Friday, October 6, 2023

Infringing Intermediate or de minimis Use?

Deeproot Green Infrastructure, LLC v GreenBlue Urban North America Inc 2023 FCA 185 Mactavish JA: Gleason, Woods JJA affg 2022 FC 709 McDonald J

            2,552,348 / 2,829,599 / Integrated Tree Root and Storm Water System

DeepRoot’s 348 and 599 patents relate to a landscaping system to promote healthy urban trees using a subsurface structural cell system that supports the hardscape (eg sidewalk and paving), enables stormwater retention and filtration as well as allowing tree roots to grow in uncompacted soil. In DeepRoot v GreenBlue 2021 FC 501 (see here) McDonald J held that DeepRoot’s patents were infringed by GreenBlue’s RootSpace product. She consequently enjoined GreenBlue from infringing. After this decision, GreenBlue revised its product by adding the “Airform” component in an attempt to design around the patents. In response, DeepRoot brought a contempt motion against GreenBlue. McDonald J dismissed the motion in a decision that turned entirely on the facts, on the basis that DeepRoot had not proven beyond a reasonable doubt that GleenBlue’s revised system was infringing: 2022 FC 709 discussed here. The FCA has now affirmed, generally on the basis that McDonald J’s decision turned on the facts and DeepRoot had not established any palpable and overriding error. Some interesting points are nonetheless raised tangentially. This post discusses the issue of non-functional design-around and the question of the infringing intermediate. A subsequent post will discuss the import / export question, which was not really addressed at trial.

The GreenBlue RootSpace structural cell that was found to infringe is shown here [10]:


The claims require that “at least approximately 85% of the volume” in the cells in the structural cell system be available to be filled with soil. This was held to be an essential element, and was construed as meaning at least 84.5% [15]. After the initial infringement determination, GreenBlue began selling the RootSpace AirForm system as a design-around. It is the same as the infringing system except for the addition of the “AirForm Insert” which reduces the space available to be filled with soil. The AirForm Insert is shown on the right:


GreenBlue would be in contempt if the RootSpace AirForm system infringes. There were two main questions on appeal. One was whether DeepRoot had established beyond a reasonable doubt that the cells in the RootSpace AirForm system have available volume space of at least 84.5% [51]. McDonald J found on the facts that the available volume was less than 84.5%, and this was affirmed on the basis that she had not made any palpable and overriding error [92].

The other issue was whether it is appropriate to include the volume of the AirForm Insert in the calculation of the available volume in the cells in the RootSpace AirForm system [51]. (If not, the new system would clearly infringe.) The way it was argued, this turned on whether the AirForm Insert formed part of the structural cell itself [52]. On the facts, this was unclear and therefore DeepRoot had not carried it’s burden of establishing infringement beyond a reasonable doubt [62]. This factual finding was not tainted by a palpable and overriding error [64].

Structural element

The tangential issue is whether it is really necessary that the AirForm Insert is structural. I don’t see why it should be. The 85% requirement was express in the claim and was an essential element. This implies it is important. It seems obvious enough that the more dirt volume the better for the tree roots. So perhaps the prior art had systems where only 80% of the volume was available, and the breakthrough lay in devising a system with 85% volume. In that case, so long as the RootSpace AirForm system was in fact used with the AirForm insert in place, the essence of the invention would not be taken, and I don’t see why it should be considered infringing, whether or not the AirForm insert is structural. More broadly, since the requirement that “at least approximately 85% of the volume can be filled with soil” is essential, I don’t see why a system should be considered infringing so long as that requirement is not met, regardless of reason that is it not met.

I will note that there is a claim construction point here. This argument applies directly to Claim 1 of the 348 Patent requires that “at least approximately 85% of the volume can be filled with soil.” If the system is used with the AirForm insert, that is not true, and I don’t see why it should matter whether the insert is structural. Claim 1 of the 599 Patent requires a base and periphery support members “said support members being sized and arranged so that at least approximately eighty five percent of a volume defined by outer edge of said cell is a void space.” This might possibly be construed as requiring that “void” is defined solely by structural members. However, this does not seem to have been the basis for the argument, since the structural argument applied to both the 248 and 599 patents [52].

