T 1982/18 (Fabric conditioner containing a hydrophobic agent / UNILEVER) of 2.3.2021

Summary of Facts and Submissions

I. The appeal of the opponent (the appellant) is against the decision of the opposition division to reject the opposition against European patent no. 2 646 533, with independent claims 1 and 12 reading as follows:

"1. An aqueous fabric conditioner composition comprising

(a) from 2 to 9 wt% of a fabric softening active, by weight of the total composition, wherein the fabric softening active is an ester-linked quaternary ammonium compound having fatty acid chains comprising from 20 to 35 wt% of saturated C18 chains and from 20 to 35 wt% of monounsaturated C18 chains, by weight of total fatty acid chains;

(b) a fatty alcohol; and

(c) from 0.05 to 1.0 wt% by weight of the total composition of a hydrophobic agent having a ClogP of from 4 to 9, which is not a fatty alcohol,

wherein the aqueous fabric conditioner composition has a stable viscosity of greater than 50 cps, preferably from 55 to 200 cps as measured on a cup and bob viscometer; the viscosity being continuously measured under shear at 106s**(-1) for 60 seconds, at 25°C."

"12. Use of a hydrophobic agent in an aqueous fabric conditioner composition to improve viscostability wherein the hydrophobic agent has a ClogP of from 4 to 9, preferably from 4 to 7, most preferably from 5 to 7 wherein the fabric conditioner composition is defined by any one of claims 1 to 11."

II. With its statement of grounds the appellant filed new documents labeled D21-D25 and raised objections under articles 100(a) and (b) EPC.

III. With its reply the patent proprietor (the respondent) defended the patent as granted and filed amended sets of claims as auxiliary requests 1 to 17 as well as new documents labeled D26 and D27. Moreover, it requested that documents D21-D25 not be admitted into the proceedings.

Claim 1 of auxiliary request 1 differs from claim 1 as granted by the following amendments put in evidence by the board:

"...(b) a fatty alcohol; [deleted: and]

(c) from 0.05 to 1.0 wt% by weight of the total composition of a hydrophobic agent having a ClogP of from 4 to 9, which is not a fatty alcohol; and[deleted: ,]

(d) optionally, a non-ionic alkoxylated material having an HLB value of from 8 to 18;..."

Claim 1 of auxiliary request 2 differs from that of auxiliary request 1 by the following amendment put in evidence by the board:

"...(d) optionally, 0.01 to 0.5 wt% of a non-ionic alkoxylated material having an HLB value of from 8 to 18;..."

The wording of claims 1 and 12 of auxiliary request 3 differs from those as granted in that the ClogP value of the hydrophobic agent is restricted to from 4 to 7.

Dependent claims 2 to 11 of this request refer to particular embodiments of the claimed fabric conditioner composition.

IV. Following documents are relevant for the present decision:

D2: WO 2012/072370 A1

D2b: WO 2012/052349 A1

D3: WO 00/004118 A1

D4: WO 00/06690 A1

D5: US 6 664 223 B2

D6: EP 0 845 523 A2

D7: WO 2010/079100 A1

D8: EP 2 053 119 A1

D11: WO 2011/149475 A1

D12: Bailey's Industrial Oil and Fat Products, 6th Ed., 2005, pages 161-164

D13: Monti Stein's Technical Report, 25 January 2018

D14: Unilever's statement of grounds of appeal of 19 October 2017 in case T 1900/17

D18: "What is HLB? How is it applied to formulate emulsions?" Answer ID 3277, updated 10/18/2017 by Dow Answer Center, The Dow Chemical Company

D19: Surfactants datasheet by BASF

D21: Experimental Report

D22: Glyceryl tricaprylate data sheet

D23: Ultra fabric softener dispersion using 15% solids Stepantex VT-90, by Stepan Company, 2010

D24: Stepan China Surfactant Catalog, 2010

D25: Statement of Marc Declercq dated 2 October 2018

D26: RSC Paperbacks, chemical formulation by Tony Hargreaves, 2003, pages 79-81

D27: Surfactants Classified by HLB Numbers by Merck KGaA, 2019.

