Monday, October 4, 2010

BPAI finds claim reciting structures "constructed in CMOS technology" to be indefinite (Ex parte Sewald)

Takeaway: The BPAI affirmed an indefiniteness rejection for a dependent claim which required two of the structures in the independent claim -- a container and electrodes -- to be "constructed in CMOS technology." The Board found that a POSITA "would not be able to readily ascertain the specific aspects of CMOS technology that are used to construct the container configuration and electrodes and how such technology is specifically implemented."

Details:
Ex parte Sewald
Appeal 200900658, Appl. No. 10/200,705, Tech. Center 1700
Decided March 30, 2009

The independent claim recited:
13. A device for identifying molecules in a carrier liquid, comprising:
a container configuration for holding the carrier liquid;
a tunable radio-frequency oscillator;
an evaluation circuit;
two electrodes disposed in said container configuration and connected to said tunable radio-frequency oscillator and said evaluation circuit.

The subject of the indefiniteness rejection was dependent claim 19:
19. The device according to claim 13, wherein said container configuration and said electrodes are constructed in CMOS technology.

When the indefiniteness rejection was first introduced, the Examiner used the "omitted essential steps" rationale: "[T]he claims do not positively recite any specific method steps. It is unclear as to how the device is fabricated using CMOS technology."

The Applicant responded by introducing a technical definition of CMOS ("a device formed by the combination of a PMOS and with an NMOS") and a discussion of the CMOS fabrication sequence from the textbook "VLSI Process Integration."

The Examiner maintained the indefiniteness rejection in the final Office Action, asserting that:
The metes and bounds of this claim are unclear in that the claim does not positively recite any specific structural limitations for the claimed device. Is the specific device structural configuration comprising a certain metal oxide semiconductor layered structure? The structure which goes to make up the device must be clearly and positively specified. The structure must be organized and correlated in such a manner as to present a complete operative device. 
The Applicant simply repeated the previous argument in the Appeal Brief. The Examiner's Answer noted that the arguments about the meaning of the term CMOS were not based on limitations actually recited in the claim:
Claim 19 does not recite either a NMOS or PMOS transistor element. It is unclear as to how these additional elements are associated with the recited container configuration and the electrodes actually recited in the claim to form an operative device.

The Board adopted the Examiner's position, summarily stating that a POSITA "would not be able to readily ascertain the specific aspects of CMOS technology that are used to construct the container configuration and electrodes and how such technology is specifically implemented."

My two cents:  When I first read this case, I thought: "What's next – indefiniteness rejections for dependent claims which specify 'the network is Ethernet', because the app doesn't explain how to build an Ethernet network?" That is, I thought the rejection was completely off base.

But I don't work much with semiconductors, and it turns out that my analogy to network-is-Ethernet was misplaced. After discussing the case with various commenters on this blog, I've revised my stance. And as a consequence, I revised this post.


I see that "constructed in CMOS" means made using three specific layers (Metal, Oxide, Semiconductor). And I've been persuaded by folks with semiconductor expertise that the claim at issue in Sewald doesn't make sense because a POSITA wouldn't understand how the claimed elements electrode and container were "constructed in" those particular layers.

I'm still interested in this case, but for a different reason: I'm wondering how to get "microelectronic" in there as a limitation. Because from reading the spec, it appears that the invention was really directed a microelectronic implementation. This aspect was described as follows in the spec:
  • [0004] Biosensors, in which the evaluation of electronic and physical quantities are employed as indicators in order to obtain the desired information, also seem to be very promising for this. In terms of process engineering, this may involve utilizing matricial configurations of microelectronically integrated sensors, which are used as an interface between the organic substances and various electronic measurement methods.
  • [0025] Preferably, the container configuration and the electrodes may be produced in CMOS technology. The entire circuit, including the electronic components, can in this way be integrated particularly well on a single chip.
  • [0054] Especially for a microelectronic embodiment of this method, transmission measurement is preferable because the sample containers can have very small dimensions. Moreover, better results may be expected from this type of measurement.

