Thursday, 8 September 2011

T 390/08 – Not For Laymen


The patent proprietor filed an appeal against the decision of the Opposition Division (OD) revoking the opposed patent.

Claims 1 and 9 of the application as filed read (my emphasis):
1. Process for the production of a hydrogen rich gas without formation of hydrocarbons comprising water gas shift conversion of a gas containing carbon monoxide and steam at a temperature of between 400°C and 850°C in the presence of a catalyst, which catalyst comprises one or more of the elements Mg, Mn, Al, Zr, La, Ce, Pr, and Nd, being able to form basic oxides, and mixtures thereof.

9. Process according to anyone of the preceding claims, wherein said catalyst further contains alkali metals.
Claims 1 and 8 as granted had an almost identical wording.

Claim 1 of the main request before the OD read (my emphasis):
Process for the production of a hydrogen rich gas without formation of hydrocarbons comprising water gas shift conversion of a gas containing carbon monoxide and steam at a temperature of between 400° and 850°C in the presence of a basic metal oxide catalyst, wherein the metal of said catalyst consists of one or more of the elements Mg, Mn, Al, Zr, La, Ce, Pr and Nd, and mixtures thereof, and said catalyst optionally further contains alkali metals.
In its decision, the OD held this claim not to comply with the requirements of A 84, because the feature “the metal of said catalyst consists of one or more of Mg, Mn, Al, Zr, La, Ce, Pr or Nd” – which explicitly excludes other metals – was in contradiction with the further feature that “said catalyst optionally further contains alkali metals”.

The OD further held the main, first auxiliary and second auxiliary requests then on file not to comply with the requirement of sufficiency of disclosure (A 100(b)). It argued that the feature “without formation of hydrocarbons”, which was present in claim 1 of all these requests, did not need any interpretation, as it only bore the meaning that no hydrocarbons were formed irrespective of the method which was used to determine their presence. Thus, the skilled person should be able to obtain the desired result – a hydrogen rich gas without hydrocarbons – in a reliable way, irrespective of the method of evaluating the composition of the gas product. In the case at issue, by using the gas chromatography method for evaluating the said gas, the skilled person would not be able to perform the invention “without” formation of hydrocarbons as it could not determine unambiguously whether hydrocarbons were formed or not. The argument that formation of trace amounts of hydrocarbons did not impede the possibility of carrying out the claimed process could not be accepted, because the aim of the patent-in-suit was explicitly to avoid concomitant formation of hydrocarbons.

A further lack of clarity on how the process was to be conducted “without formation of hydrocarbons” arose in particular from examples 1 to 4, 12, 13, 16, 20, 21 and 41, which were defined as embodiments according to the invention, but showed formation of methane well above the detection limit of 15 ppm. So, it was not possible on the basis of the examples to determine which process conditions were necessary to produce a hydrogen rich gas without concomitant formation of hydrocarbons.

Together with its statement setting out the grounds of appeal, the patent proprietor submitted a main request claim 1 of which read (my emphasis):
1. Process for the production of a hydrogen rich gas without formation of hydrocarbons comprising water gas shift conversion of a gas containing carbon monoxide and steam at a temperature of between 400° and 850°C in the presence of a basic metal oxide catalyst, wherein the metal of said catalyst consists of one or more of the elements Mg, Mn, Al, Zr, La, Ce, Pr and Nd, and mixtures thereof.
Dependent claim 8 of the same request read (my emphasis):
8. Process according to any one of the preceding claims, wherein said catalyst further contains alkali metals.
In preparation to the oral proceedings, the board expressed its provisional opinion that the wording of the above claims appeared not to fulfil the clarity requirement of A 84.

The patent proprietor then filed a new main request claim 1 of which read (my emphasis):
1. Process for the production of a hydrogen rich gas without formation of hydrocarbons comprising water gas shift conversion of a gas containing carbon monoxide and steam at a temperature of between 400° and 850°C in the presence of a basic metal oxide catalyst, wherein the metal of said catalyst consists of one or more of the elements Mg, Mn, Al, Zr, La, Ce, Pr and Nd, and mixtures thereof, wherein said catalyst optionally further contains alkali metals.
Will this claim – which is almost identical to the claim found unallowable by the OD; the only difference being the replacement of “and” by “wherein” in the last feature – overcome the objections?

