Thursday, March 12, 2015

Board finds non-analogous art when Examiner mistakes advantage in reference with problem solved by reference

Takeaway: In appealing an obviousness rejection of a semiconductor device, the Applicant argued that one of the references was non-analogous art. The Examiner asserted that the anti-fogging film disclosed in the reference had the same purpose – excellent durability and wear resistance – as the claimed invention, and was therefore analogous. The Board found that the Examiner had erred in characterizing the problems. The Board found that the problem solved by the invention was "protecting a metal layer of a semiconductor from corrosion with an aluminum complex oxide layer". The Board found that reference solved a different problem: fogging of window glass, mirrors, and lens. The Board therefore reversed the § 103 rejection. (Ex parte Seo, PTAB 2013.)

Details:
Ex parte Seo
Appeal 2011-005820; Appl. No. 11/378,799; Tech Center 2800
Decided:  October 11, 2013

The application on appeal was directed to a semiconductor device for an LCD display. The independent claim on appeal read:
     1. A display device comprising:
     a substrate;
     a metal layer formed on the substrate and including a top surface and a side surface, wherein the metal layer comprises aluminum or an aluminum alloy;
     an insulating layer covering the metal layer; and
     an aluminum complex oxide layer disposed between the top and side surfaces of the metal layer and the insulating layer, wherein the aluminum complex oxide layer is formed by applying to the metal layer a coating solution that comprises at least one selected from the group consisting of zirconium, tungsten, chromium and molybdenum.
The Examiner rejected the originally filed independent claim 1 as obvious over Doushita in view of Applicant Admitted Prior Art. The Examiner relied on Doushita for the aluminum complex oxide layer element (location and composition), and AAPA for everything else. As a reason for combining, the Examiner asserted that Doushita's coating would provide "the structure of the admitted prior art with 'excellent durability, wear resistance, anti-fogging property and anti-fogging sustainability property'. "

During prosecution, the Applicant made various amendments, but the Examiner maintained the obviousness rejection using Doushita and AAPA. The Applicant also made one argument: that Doushita was non-analogous art. As a preliminary matter, the Applicant noted that Doushita and the invention were in different classifications: Class 428 "Stock material or miscellaneous articles" as compared to Class 438 "Semiconductor device manufacturing: process." Moving on to the test for analogous art, the Applicant first discussed the respective fields of invention. Doushita disclosed an anti-fogging article used for "buildings, vehicles, optical components, [etc.]," but this list did not include a semiconductor device (Applicant's field). The Applicant then discussed the purpose or problem solved: for Doushita, anti-fogging, with durability and wear resistance as secondary characteristics; for the Applicant, corrosion of aluminum, with secondary characteristics of electrical resistance, electrical conductivity, and adhesive strength. Therefore, "Doushita is not reasonably pertinent to the specific problem with which the Applicants were involved."

During prosecution, the Examiner did not directly respond to all of the Applicant's non-analogous art arguments. The Examiner did clarify that Doushita was "not used for the claimed structure but for the substitution of the conventional aluminum oxide layer formed on the line." The Examiner also explained that several of Doushita's applications for the anti-fog coating (i.e., "optical component," "medical equipment," "mirror," "lens") amounted to an application for semiconductor devices.
 
After several rounds of prosecution (including an RCE), the Applicant appealed. In the Appeal Brief, the Applicant reiterated the previously made non-analogous art argument, concluding with: 
Applicants fail to understand how the problem of anti-fogging of window glass, mirrors, and lens is reasonably pertinent to the problem of protecting a metal layer 212 of a semiconductor from corrosion with an aluminum complex oxide layer 213.
 The Applicant separately argued that the Examiner's rationale for combining was insufficient:
     Doushita simply discloses an aluminum oxide used as an anti-fogging article. Nowhere does Doushita disclose the structural relationship of the aluminum oxide layer relative to any other layer(s). Doushita fails to disclose the aluminum complex oxide layer is disposed between the top and side surfaces of the metal layer and the insulating layer. As such, without more structure specifics, Applicants submit the complex oxide of Doushita is functionally and structurally different from Applicants' aluminum complex oxide layer used to prevent a corrosion of the metal layer.
     Therefore, there is no suggestion to combine AAPA and Doushita with each other so as to obtain the technical features of Applicants' claimed subject matter.
Notably, the Applicant did not argue that the combination failed to disclose the claimed limitations.

The Examiner's Answer include (for the first time) a specific response to the non-analogous art argument. The Examiner asserted that Doushita's teaching of an aluminum complex oxide layer with "excellent durability and wear resistance" was "the same purpose as the claimed invention" and was therefore "reasonably pertinent and has enough motivation to be considered." Addressing Applicant's argument about classifications, the Examiner had "considered or cross referenced" the "miscellaneous articles" referred to in the title of Class 428. "Further, 'optical component' is a semiconductor device; and  Class 438 'Semiconductor device manufacturing: Process' deals with this device formation."

The Board  reversed the obviousness rejection. The Board first explained that the field of endeavor was determined by looking at "explanations of the invention’s subject matter in the patent application, including the embodiments, function, and structure of the claimed invention.” (In re Bigio, 381 F.3d 1320, 1325 (Fed. Cir. 2004)). The Board then characterized the problem solved by the reference (fogging of window glass, mirrors, and lens) and the problem addressed by the claimed invention (protecting a metal layer of a semiconductor from corrosion with an aluminum complex oxide layer).

