Sometimes an applicant gets caught in the grinding wheels of federal bureaucracy which spits their patent application out the other side, or at least that is how it can feel. Today we review application 15/988,262 by Semiconductor Components Industries. The application relates to a method of singulating a plurality of die from a semiconductor substrate. Claim 10 on appeal sets forth a method that utilizes sonic energy with the substrate mounted to a support tape and spindle, where the sonic energy is applied through the spindle to singulate the die at a damage layer formed from irradiation.

10. A method of singulating a plurality of die from a semiconductor substrate, the method comprising:
irradiating a die street with a laser beam at a focal point within the semiconductor substrate at one or more spaced apart locations beneath the surface of the die street to form a damage layer beneath a surface of a die street, the die street connecting a plurality of semiconductor die, the plurality of semiconductor die formed on the semiconductor substrate;
mounting the semiconductor substrate to a support tape and to a spindle;
exposing the semiconductor substrate to sonic energy using a single sonic energy source applied through the spindle; and
singulating the plurality of die at the damage layer using the sonic energy.

The examiner relied on a “obvious to try” rejection. Both the final rejection and the answer laid out a short and simple rejection. The examiner relied on a primary reference (Obata) to show most of the features but was missing the semiconductor aspect. The examiner turned to a secondary reference (Murata) that showed various other wafer division methods. The examiner was also missing aspects related to the sonic energy via the spindle, and so turned to Sharma. To combine each of the two secondary references, the examiner relied on KSR and “obvious to try” reasoning:

The rejection, therefore, was quite clearly based on the obvious to try rationale. However, for some of the PTAB judges on this panel that wanted to affirm the rejection, this rationale was problematic as the Examiner did not actually support the rejection. Specifically, and importantly, the examiner failed to identify, as just one example, a recognized problem or need in the art before concluding that Sharma’s spindled vibrator would have been “obvious to try.” Rather than reverse (or affirm with a new ground of rejection), the PTAB took a short cut that seems dubious at best (internal citations omitted):

With regard to Appellant’s arguments pertaining to “obvious to try” rationale, we view the Examiner’s characterization of the rationale as being “obvious to try”, as harmless error. While we appreciate the dissenting position that the MPEP mandates that the Examiner articulate there had been a recognized problem or need in the art when relying upon this rationale, we believe that the Examiner de facto relied upon and/or articulated that the design need and/or market pressure is for any appropriate ultrasonic agitation. The Examiner’s position that it has been held that choosing from a finite number of identified, predictable solutions (using the vibrator with spindle or without spindle for supplying ultrasonic energy to a wafer), with a reasonable expectation of success, is obvious may also reasonably be viewed as the predictable use of known prior art elements or steps performing the same functions they have been known to perform. In any event, Appellant’s arguments are not persuasive of reversible error in the Examiner’s determination because they fail to account for “the inferences and creative steps that a person of ordinary skill in the art would employ.” The claimed subject matter appears to be no more than the predictable use of known prior art elements or steps performing the same functions they have been known to perform and the Supreme Court has stated such a situation is normally obvious; that is, the combination of familiar elements/steps is likely to be obvious when it does no more than yield predictable results. Accordingly, Appellant has not shown reversible error in the Examiner’s obviousness determination.

It seems the applicant had to do more than account for the inferences and creative steps of the skilled artisan, but also the creative legal analysis of PTAB judges who realized the examiner’s rejection was flawed but still found the claims obvious. In any event, some PTAB judges will go so far as to equate “obvious to try” with a “predictable use of a known element performing the same function” to affirm.