V. In response to the board's communication the appellant filed decision T 1227/13.

VI. During the oral proceedings held on 2 March 2021 the parties discussed in particular novelty and inventive step of the main request and inventive step of auxiliary request 3. As regards sufficiency of disclosure the appellant referred to its written submissions.

VII. The final requests of the parties were as follows:

The appellant requested that the decision under appeal be set aside and that the patent be revoked.

The respondent requested that the appeal be dismissed (Main Request), in the auxiliary, that the patent be upheld on the basis of one of auxiliary requests 1 to 17 all filed with letter dated 19 February 2019.

Reasons for the Decision

1. Admissibility of documents D21-D27

1.1 It is apparent that documents D21 to D24 have been filed as a response to the arguments in the decision under appeal (Section 17) regarding inventive step, whilst D25 similarly has been filed with respect to the discussion regarding the meaning of the term "hydrophobic agent" in claim 1 or 12.

The Board thus decided under Article 12(4) RPBA 2007 to admit all these documents into the proceedings.

1.2 D26 and D27, having similarly been filed by the respondent with regard to the discussion about the term "hydrophobic", they were also admitted into the proceedings under Article 12(4) RPBA 2007.

2. Sufficiency of disclosure of the invention

The appellant during the oral proceedings before the board relied on its written submissions in which it raised three different objections.

2.1 First, it argued that the patent was not sufficiently disclosed because of the ambiguity surrounding the calculation of the ClogP parameter of the hydrophobic agent, the patent disclosing (paragraph [0115]) only a program (CLOGP) for measuring the ClogP of perfume materials but not one for measuring the ClogP of hydrophobic agents such as those of the patent. The use of the program CLOGP would moreover not enable the skilled person to identify with certainty those hydrophobic agents suitable for achieving the promise of the invention, namely an increased viscostability, since as shown in D13 and D14 the determination of the ClogP value by using other types of software or updated versions thereof may lead to different results, so that a given hydrophobic agent may or not fall under the scope of claim 1, depending on the used software.

2.1.1 The Board does not accept this argument and agrees in this respect with the decision under appeal that the description of the patent clearly teaches to use the program CLOGP, not only for determining the ClogP of perfumes, but also the ClogP of the hydrophobic agents defined in claims 1 and 12 at issue. Paragraph [0115] in this respect reads: "As used herein, the term "ClogP" means ... ClogP values can be readily calculated from a program called "CLOGP" available from Daylight Chemical Information System Inc., Irvine Calif., USA".

It is not in dispute that this program was known and available to the skilled person, so the skilled person was able at the priority date of the patent to measure the ClogP of a given hydrophobic agent following the teaching of the patent and to find suitable ones achieving the promise of the invention.

Since claims 1 and 12 moreover do not contain any limitation as to the method to be used for determining said ClogP, any method known to the skilled person for determining the ClogP of a given hydrophobic agent at the priority date could have been used for finding out whether a given hydrophobic agent had a ClogP as required in claims 1 and 12.

It is thus irrelevant whether the measured ClogP may vary depending on the software (or its updated version) used.

2.1.2 Therefore, this objection rather concerns the clarity of claims 1 and 12 than sufficiency of disclosure of the invention.

2.2 The appellant further argued that it was not possible for the skilled person to identify the "hydrophobic agents" suitable for obtaining the promise of the invention because this term was not further defined in the patent. If this term were not supposed to identify any agent having a ClogP value within the claimed range, it remained totally unclear whether it may have also other meanings, as stated in the decision under appeal (page 8, lines 10-12), such as for example "being not miscible in water".

2.2.1 For the Board, as specified on page 8 (lines 10-12) of the decision under appeal, the term "hydrophobic" is well known in the art and generically identifies compounds which have no affinity to water since - as expressed in the decision - "water repels such compounds". Even though this term may have a relative meaning, the skilled person would however with certainty recognise hydrophobic compounds such as those disclosed in the patent on the basis of their behaviour with water.