Main request – Amendments

[1] Claim 1 of this request results from the straightforward combination of claims 1 and 9 of the application as filed, and so it meets the requirements of A 123(2).

The respondent argued that claim 9 as originally filed defined the catalyst as further containing alkali metals, hence any amendment to include alkali metals should not define the latter as optional but rather as a further definite component. This argument is not accepted by the board, because when a feature is included in a dependent claim, it is conventional in patent law to consider said feature as optional in comparison to the features defining the subject-matter of the independent claim on which said feature depends.

Clarity

[2] The definition of a composition as “consisting of” compulsory and optional components meets the requirements of A 84, as such a composition defines a “closed” composition from which elements other than those mentioned are excluded (see in particular decisions T 1998/07 [2.1-2] and T 1190/01 [2]). This wording furthermore prevents contradiction between an independent claim and a dependent claim relating to a preferred embodiment including a further element not yet defined in the independent claim (see in this respect T 1998/07 [2.1]).

Therefore, claim 1 meets the requirements of clarity of A 84.

Sufficiency of disclosure

[3.1] According to A 83 and its counterpart in A 100b), the requirement of sufficient disclosure means that an invention shall be disclosed in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art.

[3.2] In the case at issue, the decision to revoke the patent for lack of sufficient disclosure of the invention was exclusively based on the feature “without formation of hydrocarbons” that the OD held as not needing any interpretation, as it only bore the meaning that no hydrocarbons were formed irrespective of the method which was used to determine their presence.

[3.3] The board observes that in the absolute it is true that a layman reading claim 1 at issue would literally understand that the result of the process features claimed - namely the “water gas shift conversion of a gas containing carbon monoxide and steam at a temperature of between 400º and 850ºC in the presence of a basic metal oxide catalyst, wherein the metal of said catalyst consists of one or more of the elements Mg, Mn, Al, Zr, La, Ce, Pr and Nd, and mixtures thereof, wherein said catalyst optionally further contains alkali metals” - would be the total absence of hydrocarbon by-products in the hydrogen-rich product gas.

However, a patent specification is not aimed at a layman, but at a skilled person with common general knowledge in the technical field concerned. Furthermore, sufficiency of disclosure of a patent is not to be assessed on the basis of the claims alone, but on the patent specification as a whole (see e.g. T 14/83 [3]; T 202/83 [2]).

[3.4] In the present case, the patent specification (paragraph [0017]) describes the invention as the provision of “a process for producing a hydrogen rich gas by contacting an effluent gas from a steam reforming unit with a basic metal oxide catalyst at high temperatures, preferably from 400°C to 850°C, with significantly less hydrocarbon by-product formation than may be accomplished by contact with a conventional iron-chromium high temperature shift catalyst” (emphasis added by the board). In paragraphs [0021] and [0022], it further explains that “it was surprising that with catalysts comprised by basic oxides of main group metals, rare earth metals or mixtures thereof in crystalline or amorphic form, significant CO-conversion was observed while essentially no hydrocarbons (emphasis added) were formed, […]. One of the most active catalysts was a catalyst comprised by magnesium oxide stabilised with alumina (catalyst B). Even at a very low steam/carbon ratio, no detectable amount of hydrocarbons was formed within 24 hours on stream at 650°C.”

[3.5] In summary, the skilled reader of the patent specification is taught that, in comparison to a conventional iron-chromium high-temperature shift catalyst, the catalysts of the alleged invention are supposed to produce a hydrogen rich gas containing “essentially no hydrocarbons”, or even “no detectable amount of hydrocarbons” as regards the “most active catalysts” based on magnesium oxide stabilised with alumina.

Hence, the skilled person learns from the description of the patent specification that a broader technically meaningful interpretation is to be given to the expression “without formation of hydrocarbons” than the strict literal interpretation held by the OD.