On a related point, GreenBlue’s witnesses at the contempt hearing “explained that the trial judgment drew ‘a line in the sand’, allowing the company to know the parameters that it needed to follow in designing a new product that had less soil volume than that of the cells in its old RootSpace cell system” [21]. The obvious rejoinder is that it is the claims of the patents themselves that drew the line in the sand. It is clear that GreenBlue designed AirForm for the purpose of avoiding infringement [22], but as Mactavish JA pointed out, this is entirely permissible: “One can deliberately attempt to avoid infringing a patent of which one has knowledge by “designing around” the patent. Whether or not that effort succeeds will depend on the construction of the claims of the patent at issue and not on one’s intention” [42].

With all that said, I don’t see Mactavish JA’s decision as endorsing the view that the Airform Insert had to be structural for the system to be non-infringing. As I read it, she was simply addressing the issues the way the case was argued. She did say that “In assessing whether the space taken up by the AirForm Insert should be included in calculating the volume available in GreenBlue’s RootSpace AirForm system, DeepRoot had to establish that the AirForm Insert did not form part of the structural cell itself” [52]. I don’t read this as a holding that this is required as a matter of law. Read in conjunction with the following sentence—”If the AirForm Insert did not constitute part of the structure of the cell, DeepRoot contends that it could be ignored in calculating the available volume in the RootSpace AirForm cells” (my emphasis), I read this as simply setting out DeepRoot’s argument, not as endorsing it.

Infringing intermediate

DeepRoot also argued that “a RootSpace cell could be made as an ‘infringing intermediate’ if an old RootSpace cell was assembled before adding the AirForm Insert to complete the new RootSpace AirForm cell” [98]. This was dismissed on the facts, on the basis that there was no evidence that RootSpace induced infringement by instructing customers to install the system without the inserts (indeed, the evidence was to the contrary).

The more interesting issue is DeepRoot’s argument that “the creation of infringing RootSpace structural cells during the assembly of GreenBlue’s RootSpace AirForm structural cell systems constitutes direct infringement . . . [e]ven if the structural cell was subsequently rendered non-infringing by installation of the AirForm Insert” [103]. This was dismissed on the basis that if GreenBlue’s assembly instructions were followed, not even a single structural cell would be created during assembly, as the AirForm Insert is inserted before the final structural panel [120]. Any evidence of a different assembly sequence did not involve GreenBlue’s participation [116].

While the argument failed on the facts, it raises an interest issue of de minimis use. Citing Pfizer v Apotex 2009 FC 671 [48], DeepRoot argued that “a party is liable for patent infringement if any intermediates made during the process to make its end product are found to infringe the patent in question” [103]. However, as pointed out in Pfizer-Atorvastatin 2007 FC 898 [91] this rule applies when the role of the intermediates in the manufacturing process is “not insignificant or incidental.” Intermediates are typically important because they are part of a more efficient process. Incidental production of intermediate might not be considered infringing as being de minimis. This view is supported by Bayer v Fresenius Kabi 2016 FC 581. DeepRoot cited Bayer v Fresenius Kabi [164], which states “infringement occurs where a patented substance is produced at an intermediate stage, even if the intermediate is not in the final drug product,” in support of its argument. But Brown J went on to say that even if the patented compound was present in the manufacturing process, that use would be insufficient to establish infringement by importation [168] because it was it was only transiently present, and “trivial and merely incidental” in the manufacture of the product sold into Canada [171], [182]. In my view, both cases relied on by DeepRoot support the view that occasional inadvertent creation of an infringing cell during the assembly process—eg if a worker who was new on the job mistakenly assembled the full cell before inserting the AirForm Insert—would be considered de minimis and therefore non-infringing.

No comments:

Post a Comment