In this respect the skilled person would with certainty not consider the "hydrophobic agents" of claim 1 at issue to include water-soluble compounds, which are manifestly hydrophilic and not hydrophobic, even if they might have a ClogP within the limits of claim 1 at issue. The parameter ClogP of claim 1 thus further restricts the class of hydrophobic agents under protection and is definitely not to be used for qualifying any agent as "hydrophobic", as stated by the appellant and in D25. Therefore, also this objection concerns, if at all, the clarity of the claims but not sufficiency of disclosure.

2.2.2 As regards decision T 1227/13, this merely states (point 2.2 of the reasons) that in the context of dyes the term hydrophobic has no absolute, but merely a relative meaning. Thus this decision confirms, if at all, that the term "hydrophobic" has a relative meaning as already explained above. It is further to be emphasised that this decision concerns clarity and not sufficiency of disclosure; it is thus not relevant to the present case.

2.3 The appellant further raised an objection against claim 12, directed to the use of said hydrophobic agents to improve viscostability of the claimed aqueous fabric conditioner composition. This objection was supported by the filing of experimental report D21 which was supposed to show that it was not possible to obtain an improvement in viscostability across the entire scope of claim 12.

The respondent further contested the admissibility of this objection.

2.3.1 The board observes that both parties agreed that this objection was raised the first time during oral proceedings before the opposition division (see point 3 of the minutes referring to the arguments that "there was no proof that the object of the claim could be achieved over the entire scope" and that "the data of the patent were based on one specific type of ester quat and a specified amount of fatty alcohol"). The opponent however did not submit any evidence supporting this allegation.

As this objection was unsubstantiated and thus not considered by the opposition division, which however discussed the sufficiency of claim 12 in its decision (Section 14) with respect to the objection raised in respect to the ambiguity linked to the use of the parameter ClogP, the appellant attempted to substantiate this late filed objection by means of the new experimental report D21 (which - as stated on page 1 of the grounds - had been filed in reaction to the discussion of inventive step in the decision under appeal).

2.3.2 For the board the arguments submitted with the grounds of appeal on the basis of D21 amount to a fresh case and an attempt to expand the reasoning supporting the ground under article 100(b) EPC beyond what had been submitted in due time before the opposition division. Therefore, this new objection is not admitted into the proceedings under Article 12(4) RPBA 2007, as already indicated in the board's preliminary opinion.

2.4 It follows from the above considerations that since the skilled person would have been able to prepare a composition as claimed by following the teaching of the description, and since the data of the patent confirm that viscostability is improved by adding the required hydrophobic agent, the board has no reason to conclude that the invention cannot be carried out by a skilled person without undue burden by following the teaching of the patent and using common general knowledge.

The ground under Article 100(b) EPC thus does not prejudice the maintenance of the patent as granted.

3. Main request - Novelty (Articles 100(a) and 54 EPC)

Appellant cited in this respect documents D2, D2b, D3, D4 and D5 against the novelty of claim 1.

3.1 D2 and D2b have both been published after the filing date of the patent ; these documents are thus prior art in virtue of article 54(3) EPC only.

3.1.1 The appellant cited in particular examples 1 and 4 of D2 and example 2 (composition B) of D2b, which relate to compositions comprising the non-ionic surfactants Lutensol AT25 or Dehydrol LT7. According to the appellant these non-ionic surfactants had a ClogP according to granted claim 1 (see D2b: page 49) and were thus hydrophobic agents (c) in the sense of the patent.

The board notes in this respect that, as stated in the decision under appeal (page 7) and not disputed by the appellant, the surfactant called Dehydrol LT7 is identical to the one identified Dehydol LT7 on page 49 of D2b.

3.1.2 The appellant further argued that during opposition the proprietor itself acknowledged (letter of 12 December 2016, page 2, section "Regarding D2") that the above surfactants were hydrophobic agents in the sense of the patent.

