Wednesday, July 10, 2013

Board affirms obviousness after finding that melt index was recognized as result effective variable

Takeaway: The Examiner rejected as obvious various claims to a method of manufacturing foam-paperboard laminates. The Examiner used the routine optimization rationale, alleging that since melt index was a result effective variable for viscosity, the claimed melt index value was obvious. The Applicant argued that the Examiner had used hindsight because the prior art had not recognized melt index as a result effective variable. According to the Applicant, "none of the prior art of record recognized or appreciate the effects melt index have on web speed, as discovered by the inventors of the present invention." The Board affirmed the obviousness rejection after finding that melt index was recognized by the prior art to "affect viscosity, among other blend properties, and thus ability to extrude or process the blend." (Ex parte Swoboda, PTAB, 2012.)

Details:

Ex parte Swoboda
Appeal 2010012538; Appl. No. 11/454,570; Tech. Center 1700
Decided:  May 28, 2012

The application was directed to a method of manufacturing foam-paperboard laminates, such as those used in insulated beverage containers. One of the limitations argued on appeal was "providing a melted LDPE [Low Density Polyethylene] having a melt index of at least about 8.0 g/10 min."

According to the specification, the claimed melt index was unconventional in an extrusion process, "because the resulting coating would be thought to be too soft to provide an article with useful properties." However, the inventors found that using the claimed melt index in combination with other claimed processing parameters allowed an increase in extrusion speed (recited in step c of the claim) and improved the foam quality. 

The Examiner rejected the independent claims as obvious. The Examiner relied on a primary reference, Iioka, for everything except the claimed melt index, for which the Examiner relied on a secondary reference, Roox. The Examiner asserted that "melt index is a result effective variable of viscosity controlling at least ease of extrusion and is therefore subject to routine optimization."

The Applicant appealed and made several arguments. With respect to result effective variables, the Applicant argued that Roox did not recognize melt index as a result effective variable "for viscosity" as asserted by the Examiner. Instead, the Applicant argued that Roox merely taught that melt index should be "sufficiently high to permit dispersion of one component of a resin into another component of a resin." According to the Applicant:
Insofar as the record shows, the only evidence that shows melt index is a recognized, result determinative variable has been gleaned from the Applicant's own specification, which is nothing short of impermissible hindsight. Indeed, none of the prior art of record recognized or appreciate the effects melt index have on web speed, as discovered by the inventors of the present invention.
In the Answer, the Examiner further explained his position on the result effective variable:
Roox teaches that the melt index of the polymer blend may be selected from a large range (.1 to 500 for the blend (col. 2, lines 38-41 )) but does not state the given effect. However, the effect has been articulated; specifically, melt index is a result effective variable of viscosity controlling at least ease of extrusion."
The Board found that melt index was a result-effective variable recognized by Roox:
 The Examiner finds that Roox teaches a range of melt indices that overlaps Appellants’ claimed range (Ans. 4, 9, 13). The Examiner finds that Roox’s focus on the melt index of the blend indicates that the melt index is an important variable for controlling viscosity and thus extrusion. Id. Roox further discloses using higher melt indices to permit dispersion within the blend (col. 3, ll. 12-15). Based on these findings, we agree with the Examiner that Roox recognizes melt index is a result effective variable for controlling the ability of extruded materials to disperse. An improved dispersability of the materials in the blend would affect viscosity, among other blend properties, and thus ability to extrude or process the blend.
(Emphasis added.)
The Board was not persuaded by any of the other arguments made by the Applicant, and so affirmed the rejection.

My two cents: Materials science isn't my area of expertise, but seems to me that the Board went a bit beyond the Examiner's actual findings. I don't see that the Examiner said Roox "focused on the melt index of the blend" – all I saw was that the Examiner relied on Roox's blend melt index to teach the claimed LDPE melt index. I certainly don't see how this "focus" led the Examiner to the conclusion that Roox taught that viscosity was controlled by melt index. All I saw was the Examiner's assertion that viscosity was controlled by melt index.

On the other hand, Roox certainly did teach that melt index affects dispersion. The Applicant even acknowledged this teaching in the Appeal Brief. Plus, Wikipedia says that melt index is inversely proportional to viscosity. Materials science is not my thing, but this suggests to me that "melt index is not a results-effective variable recognized in the prior art" is a losing argument.

Perhaps the Applicant was really making an unexpected result argument:
Indeed, none of the prior art of record recognized or appreciate the effects melt index have on web speed, as discovered by the inventors of the present invention.
(Appeal Brief.)
 If so, the Applicant didn't frame the argument very well – the section title for the argument was "result determinative variable" rather than "unexpected results." And the Applicant didn't carry the evidentiary burden either. The Specification did include various statements that the claimed LDPE properties led to an increase in web speed. The Specification even included some technical explanation of why a POSITA would have found the results unexpected. However, case law also requires a comparison of the claims with the closest prior art. In re Baxter Travenol Labs., 952 F.2d 388, 392 (Fed. Cir. 1991). No such comparison was present in the Specification, nor did the Applicant file a § 1.132 declaration making a comparison.

1 comment:

  1. Karen I was astonished to read that it was in 1991 that the US caselaw started to require a comparison of results between what the claimed subject matter delivers and what the "closest prior art" can deliver.

    It has been that way at the EPO since the very beginning (1978). So can you tell me, did the Baxter case in 1991 simply summarise usefully the established predecessor US law?

    I had thought from my daily dealings with US patent attorneys that the expression "closest prior art" causes them deep anguish, and so should never be uttered in polite society.

    ReplyDelete