rights for partaking in the Quid Pro Quo.
I cannot recall if those Prior User Rights were ever actually used. I do not recall even a single case of any such use here at the O.
The AIA change I noted above is the amendment of 35 USC 273 “Defense to Infringement Based on Prior Commercial Use.” One could use it if you commercially used the subject matter in the United States, either in connection with an internal commercial use [this covers one’s secret use] or an actual arm’s length sale or other arm’s length commercial transfer of a useful end result of such commercial use, more than a year before the patent being challenged was filed on. But, there are conditions on this defense in this statute far more restrictive than a 102 or 103 prior art defense. It could not be asserted in a reexam or IPR either. I have never seen any Fed. Cir. decision indicating that this statute was asserted below.
Like I always tell clients: at the very least, file a provisional before you make it, use it, sell it, or offer to sell it! And that includes disclosing it to the public at trade shows, in publications, etc.
I wonder, does the case provide a matrix of fact suitable for debating what a patent system is for, what is the public interest?
If the patent system is to encourage early filing and, 18 months later, the disclosure to the public of an enabling disclosure at the earliest achievable date? Or is it to stimulate technological innovation in particular by allowing inventors to develop in secret processes of manufacture of products which, when exploited commercially, enrich the economy and then, later, when others are close to discovering the secret, filing on the process to get another 20 years of exclusivity? Are patents to promote inventive activity or trading activity?
Of course, the risk of delaying filing is that a competitor gets to the USPTO first, and comes out of it with a not invalid and not unenforceable patent with wide claims. Often, as Dennis observes, it is too great a risk, to hold back filing a patent application. So does it make any difference in practice to the speed and scale of technological innovation, whether the world adopts or does not adopt an on sale bar? I doubt it. Let’s just keep the patent system as simple as possible, shall we?
Let’s not overlook the differences in Quid Pro Quo (and why the US still maintains a No-Publication route).
In the US, that Quid is a granted patent for the Quo of publication.
Your EPO offers the far less Quid of a chance at grant for the Quo of publication.
My question is why the “on sale bar” in the US patent system, and only in the USA, even though the USA has now replaced First to Invent with First to File
In other words, why (only in the USA) in the event of a prior sale, the Quid (upon delivery of the Quo of publication) fails to materialise. What public policy objective is delivered by the on sale bar, that USA sees as vital but ROW fails to see at all. In other words, the on sale bar, in this day and age, what does it bring?
It brings continued conflation of two very different types of “time protection.”
Note as well that this confusion is rooted in a “make” scenario whereas time protection under patent has no actual “make” requirement.
Further, the notion of “another “X” years” confuses and conflates two very different protection regimes.
Trade secret protection regime stems from the Commerce Clause and is ruled over on a State by State basis.
Patent protection falls under a very different Constitutional clause and exists only at the Federal level.
No guarantee to the length of time of Trade Secret protection and certainly not having any type of mandatory “for limited times only.”
While you MaxDrei at least have some thin excuse (not much of one, since I know that I have shared these points with you), those attorneys (and judges) in the US have no excuse – not even ‘historic’ ones for still confusing and conflating these very different periods.
The stated Constitutional purpose of the patent system is to promote the progress of the usful arts. That is not promoted by rewarding with patent protection inventors who prevent disclosure of that progress, and deny its competitive benefits to the consumer-public, for as many years as they could or wanted to, be so protected by trade secrecy laws.
The two ARE separate and should not be conflated.
The only actual question – at the time of filing and examination – is does this or does this not promote the existing progress for all.
There IS value in promoting to be open that which was secret.
This is not difficult.
Sooner than anyone is STILL, well, sooner than anyone.
The fact that Giants once slept and confused and conflated two very different times of protection should not serve as a “let’s just shut down our brains” excuse.
In much shorter terminology, “New to You,” is still “New.”