3.1.3 The board observes in this respect that, as explained above, the skilled person would understand claim 1 at issue as relating to a composition comprising as the component (c) a hydrophobic agent further characterised by its ClogP, and not simply a compound (which might also be hydrophilic) having the required ClogP.

Therefore it is the conviction of the board that a skilled person would not consider hydrophilic water-soluble non-ionic surfactants to be "hydrophobic agents" within the meaning of claim 1 at issue. In this respect it has been amply supported by evidence (D18, D26 and D27) that non-ionic surfactants having an HLB above 10 have more affinity for water and are thus water-soluble. In particular the surfactants Lutensol AT 25 and Dehydol LT7, which both have an HLB value (see D19) of approximately 16 and 12, respectively, are therefore undisputedly water-soluble surfactants.

Evidence to the contrary has not been provided by the appellant. Also if different methods of calculation might lead to different values of HLB for a given compound, as stated by the appellant during oral proceedings, the fact that a given surfactant is water-soluble disqualifies it already as a hydrophobic agent according to the patent.

3.1.4 Regarding the proprietor's statement in opposition, the board observes that this statement merely informs the reader that "to the extent that Examples 1 and 4 of D2 destroy novelty of the Claims of the Opposed Patent, the Opposed Patent validly claims partial priority to the priority document". So, this passage does not contain any explicit acknowledgement that the cited surfactants are hydrophobic agents, but rather states that, in case these examples were considered to be novelty destroying, they could not be cited against the patent because of the partial priority to be accorded.

3.1.5 It follows that the surfactants of the cited examples of D2 and D2b cannot be considered to be hydrophobic agents within the meaning of claim 1 at issue. Consequently, already for this reason, D2 and D2b do not destroy the novelty of claim 1 at issue.

3.2 D3, example 2, discloses a composition (composition B of Table 1) comprising the ester-linked quaternary ammonium compound (in the following ester quat) Tetranyl AT1-75.

3.2.1 However, as explained below, D3 does not disclose directly and unambiguously that the ester quat Tetranyl AT1-75 has from 20 to 35 wt% of saturated C18 chains and from 20 to 35 wt% of monounsaturated C18 chains, as required by granted claim 1. The appellant by the way admitted in its statement of grounds (page 9, lines 16-17) that the exact values for the amounts of such C18 chains cannot be derived from D3.

3.2.2 As further stated in D3 (page 8, lines 5-6) this ester quat is based on a mixture of about 25% hard tallow and about 75% soft tallow, i.e. on a similar mixture as Stepantex VT-90 (D11, paragraph [0077]), which is an ester quat according to the patent (see paragraph [0026]).

3.2.3 In the appellant's view the C18 chain distribution of the ester quat of D3 thus had to be similar and could be calculated using the data contained in D12 (page 163, Table 1) for typical fatty acid compositions of animal fats such as beef tallow and mutton tallow, which calculation resulted in relative amounts of the C18 chains largely overlapping with or even being within the range claimed.

3.2.4 The board cannot accept this argument because there is no disclosure in document D3 that the tallow source used for the preparation of the cited ester quat is similar or identical to the one used for the preparation of Stepantex VT-90 or to the beef or mutton tallow described in D12.

3.2.5 Therefore, in the absence of such information in D3, it is not possible to directly and unambiguously derive the amount of saturated and monounsaturated C18 chains contained in the cited ester quat. Already for this reason the claimed subject-matter is novel over D3/example 2.

3.3 D4 and D5. The cited composition of D4 (page 17, lines 5-8) contains the same ester quat as the one disclosed in D3 and the composition cited in Table 12 (column 14, lines 5-20) of D5 comprises the ester quat Tetranyl AT2-75, which according to the appellant is Tetranyl AT1-75, as specified in column 4, lines 26-31 of this document, i.e. the same ester quat as the one used in example of D3.

3.3.1 It follows, for the same reasons as those exposed with respect to D3, neither D4 nor D5 directly and unambiguously discloses an ester quat according to granted claim 1. Therefore, these documents are not novelty-destroying either.