  • [0067] All the components of such an individual cell may be integrated on a chip in CMOS technology. Furthermore, a plurality of such individual cells may be integrated in a single configuration on a chip. The individual outputs 4 of the individual measurement cells are then fed to a multiplexer 22, which connects the individual outputs 4, 4' etc. to an analog/digital converter 5 that, in turn, is connected to a signal processor 6. The circuit components may also be integrated on the chip.
  • [0068] CMOS technology is most suitable for the integration on the chip.
Without some sort of microelectronics limitation, you could easily envision a macro-scale container and electrodes when reading the independent claim. And I think there's value in having a claim which gets across "microelectronics" or "integrated circuit," in order to orient the Examiner, judge, jury, and/or licensee.

So how do you claim the microelectronic aspect?

What about this independent claim:
A microelectronic sensor comprising: [elements in independent claim].
Microelectronic appears only in the preamble, so it may not get patentable weight – but it might be enough to orient the reader.

What about this dependent claim:
...wherein the container configuration is a microelectronic structure.
Would that be considered a process limitation, and thus given no patentable weight?

Would this language be treated any differently:
13. A device for identifying molecules in a carrier liquid, comprising:
a microelectronic container configuration for holding the carrier liquid; ...

19 comments:

  1. Former Examiner, Current ProsecutorOctober 4, 2010 at 11:38 PM

    I have a whole book on this- Karen agree with you entirely.

    ReplyDelete
  2. Former Examiner, Current Prosecutor said:
    >I have a whole book on this- Karen agree
    >with you entirely.

    Have you been successful in convincing Examiners to drop the rejection?

    BTW, love your "Former Examiner, Current Prosecutor" handle. Anonymous yet descriptive.

    ReplyDelete
  3. I have to disagree with you on this one... I think the BPAI got this one right. A PHOSITA would find "a container, an oscillator and electrodes -- to be constructed in CMOS technology" confusing. The "MOS" in CMOS means "metal-oxide-semiconductor." It isn't clear how a container or electrodes are made using a metal layer, an oxide layer, and a semiconductor layer. Moreover, CMOS specifically refers to a combination of NMOS and PMOS devices to form digital logic. An oscillator, container, and an electrode are not ordinarily thought of as being made using NMOS and PMOS devices. Ring oscillators can be made using NMOS and PMOS devices, but it isn't really clear that that is what was intended to be claimed because other oscillators can be made using capacitors -- potentially made using metal-oxide-semiconductor. In sum -- it just isn't clear what is being claimed. Also, saying that something is constructed in a technology is strange. So, the claim just sounds like nonsense.

    ReplyDelete
  4. >isn't clear how a container or electrodes are
    >made using a metal layer, an oxide layer, and
    >a semiconductor layer.

    I don't have experience with semiconductors,
    and now that I read your comment, it looks like I just didn't appreciate how specific the term CMOS is. And more importantly, how strongly it connotes structure.

    Your explanation makes sense, so I'll defer to your judgment on this one.

    However, I do think there's value here in having a claim which gets across "microelectronics" or "integrated circuit", if only to orient the Examiner, judge, or jury. Otherwise, you might envision a macro-scale container and electrodes.

    So what about a claim that wasn't as narrow -- and didn't connote as much structure -- as CMOS. How could you get "microelectronic" in there? It seems the Examiner could easily call it a process limitation and ignore it. In fact, it seems like the "CMOS technology" limitation could be treated that way.

    ReplyDelete
  5. >How could you get "microelectronic" in there? >It seems the Examiner could easily call it a >process limitation and ignore it. In fact, it >seems like the "CMOS technology" limitation >could be treated that way.

    I would agree that this limitation won't add to the point-of-novelty. However, assuming the independent claim issues, this type of process limitation dependent claim could prove a useful guide to a judge for claim interpretation purposes.

    ReplyDelete
  6. Anonymous @ October 5, 2010 12:10 AM continues...

