When a Patent Sets a Numerical Range but Never Says How to Measure It, the Claims May Not Survive

COCONA, INC. v. VF OUTDOOR, LLC

Authored by: Jeremy J. Gustrowsky

The Federal Circuit has affirmed a ruling that claims covering a waterproof, breathable fabric membrane are invalid because the patent never told readers how to perform the one test that determines whether a product falls within the claims. The decision is a reminder that reciting a precise-sounding number is not enough if the measurement behind that number can be taken several different ways.

Cocona, Inc. owns U.S. Patent No. 8,945,287, titled “Active Particle-Enhanced Membrane and Methods for Making and Using the Same.” The patent addresses a familiar problem with rain gear: material that keeps rain out also tends to trap sweat in. The solution described in the patent is to embed active particles, such as activated carbon, in a waterproof membrane so that water vapor can move through more easily while liquid water is still repelled. Each asserted claim required the membrane to have “a moisture vapor transmission rate from about 600 g/m2/day to about 11000 g/m2/day.” Moisture vapor transmission rate, or MVTR, measures how quickly moisture passes through a material.

Cocona sued VF Outdoor, LLC (doing business as The North Face) and Columbia Sportswear Company in Colorado federal court. VF Outdoor also challenged a group of claims in an inter partes review, and the Patent Trial and Appeal Board found most of them obvious, a decision the Federal Circuit previously affirmed. The district court case then proceeded on claims 1, 38, and 39. During claim construction, Cocona asked the court to read a detailed testing protocol into the MVTR requirement, including the ASTM E96 cup water test, 40 degrees Centigrade, 20 percent relative humidity, a 24 hour period, and instructions to use the upright method for low moisture contact and the inverted method for high moisture contact. The district court declined and instead held the claims indefinite.

The Federal Circuit agreed. Starting with the claim language, the court observed that the claims say nothing at all about how to measure MVTR, and none of the detail Cocona wanted to import appears anywhere in them. The specification was no better. It lists three measurement approaches (the Mocon method, the upright cup method, and the inverted cup method), says only that upright is for “low moisture contact” and inverted is for “high moisture contact,” and never explains how to tell low from high. The temperature and humidity figures Cocona pointed to appear in permissive language about what “may be measured,” not as requirements. The referenced ASTM E96-94 standard itself allows a range of testing temperatures and two different humidity settings, and it warns that a value obtained under one set of conditions may not reflect the value under another.

What sealed the outcome was the patent’s own data. A table in the patent reports a control sample with an MVTR of 470 g/m2/day using the upright method, which is outside the claimed range, but 6,356 g/m2/day using the inverted method, which is inside it. The provisional application incorporated by reference, U.S. Provisional Application No. 60/799,426, contained an even more striking example: a sample containing activated carbon measured at 1,394 g/m2/day inverted (inside the range) and 223 g/m2/day upright (outside the range). Drawing on earlier decisions involving measurements of slope and molecular weight, the court explained that when multiple methods yield different results and the patent gives no guidance on which to use, the method chosen could decide whether a product infringes. That uncertainty is what the definiteness requirement forbids.

The court also rejected Cocona’s remaining arguments. Expert testimony from the defendants’ side supported the conclusion, and the district court permissibly found Cocona’s expert declaration conclusory for failing to explain how a skilled reader would turn permissive language into a mandatory protocol. Prior decisions Cocona cited were distinguishable, since those patents either identified a single established measurement method or gave enough direction to choose among several. And while the district court used phrases like “mandatory guidelines” and “imperative language,” the appeals court found it had stated and applied the correct legal test.

One caveat deserves emphasis. The Federal Circuit expressly declined to hold that disclosing more than one measurement method automatically makes a claim indefinite. The holding is tied to this record, where the combined intrinsic and extrinsic evidence clearly and convincingly showed that a skilled artisan could not determine the scope of the claims with reasonable certainty. The practical lesson for drafters is straightforward: if a claim limitation depends on a measured value, the specification should state the test method and the conditions in terms that read as requirements rather than suggestions.