3.4 The board therefore concludes that the claimed subject-matter is novel over the cited prior art.

4. Main request - Inventive step (Article 56 EPC)

The present invention concerns (claim 1) an aqueous fabric conditioner composition of specified viscosity comprising from 2 to 9 wt% of a specific ester quat as fabric softening active, a fatty alcohol and 0.05 to 1 wt% of a hydrophobic agent having a ClogP of from 4 to 9, which is not a fatty alcohol.

4.1 As stated in the patent (paragraph [0013]), fabric softening compositions comprising an ester quat having fatty acid chains comprising 20-35 wt% of saturated C18 chains and 20-35 wt% of monounsaturated C18 chains suffer continued stability issues upon storage. The purpose of the patent in suit is thus to improve their viscostability upon storage.

4.1.1 The appellant referred to D21 and D23 and disputed the the technical problem addressed in the patent.

4.1.2 For the board, as correctly noted by the respondent, D21 cannot be compared with the experimental evidence contained in the patent since it does not even indicate whether the compositions were prepared in the same way. Moreover, D21 only identifies in a generic way the ester quats used without indicating the precise content of C18 chains, and without specifying whether they were similar to the one tested in the patent (TEP-88L) or to any other ester quat specifically disclosed in the patent (paragraph [0026]), thus rendering impossible a verification of the experimental data.

The board remarks also that in the patent (paragraph [0140]), the viscosity was measured initially and after 7, 25, 55 and 85 days of storage at 20ºC whilst in D21, only the initial viscosity and the viscosity after 5 weeks (35 days) at 20ºC and at 50ºC were measured, with these latter values being with certainty not directly comparable with those measured in the patent at a different temperature and after 55 or 85 days of storage.

Therefore, since D21 concerns tests diverging in several respects from those of the patent, its validity for proving that the technical problem underlying the patent does not exist at least across the whole claimed scope is doubtful.

4.1.3 As to D23, also cited in this respect by the appellant, the board notes that this document concerns a specific dispersion of the softener Stepantex VT-90 (one of those falling under claim 1 at issue), in which calcium chloride was used to reduce viscosity and stabilise the system. In the last paragraph of its first page, D23 further establishes that the final viscosity may vary depending on the processing temperature, mixing rate and time, and that fragrance and other ingredients may affect viscosity, and that therefore monitoring of the viscosity was recommended.

Hence D23 clearly discloses that this softening active is not stable and needs a viscosity stabiliser, which thus confirms the technical problem identified in the patent in suit.

4.1.4 Therefore it can be concluded that the technical problem underlying the patent is credible and can serve as a basis for selecting the document representing the closest prior art.

4.2 As possible starting points for the evaluation of inventive step, the appellant cited D3, D6, D7 and D8.

4.2.1 As indicated in the board's preliminary opinion, D8 is the only document having a similar goal as the patent in suit, since it concerns (paragraphs [0001] and [0005]) fabric softening compositions having improved resistance to shear as well as excellent stability and viscosity characteristics upon prolonged storage, and aims in particular (paragraph [0003]) at overcoming the hydrolytic instability of ester quats upon prolonged shelf storage. D8 thus represents the most suitable starting point for the evaluation of inventive step, which was not disputed by the appellant.

4.2.2 As to the closest prior art, this is represented by the fabric softening composition of its example 2, which contains the ester quat Rewoquat V3282 in amounts in accordance with granted claim 1 and 0.3 wt% of the hydrophobic ester glycerol tricaprylate, which has a ClogP of 7.066 as measured with the Crippen Method (see D22).

4.2.3 Since claim 1 at issue does not contain any limitation as to the method to be used for determining the ClogP, glycerol tricaprylate is clearly a hydrophobic agent according to claim 1 at issue, namely one requiring a ClogP from 4 to 9.

4.2.4 The closest prior art thus differs from claim 1 as granted in that it neither contains a fatty alcohol nor discloses the amounts of C18 saturated and monounsaturated fatty acid chains of Rewoquat V3282 and the viscosity of the composition.