The Board concluded that the anti-fogging properties of Doushita’s film "would not have logically attracted the attention of an artisan possessing knowledge of Appellants’ Admitted Prior Art." Despite Doushita's mention of "optical components" as a use for the disclosed anti-fogging article, the Board found that the anti-fogging properties "would have little utility on a semiconductor device."

As an independent basis for reversal, the Board agreed with the Applicant's second argument that the Examiner's rationale for combining was insufficient. Although Daushita taught the use of aluminum oxide as a coating and listed various advantageous properties of the coating, Daushita lacked the structural details of the oxide that were recited in claim 1. The Board agreed that "without more structure specifics, ... the complex oxide of Doushita is functionally and structurally different from Applicants' aluminum complex oxide layer used to prevent a corrosion of the metal layer."
My two cents: An important part of analogous art analysis is framing the problem. Here, the Examiner confused an advantage disclosed in the reference for a problem solved by the reference. Which seems to be related to focus. It's common for a reference to have a laundry list of advantages, Here, that list did include an advantage – wear-resistance – that overlapped with an advantage of the invention – corrosion protection.  But as suggested by the title ("Non-fogging article and process for the production thereof"), the problem solved by the reference was not susceptibility to wear, but fogging. Kudos to the Applicant for honing in on this winning argument.

7 comments:

  1. This one should be read by all the practitioners who say, "I never argue non-analogous art because the examiner is never persuaded by it." Yeah, the examiner wasn't persuaded, but 3 lifer APJ's were.

    Make the argument.

    ReplyDelete
    Replies
    1. >Make the argument.

      Meaning prosecute with an eye to what arguments work before the Board?

      I suppose you could pursue an Examiner-centric prosecution strategy up to the time of appeal, then switch to a Board-centric strategy in the Appeal Brief. E.g., save your Non-Analogous Art and Rationale-To-Combine arguments for the Board.

      OTOH, a completely understanding of the Examiner's position gives you the best chance of winning. Yet many Examiners don't really explain until you make explicit arguments.

      So if you wait until the Appeal Brief to argue NAA or RTC, then the Answer will probably be the first time you get a good explanation of the rejection. Which gives you only one shot -- the Reply Brief -- to make a compelling argument based on a full understanding.

      My personal philosophy is to appeal once the issues are fully developed, but not before. So I argue sooner rather than later.

      >the examiner wasn't persuaded, but 3 lifer APJ's were.

      FWIW, my gut tells me a Non-Analogous Art argument works with Examiners better than a Rationale-To-Combine argument. NAA doesn't win a lot, but more than RTC.

      Delete
    2. Make the argument.
      Oh, I make them. However, you still get the lifer APJ who writes "the Examiner stated that the prior art is analogous because they both require a power supply" -- when my art is a hard drive and the applied prior art is an electric typewriter.

      Meaning prosecute with an eye to what arguments work before the Board?
      ALWAYS write your arguments with an eye to what works before the Board -- assuming that you have a client that will give you leeway to appeal when necessary. You make the argument -- even if it is over the Examiner's head, and when you appeal, you write "Appellants argued X and the Examiner didn't respond to X (or poorly responded to X)." If the Examiner writes an Answer, you'll then pounce on what is probably a poor misstatement of the law and/or facts.

      There may be times when I'll hold back some arguments as to dependent claims during normal prosecution (e.g., if I already have plenty of arguments already made as to the independent claims), but I never hold anything back as to the independent claims.

      My personal philosophy is to appeal once the issues are fully developed, but not before.
      Two points. First (and consistent with my statement above), you should be developing all the issues from the get go. My personal philosophy is to get an examiner to write as much as possible because the more he/she writes, the more errors is reasoning and/or facts and/or law I'll be able to find.
      Second, waiting for all the issues to fully develop may be to late. You only have a very limited time to develop them before you are forced to either file a RCE or appeal. Filing a RCE permanently tolls type-B PTA (which could add years of term to your clients patent -- once issued). If I think I have decent arguments to make before the Board (no matter how well the issues are developed), I'll appeal.

      Delete
    3. >get an examiner to write as much as possible

      Absolutely! In my experience, this does require the Applicant to put something substantive on record -- the Examiner typically expands only when the Applicant goes beyond the naked assertion "does not teach." IOW, you can't count on the Examiner to do any more than address whatever specific, substantive arguments you make.

      >limited time to develop [issues] before you must RCE or appeal.
      >Filing a RCE permanently tolls type-B PTA

      Sure. And different clients take different approaches to this. [Which is why I characterized it as "personal" and why I consider it more of a "philosophy" than strategy.]

      The clients I work for aren't as concerned about patent term as about other factors, perceiving the bulk of their patents to have more value at the start of term rather than end of term.

      I have worked for clients with a one-RCE-then-appeal policy. As far as I could tell, the policy was about reducing the cost of endless rounds of prosecution rather than avoiding PTA.

      Delete
    4. As far as I could tell, the policy was about reducing the cost of endless rounds of prosecution rather than avoiding PTA.
      This is what appealing is for. If you (or your client) gets a reputation for always doling out an RCE before appeal, then that is what you will get.

      Delete
  2. As usual Karen, nice find! Thanks.

    ReplyDelete
  3. Thank you, Karen.

    ReplyDelete