    Caveat: I have not read the opinion or patent.

    I think a claim that said something like "the system of claim 1 further comprising an integrated circuit including a container, an oscillator, and electrodes" might be okay. However, "container" is not a commonly known circuit component, and so if the "container" has more of a mechanical function, then I would replace "integrated circuit" in the limitation above with "a micro-electro-mechanical structure" or something like that.

    ReplyDelete
  7. >this type of process limitation dependent
    >claim could prove a useful guide to a judge
    >for claim interpretation purposes.

    Yeah, that's what I was thinking.

    Although ... what about claim differentiation? Once you have a dependent claim that says "microelectronic", doesn't differentiation suggest the *independent* claim is something other than micro, e.g, macro?

    Maybe there's something to be said for an *independent* claim that really reads on the preferred embodiment. That might make claim scope more clear to the Examiner, the infringer, the judge, *and* the licensee.

    ReplyDelete
  8. >further comprising an integrated circuit
    >including a container, ...
    >However, "container" is not a commonly known
    >circuit component, and so if the "container"
    >has more of a mechanical function, then >[maybe] "a micro-electro-mechanical structure"

    The container in Sewald is just a space in the chip that holds the solution being tested. So you're right, the container is not itself a circuit. Though the electrodes pass current through the liquid in the container -- so doesn't that make the container part of the circuit?

    In fact, although this is not my area of expertise, the invention in Sewald seems to be in the micro-electro-mechanical(MEMS) arena. But "microelectronics" was the term used, not MEMS.

    The Sewald app claims priority to German app, so maybe MEMS got lot in the tranlslation.

    ReplyDelete
  9. Upon reading the original application, it is apparent that this whole CMOS aspect has become an unwelcome distraction in the case. To elaborate, page 13 of the spec points out: "Especially for a microelectronic embodiment of this method, transmission measurement is preferable because the sample containers can have very small dimensions" and page 14 states: "Owing to the ever-improving radio-frequency properties of new process generations, this configuration could be produced in CMOS technology."
    A POSITA will zoom in on 2 aspects of these statements (no, not the CMOS aspect!!!): 1)dimensional aspect, and 2)RF frequency aspect.
    An implementation on a chip, whether in CMOS, or better yet on an alumina/ceramic substrate (supports higher RF frequencies)provides 2 benefits: 1) dimension of the container (the DUT) is small therefore helping the transmission measurement, and 2) allows higher RF frequency implementations that may not be practical with discrete components (the mixer (as well as the S-parameter measurements)is a dead give-away that it is preferred to use higher frequency transmission testing coupled with down-conversion so as to perform low frequency measurements.)
    The claim could have been worded to indicate that the container is formed as part of a chip with the container dimensions selected to accommodate a desired transmission parameter (and/or desired RF frequency range).
    Furthermore, rather than needlessly digging in one's heels, I would have added some dependent claims that addressed two other interesting aspects of the invention: 1) detecting residual "unwanted reaction products" and 2) parallel testing over multiple containers along with a TDM measurement using the multiplexer (especially with these multiple containers arranged in a "matricial configuration" (matrix).

    ReplyDelete
  10. >this whole CMOS aspect has become an
    >unwelcome distraction

    One man's distraction is another man's area of interest :-)

    I was interested in Sewald because I thought the indefiniteness rejection was completely misplaced. That's because I saw it as analogous to rejecting "the network is Ethernet" for indefiniteness. Your comment, and others in this thread, have convinced me otherwise.

    What interests me now is how the claim could be better drafted to capture the microelectronic aspect -- now that I see "implemented in CMOS" was a bad way to do it.

    >could have been worded to indicate that the '
    >container is formed as part of a chip with
    >the container dimensions selected to
    >accommodate a desired transmission parameter

    Now *that's* what I'm talking about. Assuming the drafter was right about the general elements of novelty, how to better capture the microelectronic aspect.

    ReplyDelete
  11. This entire thread is based on Karen not understanding the rejection. She probably still doesn't get it really.