4.3 Bearing in mind that the closest prior art already provides aqueous fabric conditioner compositions containing ester quats and having viscostability upon storage and that the patent in suit does not suggest or show any improvement over this prior art, the technical problem underlying the invention (and solved by means of the composition of granted claim 1) must be reformulated in the less ambitious terms of providing an alternative softener composition having stable viscosity upon storage, for example showing a reduced drop in viscosity as submitted by the respondent. This formulation was accepted by the respondent during oral proceedings.

4.4 As regards the question of obviousness of the solution, the following is noted:

4.4.1 According to paragraph [0038] of D8, the compositions disclosed therein have a viscosity of preferably between 10 mPas and 750 mPas when measured at a shear rate of 20s**(-1)(i.e. a much lower shear rate than that used in the patent in suit for measuring viscosity). D8 however teaches (see paragraph [0037]) that said compositions may be subjected to high shear mixing or pumping without experiencing substantial loss or change in viscosity. It follows that the skilled person would expect from a composition having for example a viscosity of 750 mPas not to have a substantially different viscosity when measured at a higher shear rate, such as when measured under the conditions of the patent. Therefore it was obvious for the skilled person following the teaching of D8 to adjust the viscosity of the composition of the closest prior art also at a viscosity within the limits of claim 1 at issue.

4.4.2 D8 further teaches (paragraph [0012]) that preferred ester quats may have N,N-di(acyl-oxyethyl) groups derived from animal fats such as tallow. It would thus have been obvious for the skilled person to try commercial available ester quats of this type, instead of the one used in example 2. One of these known commercially available ester quats is Stepantex VT-90 (see D24, page 8 or D7, page 10, lines 18-19), which is an ester quat according to the patent. It was thus obvious for the skilled person to try in example 2 of D8 an ester quat having from 20 to 35 wt% of saturated C18 chains and from 20 to 35 wt% of monounsaturated C18 chains.

4.4.3 It remains thus to be decided whether it was obvious to add to the composition of D8/example 2 a fatty alcohol in the expectation of providing an alternative softener composition having stable viscosity upon storage.

D8 does not teach to use a fatty alcohol in the disclosed compositions but states at page 9, lines 36-37 that "According to another aspect of the present invention, said fabric softening composition is essentially free of non-ionic surfactant". Since this statement refers only to another aspect of the invention, it does not restrict the broader teaching of D8 and would not have dissuaded the skilled person from trying other components, including non-ionic surfactants, if found suitable for the type of compositions of D8. For the board this statement would thus not have been considered by a skilled person as an advice not to use fatty alcohol in the composition of example 2.

4.4.4 It follows that when looking for possible alternative softener compositions having stable viscosity upon storage, the skilled person would have considered the teaching of other documents relating to aqueous softener compositions based on an ester quat and aiming at improving the viscosity stability upon storage.

One of these documents is D7 (see page 1, lines 10-20; page 2, lines 25-28 and page 9, line 2) which further teaches (page 17, line 26 to page 18, line 4) that fatty alcohols are able to improve the viscosity profile of compositions containing ester quats by complexing with a mono-ester component which may affect detrimentally the initial viscosity.

Thus, even though D7 concerns fabric softening compositions which are different from those of the closest prior art D8 insofar as they comprise a polymeric thickener and do not comprise the stabilising system of D8 (namely a hydrophobic ester and a chelating agent as disclosed e.g. in paragraph [0034]), the skilled person would have taken D7 into account when attempting to formulate alternative ester quat compositions having stable viscosity and he would have considered the technical information contained in D7 regarding the usefulness of fatty alcohols for improving the viscosity profile of such compositions comprising ester quats. He thus would have found in D7 a motivation for trying fatty alcohols as additional components in the composition of D8/example 2.

4.4.5 The respondent argued that D8 itself would have dissuaded the skilled person from applying the teaching of D7 since D8 teaches (paragraph [0035], lines 48-49) that the stabilising hydrophobic ester strongly contributes to drive water molecules out from the vesicle/particles of fabric softening actives.

