T 1044/23: application of G 1/23 to un-enabled products

A recent decision from the EPO Boards of Appeal, T 1044/23, has followed the finding in G 1/23 that a product placed on the market is prior art, regardless of whether the product can be reproduced. The decision, however, makes G 1/23 potentially rather limited in terms of how far it can be applied. 

The patent in question, EP3161066, claimed a polyethylene composition in terms of various parameters. Claim 1 as granted read as follows: 

Polyethylene composition having the following features:

1. density of from 0.950 to 0.970 g/cm³, determined according to ISO 1183 at 23°C;

2. MIE from 1 to 30 g/10 min;

3. ratio MIF/MIE from 15 to 30, in particular from 20 to 29 or from 22 to 29, where MIF is the melt flow index at 190°C with a load of 21.60 kg, and MIE is the melt flow index at 190°C with a load of 2.16 kg, both determined according to ISO 1133;

4. ER values from 0.40 to 0.52

The patent was opposed on the grounds that the claimed invention covered two products that were available before the filing date, both being high density polyethylene (HDPE) resins available from the Dow Chemical Company (the opponent). Evidence for the products being available was provided by product datasheets and dated invoices. Certificates of analysis of the products were also provided to show the relevant parameters, which are summarised in the table below.

	Density	MIE	MIF/MIE	ER
Claim 1	0.950-0.970 g/cm3	1-30 g/10 min	15-30	0.40-0.52
D1 DMDA-8904 NT7	0.9516 g/cm3	4.11 g/10 min	22.8	0.50
D9 DMDA-8907 NT7	0.9516 g/cm3	7.08 g/10 min	22.3	0.49
Aux 1		1-25 g/10 min
Aux 2	0.955-0.970 g/cm3

The opposition division, following G 1/92 and the later decision T 1833/14, found that the two products were not considered prior art under Article 54(1) EPC “because their manufacture was not enabled” (point 19 of the decision). They also found the patent to be inventive over two other cited documents due to these documents not disclosing all the parameters in claim 1, in particular the ER value (relating to rheological polydispersity).

The opponent appealed the decision and argued that, according to G 1/23 (which issued after the opposition decision), the requirement for the product to be reproducible was no longer a valid criterion to exclude a commercial product from the prior art. Since both D1 and D9 resins were publicly available before the filing date of the patent, and the evidence showed that both fell within the scope of claim 1, the patent was not novel in view of either resin.

The Board referred to point 1 of the decision in G 1/23, which states:

“A product put on the market before the date of filing of a European patent application cannot be excluded from the state of the art within the meaning of Article 54(2) EPC for the sole reason that its composition or internal structure could not be analysed and reproduced by the skilled person before that date.”

The Board found that, according to the evidence provided, there was no reasonable doubt that the resins were supplied to customers before the filing date of the patent. The resins were therefore considered to be within the state of the art according to Article 54(2) EPC. There was no dispute that the resins did fall within the scope of claim 1 as granted, so the Board found that claim 1 lacked novelty. The first auxiliary request, which limited the MIE range to 1-25, made no difference and therefore also lacked novelty.

The second auxiliary request limited the density to the range 0.955-0.970 g/cm³, which was higher than either prior art resin and therefore novel. The question was then whether the commercially available resins could be considered to be the closest prior art, given that methods to prepare them were not publicly available. G 1/23 only provided guidance on that point to the extent that a commercial product could be selected as the closest prior art but depended on the specific circumstances because the relevant technical teaching that a skilled person would take from the product would always be case specific. The Board referred to comments in G 1/23 about it being obvious to add lemon juice to Coca-Cola for a less sweet taste, even though the recipe was secret. In the present case, even though the resins were not fully reproducible, there was no reason why they could not be a realistic starting point for further developments.

Starting from either resin, the problem to be solved was, in the Board's view, at best the provision of an alternative composition, given that the patent showed that increasing the density tended to result in poorer properties. In light of this, no particular pointer in the prior art would be required. Although the appellant/opponent argued that the skilled person would know how to adjust the density of the resins while maintaining other properties within the claimed ranges, the Board considered that no evidence was provided of what the skilled person would have done to adjust the parameter. It followed that it was not credible that the skilled person would know how to modify the resins to get within the scope of claim 1. The claim according to the second auxiliary request was therefore found to be inventive.

The Board also found that claim 1 of the second auxiliary request was inventive over another of the cited document (D6), which disclosed all parameters except for the ‘ER’ value within the claimed range. Again, the objective problem was considered to be the provision of an alternative composition, but the Board concluded that it was not obvious how the skilled person would obtain a composition within the claimed range starting from the resins disclosed in D6.

The end result of all this appears to be that the scope of G 1/23 for using un-enabled prior art products is going to be limited if the patent claim has the slightest point of novelty, given that it will need to be shown how the skilled person would be able to change the product without knowing how to make it. Even if it is clear that the change would result in a poorer product, this can still result in a finding of inventive step.