    In any event Karen, I doubt if this claim could have been better written so as to encompass the same thing. What they actually meant in the application is that all of those elements are incorporated into a device that also includes CMOS somewhere in it. Or, that those elements were actually made up of CMOS components. The second is less likely because it is nonsense.

    Either way, what needed to have been done was for drafting to have been better in the first place making it clear what was meant by "implemented in CMOS".

    Like the man said above, this CMOS recitation is an "unwanted distraction" and it is also indicative, when people like yourself note that it may be of interest, of a bad prosecutor. Or in your case a software prosecutor.

    ReplyDelete
  12. I think merely miniaturizing a circuit that is already known would (rightfully) draw an obviousness type of rejection, unless the miniaturization provided some unique features that cannot be achieved via the larger implementation. There should be some indication in a claim as to what is the novelty aspect achieved via a "microelectronic" implementation (e.g. measuring nano-ampere currents, or operating in GHz RF regions).

    ReplyDelete
  13. >I think merely miniaturizing a circuit that is
    >already known would (rightfully) draw an
    >obviousness type of rejection

    Agreed. I've no reason to think that the microelectronic aspect was patentable in and of itself. Instead, I'm suggesting that having at least one claim that included this aspect might be useful in orienting the Examiner, licensee, judge, and jury as to what the claim was really about.

    ReplyDelete
  14. I think merely "orienting" an Examiner is not enough - the claim(s) should contain some specific language as to the novelty associated with the smaller size. The spec does not appear to explicitly support such novelty. One may attempt to use a PHOSITA argument to introduce some specificity, but then, as can be appreciated, this approach runs the risk of enablement problems.

    ReplyDelete
  15. >claim(s) should contain some specific language
    >as to the novelty associated with the smaller
    >size.

    Nah, not what I'm talking about. The claim is about an RF circuit for evaluating the solution in a container. The novelty has something to do with that. I've no reason to think the novelty is related to the microelectronic aspect.

    All I'm saying is that you couldn't tell, from reading the claim, that it had anything to do with microelectronics. And I'm suggesting that when the commercial embodiment is clearly going to use microelectronics [the assignee is Infineon, a chip maker], there is some value to having a claim that refers to microelectronics.

    Maybe the Examiner will ignore this language, especially if it's in the preamble. But I can't help but think that if the jury saw a claim that used the word "microelectronics", they'd have an easier time finding that the accused chip infringed than if the claim merely said electrodes and container.

    ReplyDelete
  16. My apologies for inappropriately using the phrase "measuring nano-ampere currents." Seeing that this is an RF application, the more appropriate parameter would be "measuring power" - specifically in dBm (or dB form, depending on the type of measurement). Smaller structures accommodate higher RF frequencies, especially if these structures are designed as waveguide structures.

    P.S. I'm stopping further comments on this topic because it seems like I have reached the end of the line in terms of adding any significant value to the discussion :-)

    ReplyDelete
  17. I haven't read all the comments, only the lead story.

    The Board screwed up. If the device is constructed in CMOS, any CMOS (at the broadest reasonable construction that term could be given by POSITA), it falls within the claim. If the device constructed out of high fructose corn syrup, it doesn't.

    Section 112 ¶ 2 is about claim definiteness, can you answer the "yes/no" question. Breadth is not indefiniteness. "How to" is ¶ 1 for the specification.

    Further, CMOS is about as established a term of art as there is. If I can't rely on a term like "CMOS" to mean the stuff that comes out of a Silicon Valley fab, then good grief what word has any meaning?

    ReplyDelete
  18. David Boundy said:
    >If the device is constructed in CMOS, any CMOS,
    >it falls within the claim.
    >"How to" is ¶ 1 for the specification.

    I'm glad you brought this up. My very first reaction to this opinion was that the rejection should have been enablement, not indefiniteness.

    One of the Examiner's first explanations for the rejection was: "It is unclear as to how the device is fabricated using CMOS technology." Sounds to me like he's taking issue with enablement.