The board is not persuaded by this argument, since the effect mentioned (of driving water molecules out from the vesicles/particles of ester quat) is not linked to the viscostability of the composition but, as taught by D7 (paragraph [0035], lines 50-51), allows the formation of more condensed vesicles/particles which in turn allow formulating more concentrated compositions with higher fabric softening active concentration.

Therefore, in the board's view, this teaching would not have dissuaded the skilled person from trying a fatty alcohol as a complexing agent for the mono-ester component of the ester quat used in D8 and having a favourable effect on the viscosity profile of the composition as taught in D7.

4.4.6 It follows that it would have been obvious for the skilled person, faced with the technical problem posed, to try a fatty alcohol in the composition of the closest prior art because of its known effect on ester quats based compositions. The board thus concludes that the skilled person, looking for an alternative softener composition having stable viscosity upon storage, would have arrived without inventive skill to a composition having all the features of claim 1 as granted, which thus lacks inventive step.

4.5 The opposition ground under article 100(a) in combination with article 56 EPC thus prejudices the maintenance of the patent as granted.

5. Auxiliary requests 1 and 2

The subject-matter of claim 1 of auxiliary requests 1 and 2 differs from claim 1 as granted only in that it specifies an additional component which, being however optional, cannot distinguish further the claimed subject-matter from the closest prior art.

Hence, the same conclusion regarding inventive step applies to claim 1 of these requests, which are thus not allowable either.

6. Auxiliary request 3

Claim 1 of this request differs from claim 1 as granted in that the range of ClogP of the hydrophobic agent (c) is now limited to a value of from 4 to 7.

6.1 The ClogP of the glyceryl tricaprylate used in the closest prior art D8/Example 2 is 7.066 (see D22). For the board, this value can be approximated to 7.1, i.e. a value greater than the upper limit of the range of values defined in claim 1. Thus the closest prior art differs from granted claim 1 also in the hydrophobic agent (c) used.

6.2 The appellant argued at the oral proceedings that the value of 7.066 should be approximated to 7, and so glyceryl tricaprylate was a hydrophobic agent according to claim 1.

6.3 The board notes that this argument is a new one, which has been presented for the first time during oral proceedings in support of an objection which was not raised before in writing, either.

It is however not immediately obvious for the Board why the value of 7.066 should be approximated to 7 rather than 7.1. It may very well be that in this specific field, it is common practice to do so, as the appellant argued in oral proceedings, however, this being contested by the respondent, the board is not in position to verify this statement without any supporting evidence. It would thus have been incumbent on the appellant to furnish proof in this regard.

6.4 It is further noted that auxiliary request 3 was also not discussed in writing by the appellant, so that no case was presented in writing against this request. The fact that the board in its communication, in absence of appellant's submissions to this end, had not discussed the auxiliary requests, did not relieve the appellant from its duty to present a complete case in due time as required by Article 12(3) RPBA 2020 in order to enable the other party to prepare possibly counter arguments. In other words, it is not up to the board to analyse a possible inventive step attack that has not even been made.

6.5 As required by Article 13(2) RPBA 2020, any amendment to a party's case filed after notification of a summons to oral proceedings shall, in principle, not be taken into account unless there are exceptional circumstances, which have been justified with cogent reasons by the party concerned.

6.6 In the present case there are no exceptional circumstances justifying the filing of such new arguments/objections for the first time during oral proceedings nor the appellant submitted any justification therefor. Therefore, the board under Article 13(2) RPBA 2020 decided not to admit this new argument/objection presented for the first time during oral proceedings.

6.7 In the absence of any objection filed in due time against auxiliary request 3, the board has no reason to investigate this request of its own motion.

Order

For these reasons it is decided that:

1. The decision under appeal is set aside.

2. The case is remitted to the department of first instance with the order to maintain the patent on the basis of Auxiliary Request 3 as filed with the letter dated 19 February 2019, and a description to be adapted thereto.