Wednesday, January 21, 2015

PTAB decisions involving failure to specify a limit in a claim

Today's post will consider various non-prior art issues that might arise when a claim fails to include either an upper or lower limit. In some of these cases, I'm not sure the Applicant even realized a range or limit was implicated.

Ex parte Akers (BPAI 2010) involved an enablement rejection for "at least one" with no upper limit in the claim. (Essentially a recap of a post from 2012, see the full post here: http://allthingspros.blogspot.com/2012/08/bpai-enablement-full-scope-at-least-one.html).
The claim in Akers was directed to using radiation to test various materials, and the language at issue read "detecting at least one emitted prompt gamma ray." The Examiner gave an enablement rejection, asserting that "[i]t is a notorious scientific fact that statistically meaningful data requires collection of information from not one but from a plurality of events and nuclear interactions." The Applicant argued on appeal that the specification need not enable every embodiment, and the Examiner must show that all multiple ray embodiments were inoperative or not enabled.

The Board affirmed the rejection in Akers, in part because the Applicant admitted that detection did require more than one ray: "As such, Appellant has conceded that the disclosure of the present application does not provide an enabling description for the full scope of Appellant’s claims. Appellant, having chosen the broad claim language, must make sure that the broad claims are fully enabled."

The claim in Ex parte Yamazki (PTAB 2011) recited “a channel length of said semiconductor device is 0.3 m or less." The Examiner issued rejections for inoperability (a type of § 101 rejection) and indefiniteness, used the same rationale for both: the claim did not recite a lower limit for the channel length. The Board reversed both rejections, treating this as a mater of claim construction. The Examiner interpreted the claim to cover embodiments with a zero (or infinitesimally close to zero)  channel length, which would be inoperative according to the Examiner. The Board looked to the specification, which described the channel length as "typically from 0.1 to 0.3 μm." According to the Board, a POSITA informed by the specification "would not have viewed the claim as reading upon non-working embodiments." Similarly, a POSITA informed by the specification would not have found the claim indefinite.


Semiconductor devices were also the subject of the application in Ex parte Lin (PTAB 2011). A dependent claim recited "the photomask of claim 1 wherein the wavelength reducing material has a refractive index larger than 1." The Examiner rejected this claim under § 112 First Enablement, asserting that a POSITA would not know how to make a photomask having a refractive index with no upper limit. On appeal, the Applicant pointed to a section of the Specification which described materials having a refractive index ranging from about 1.3 to 3.5. The Board noted that the issue was not whether the specification described some materials with a refractive index larger than 1, but was instead enablement through the entire scope of the claim (citing In re Goodman, 11 F.3d 1046, 1050 (Fed. Cir. 1993)). The Board affirmed the rejection because the Applicant did not address this specific issue on appeal.

 The technology in Ex parte Drzal (PTAB 2011) was graphite composites. The claim language at issue was "expanded from a graphite ... by boiling the intercalcant with radio frequency or microwaves for up to five minutes." The Examiner rejected the claim under § 112 First Written Description, alleging that the specification failed to provide support for the entire range "for up to five minutes." On appeal, the Applicant pointed to these teachings in the specification:
     Heating for 3 to 5 minutes removes the expanding  chemical. ... The intercalant heats up past the boiling point and causes the graphite to expand to many times its original volume"
     In example 1, expanded graphite was prepared by exposing intercalated graphite flakes ... to microwave energy, typically at 2.45 GHz frequency, for a few seconds to a few minutes in an oven.
The Applicant then argued that since the claims did require "boiling," which must necessarily take some period of time, the claims did not encompass zero minutes.

The Board in Drzal interpreted "boiling for up to five minutes" as encompassing a fraction of second, though not zero. "Appellants have not identified, nor do we find, any basis in the Specification or claims which supports a narrower interpretation of the claimed range." Since the Specification did not disclose at what point during the exposure the intercalcant begins to boil (emphasis in original), the Board found the Applicant had not demonstrated possession of a compound as claimed.

4 comments:

  1. All good decisions. As for the Yamazki case, I'm not an electrical engineer, but I'm left wondering if a "channel" would still exist if it had no length. If a channel would cease to exist as its length approached zero, then it seems like the term "channel" itself would be a limitation that requires some non-zero value for the length. I don't know if that's a technically reasonable way to look at the matter, but it seems "cleaner" than importing limitations from the spec.

    ReplyDelete
    Replies
    1. >If a channel cease to exist as its length approached zero, then "channel"
      >itself would be a limitation that requires non-zero value
      >[this reasoning] seems "cleaner" than importing limitations from the spec.

      Gotcha. You're referring to the Board's mention of the spec when interpreting the claims.

      I think you make a good point: arguing that channel must be non-zero as a matter of *logic* is a stronger Applicant argument than relying on the spec. Especially because when an Applicant mentions the spec, the most common Board reaction is "can't import limitations from the spec."

      OTOH, non-zero doesn't really end the inquiry. The Examiner's position was also (at least implicitly) that the claim covered some very small non-zero values, yet it wasn't clear how close to zero the claim covered. The argument "non-zero as a matter of logic" doesn't address this concern. But the spec does.

      Delete
  2. "The argument "non-zero as a matter of logic" doesn't address this concern. But the spec does."

    This is where my lack of expertise in the subject matter prevents me from carrying on much of a conversation. Let's say, taken alone, the recited number range encompassed small values such as 0.001 microns (and of course smaller values too). However (speculating here) a value as small as 0.001 microns would not actually permit the functions of a channel, so it would not actually be a "channel." In that case, POSITA would have known that a "channel" must necessarily be larger than this in order to function.

    That is, the claim covers as close to zero as would perform the functions of a channel. As for the scope of the claim, then, although POSITA might not the exact numerical value at which they stop infringing, they would know the -function/property- at which they stop infringing. So it seems like the claim still puts the public on notice as to what scope is covered.

    From a different perspective, it seems like this is akin to reading the number range all alone rather than reading the number range as a further limitation of "channel."

    ReplyDelete
    Replies
    1. >claim covers as close to zero as would perform the functions of a channel.

      Gotcha. Read "channel" in context. Not the bare naked definition -- something like "narrow passage" or "narrow cut." What matters is what "channel" means to a POSITA in the manufacture of semiconductor devices.

      >So seems like the claim still puts the public on notice as to what scope is covered.

      Agree. I'm not sure it matters here, but for the record, note that the indefiniteness standard used during prosecution ("amenable to two or more plausible interpretations") is different than the one used in litigation.

      >akin to reading the number range all alone rather than reading
      >the number range as a *further limitation* of "channel."

      Reading a limitation in isolation? Gee, I've never seen *that* happen :-)

      >my lack of expertise in the subject matter prevents me from carrying on much conversation.

      To the contrary! Your additional explanation / reasoning seems cogent to me. I'm glad you contributed.

      Delete