    A later statement from the Examiner sounded more like indefiniteness: "Is the specific device structural configuration comprising a certain metal oxide semiconductor layered structure?"

    OK, I do see that sort of question from Examiners in indefiniteness rejections. My reaction: a claim doesn't have to be as clear as the Examiner wants it to be, only clear enough for a POSITA. (Paraphrasing MPEP 2173.02.)

    But after my original post, some folks with semiconductor experience told me that it wasn't clear "how a container or electrodes are made using a metal layer, an oxide layer, and a semiconductor layer."

    I don't have semiconductor experience, but I have experience with comm. networks, so one of my friends gave me this analogy: the claim in Sewald is like "wherein the Ethernet network is constructed in copper". Indefinite because a POSITA can't tell which parts are made of copper.

    Anyway, so I deferred to the commenters with semiconductor experience, and revised my stance on the case -- even updating the post to say I agreed with the Board.

    So while I can certainly see why you view indefiniteness as the wrong sort of rejection here, I think I'm still on the side of Affirm Indefiniteness.

    >CMOS is about as established a term of art as
    >there is

    No doubt about that. I'm pretty sure that the Examiner knew what CMOS was, which is why the technical definition submitted by the Applicants didn't help. The issue wasn't the meaning of CMOS, but what a POSITA would think about an electrode constructed of CMOS. The Examiner felt, as did commenters to this blog, that a POSITA would be confused by the phrase.

    ReplyDelete
  19. "What about this independent claim:

    A microelectronic sensor comprising: [elements in independent claim].
    Microelectronic appears only in the preamble, so it may not get patentable weight – but it might be enough to orient the reader.

    What about this dependent claim:

    ...wherein the container configuration is a microelectronic structure.
    Would that be considered a process limitation, and thus given no patentable weight?

    Would this language be treated any differently:

    13. A device for identifying molecules in a carrier liquid, comprising:
    a microelectronic container configuration for holding the carrier liquid; ..."

    The first one is fine and is what I thought of when you posed the question above. That's usually how it is done. You can also make reference to the preamble in the body.

    Anyway, your second and third suggestions appear to be nonsense. A "container configuration" is never a "microelectronic structure" I doubt. That is, "container configuration" is not a genus under which "microelectronic structure" fits.

    As to your third suggestion, container configurations are not microelectronic. Ever. So far as I know.

    What you're basically trying to do is put the cart before the horse in a manner of speaking. This "it can be implemented in CMOS, or microelectronic emobidments" is not specific enough to tell one of ordinary skill wtf they're talking about. Then comes 112 2nd.

    "Agreed. I've no reason to think that the microelectronic aspect was patentable in and of itself. Instead, I'm suggesting that having at least one claim that included this aspect might be useful in orienting the Examiner, licensee, judge, and jury as to what the claim was really about. "

    It would help if the claim were actually about that in the first place. You're trying to fit a square peg into a round hole trying to make it "actually about" that at all.

    Boundy, as re tarded as ever I see.

    A device as they have formed here cannot be formed "in CMOS". If you knew what CMOS was then you'd know why.

    The claim as they have written it here is like them writing this: a lightbulb and a clock contructed in bicycle technology. How the f are you supposed to know wtf they're talking about? Are they talking about a lightbulb and a clock attached to a bicycle? All they claimed was nonsense gibberish.

    "a claim doesn't have to be as clear as the Examiner wants it to be, only clear enough for a POSITA."

    lulzerskates, tell it to the hand and stop talking gibberish in your claims.

    "All I'm saying is that you couldn't tell, from reading the claim, that it had anything to do with microelectronics. And I'm suggesting that when the commercial embodiment is clearly going to use microelectronics [the assignee is Infineon, a chip maker], there is some value to having a claim that refers to microelectronics.
    "

    Yeah and guess what? You might want to tell us what that embodiment is, specifically, and structurally. Otherwise we'll reject ur sht. Every, single, time.

    ReplyDelete