The board further holds the interpretation of the OD that “no hydrocarbons were formed irrespective of the method which was used to determine their presence” as not appropriate, because it is common general knowledge that any measurement method has its technical limits, in particular as regards the detection precision, and that this limit varies from one method to the other.

[3.6] The contested patent (paragraph [0031]) further describes the analysis of the product gas, in particular as regards the presence of higher hydrocarbons and CH4, as having been made by gas chromatography using argon as the internal standard. The contested patent (page 5, line 7) also discloses the detection limit of the gas chromatography equipment to be 15 ppm of methane. So, there is no gap of information in the contested patent as regards the measurement method used and the skilled person is clearly and unambiguously informed how the formation of hydrocarbons is to be assessed.

[3.7] The contested patent (Table 1; pages 7 to 9) further discloses numerous examples – Experiments 1 to 4, 6 to 19, 30, 32, 34 and 36 to 39, 90 (sic) and 41 – which are in conformity with a technically meaningful interpretation of claim 1 at issue and so provides a “hydrogen rich gas without formation of hydrocarbons”, since the gas produced contains - depending on the catalyst used and the operating conditions - from 0 to 121 ppm methane and no higher hydrocarbons.

In comparison, the iron-chromium catalyst - considered as reference in the introductory part of the patent specification - when tested under comparable conditions was assessed to produce a hydrogen rich gas containing 290 ppm CH4 (Experiment 23), i.e. more than twice as much hydrocarbons as the worst catalyst according to the alleged invention.

In this context, and owing to the fact that the contested patent thus contains not less than 27 specific embodiments falling under the wording of claim 1 at issue and not less than 14 specific embodiments – Experiments 6 to 11, 14, 15, 17 to 19, 36, 39 and 90 – in which the gas produced was analysed to contain 0 ppm CH4 and no higher hydrocarbons, the contested patent offers sufficient guidance since it describes in detail different ways of carrying out the invention claimed.

[3.8] For the above reasons, the board holds that the invention as now claimed, interpreted in a meaningful way, is disclosed in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art pursuant to the requirements of A 100(b) together with A 83.

The case was then remitted to the OD for further prosecution.

I find the clarity part somewhat surprising. For the Board, when the discrepancy between the main claim and the dependent claim is imported within the main claim, the lack of clarity disappears. Interesting. I may be nit-picking, but the claim did not read “… wherein the metal of said catalyst consists of one or more of the elements Mg, Mn, Al, Zr, La, Ce, Pr and Nd, and mixtures thereof, and optionally further alkali metals” as the Board appears to have read.

Should you wish to download the whole decision, just click here.

The file wrapper can be found here.

4 comments:

Anonymous said...

It is not so odd that having the optional alkali metal only in a dependent claim would lack clarity. Then the main claim would have told you that the catalyst consisted of A, and only from the dependent claim it would have become apparent that it could also consist of A+B. Such a dependent claim would actually be broader than the main claim, and therefore not truly dependent.

Anonymous said...

Part 1/2

The prosecution of this case leaves me with a general feeling of sloppiness.

The non-BEST EP search report consists of, or comprises only, (whatever you prefer), "A" (as in "Allgemeine") documents. This could have prompted the ED to grant without giving any further consideration to substantive matter. However the opponent relied on some of the SR documents for novelty and inventive step, and the arguments developed seem serviceable.

The patent was issued nearly in the form in which it had been filed, with the only amendments undertaken by the examining division. These can be seen on page 17 of the Druckexemplar; page 5 was also trivially amended.

Claim 3 was simply struck by the 1st examiner, reportedly after a phone discussion on 12.03.03 with Mrs. (sic) Seitz. There is no trace of a record of this phone conversation in the file. Did the applicant protest?

He could have refused the communication under Rule 51(4) EPC 1973. In that case a proper examination would have had to be undertaken, with a formal communication under Art. 96 EPC 1973. Or perhaps the prospect of a quick grant with a broad claim scope overruled any serious consideration about what could happen down the road.

The substance of initial claim 3 was reintroduced during the opposition procedure, and its features form the basis of the BoA decision discussed here. This subject-matter seems crucial to the proprietor. The opponent did raise Rule 57a EPC 1973 (now as Rule 80) objections, but apparently not against this particular amendment.

The wording of claim 1 was also amended to claim the "basic metal oxides" of the enumerated elements, instead of merely stating that the said elements are "able to form basic oxides". I'm not a chemist (and therefore a "layman" in the sense of the decision), so I had to reach for my bookshelf (and the internet) to look up the meaning of "basic metal oxide". I learned that at least Al2O3 and ZrO2 are classified as "amphoteric", i.e., can show either basic or acidic properties depending on the solvent. I'm not sure whether "amphoteric" can be subsumed uder "basic". Art. 83 problem, or am I just out of my depth?

The decision appealed is ONLY grounded in Articles 83 and 84, and no consideration have been given to novelty and inventive step at any point, even as a backup argument. It seems to me that sticking to more formal points (including the particular Art. 83 reasoning) plays into the hands of the proprietor, and unduly prolongs the proceedings at the expense of legal certainty. Considering that both the opposition and appeal procedures in this case each lasted more than three and one half years, the final outcome of the case might only be settled close to the patent's expiration...

The OD did act promptly after the decisions, by issuing summons on 6.9.2011 for an OP scheduled on 8.5.2012. Eight mongths! That's almost as long as a pregnancy. The annexes to the summons are not online yet, they are probably in the process of being scanned.

Anonymous said...

Part 2/2

I also looked at foreign parents to this patent.

A Canadian patent was recently issued, just before the BoA issued its decision.

The CA-C claims are interesting:

1. Process for the production of a hydrogen rich gas without formation of hydrocarbons comprising water gas shift conversion of a gas containing carbon monoxide and steam at a temperature of between 400°C and 850°C in the presence of a catalyst consisting essentially of zirconium oxide, magnesium oxide deposited on an aluminum oxide support, or mixtures thereof.

2. Process for the production of a hydrogen rich gas without formation of hydrocarbons comprising water gas shift conversion of a gas containing carbon monoxide and steam at a temperature of between 400°C and 850°C in the presence of a catalyst consisting essentially of (1) alkali metals and (2) zirconium oxide, magnesium oxide deposited on an aluminum oxide support, or mixtures thereof.


The dependent claims are formulated like this:

xx. Process of claim 1 or 2, wherein [...]

I'm not sure whether this would satisfy Art. 123(2) EPC, or the equivalent disposition of the Canadian Patent Act, but it does show what the invention really is. The selection of the particular oxides was part of the proprietor's requests in the procedure.

I don't know how much these claims were influenced by the foreign procedures. Did the CIPO examiner find a killer document, or just looked at what happened abroad? There is not online file inspection in CA, unfortunately.

The US procedure under publication number 20010055560 is also interesting. The file wrapper is not available in PAIR, but the transaction history is revealing:

05-02-2003 Mail Express Abandonment (During Examination)
05-02-2003 Express Abandonment (during Examination)
04-25-2003 Letter of Express Abandonment Filed
02-28-2003 Mail Non-Final Rejection
02-25-2003 Non-Final Rejection
08-13-2001 Information Disclosure Statement (IDS) Filed
04-23-2001 Information Disclosure Statement (IDS) Filed
04-23-2001 Preliminary Amendment

"Express abandonments" are to my knowledge uncommon at the USPTO, applicants rather let their applications die by not responding to an Office Action, very much like what happens at the EPO. Here the application was withdrawn less than two months after the OA. Did the US examiner uncover a killer document too? From the IDS? Was the OA scathing? I notice that the time frame coincides with that of the R. 51(4) communication from the EPO.

There are no Continuations or CIPs.

Why keep prosecuting a CA patent whilst abandoning the US counterpart? Money does not seem to have been the problem.

As a side note, I also have doubts about the validity of the priority, the range disclosed in the DK priority is even broader than that initially claimed in the EP application.

oliver said...

Quite impressed ... by your in-depth case study !