T 0987/05 (Pentoxides/CABOT) 14-10-2009

Reasons for the Decision

Main request

1. Novelty - Claim 6 as granted

1.1 The documents numbered "D5-.." were cited to prove that tantalum pentoxide according to claim 6, alternative (b) as granted, i.e. having a BET surface area greater than 3 m**(2)/g and a packed bulk density less than 3 g/cm**(3), had been made available to the public before the priority date of the patent in suit (12.10.1995) by virtue of sales of a specific Ta2O5 powder having the claimed properties.

1.2 Document D5-A

1.2.1 Having regard to the technical characterisation of the specific Ta2O5 powder in question, appellant I essentially relied on document D5-A. D5-A is a screen dump (printout of a computer screen) displaying data retrieved from the internal product databases ("Produktregister") of Starck Germany (appellant II). It shows a set of technical and commercial data relating to a specific Ta2O5 material.

1.2.2 The said material is referred to as "Ta2O5 LT micron" and belongs to a lot having the number "930601" (see top data row).

1.2.3 The material is characterised inter alia by a "BET" value of 3,58. Moreover, FSSS values are also mentioned in D5-A (see top left field "Bemerkungen" and "FSSS" field in the lower right part of D5-A, 7th field from the bottom). BET surface areas and FSSS sizes are properties often used to characterise metal or metal oxide powders. These indications thus confirm the assertion of appellant I, that the material in question referred to in D5-A is a Ta2O5 powder with the given BET surface area.

1.2.4 Since in the field of metal oxide powders the surface area measured according to the well known BET method is usually expressed in m**(2)/g, the board has no reason to doubt that the BET surface area value stored in the product database of appellant I is indeed displayed on screen (D5-A) in terms of m**(2)/g.

1.2.5 The board also has no reason to call into question the assertion of appellant I that the value of 1,92 displayed in the field "KD" of D5-A (lower right part, 7th field from the bottom) stands for tapped density ("Klopf-Dichte"), i.e. a packed bulk density in the sense of claim 1 of the patent in suit, which is a property often used to characterise oxide powders. More particularly, the board is not aware of another possible meaning of the acronym "KD" in the context of tantalum pentoxide. The fact that a KD value is indicated confirms that D5-A refers to a Ta2O5 powder. Having regard to the unit of the KD value indicated, the board notes that in Germany (seat of appellant I) density values are usually expressed in g/cm**(3), at least in the technical field concerned. Moreover, a packed bulk density value of 1.92 g/cm**(3) is plausible in connection with Ta2O5 in powder form. Therefore, the board has no reason to doubt that the KD value stored in the product database of Starck Germany is indeed displayed on screen (D5-A) in terms of g/cm**(3).

1.2.6 In view of the total information content of document D5-A, the board is thus convinced that the data displayed relate to a specific lot (No. 930601) of Ta2O5 powder having a BET surface area of 3.58 m**(2)/g and a packed bulk density of 1.92 g/cm**(3), i.e. a tantalum oxide powder falling within the ambit of claim 6, alternative (b), as granted, that was at the disposal of Starck Germany.

1.2.7 The commercial data shown in D5-A in connection with the specified lot of Ta2O5 include a reference to the withdrawal ("Ausgang") of an amount of "200" from said lot no. 930601 for the client "Asahi Optical" on "16.06.1993".

1.3 Documents D5-K, D5-J and D5-L

1.3.1 These documents were cited to show that 200 kg of the specific material referred to in D5-A destined for the final customer Asahi Optical were shipped to Starck Japan (H.C. Stark - V Tech Ltd in Tokyo).

1.3.2 Document D5-J is a copy of an "Inspection Certificate 3.1.B" of Starck Germany dated "06/04/1993" and relating to "Our order No. 14 04641 3". The certificate displays analytical results obtained for a product designated as "Ta2O5 Grade HPC Micron". Also appearing on D5-J are the customer "Asahi Optical", the "Lot No. 930601", the "quantity delivered" of "200 kg" and the number "7150" referred to as "Your order No".

1.3.3 D5-K is a copy of a "Dispatch note" of Starck Germany dated 16 June 1993, addressed to Starck Japan, which according to appellant II is the Japanese distributor company for H.C. Starck products. The dispatch note bears the number "14 04641 3". The dispatch note relates to an order of "200 kg" "Ta2O5 Grade HPC Micron" destined for "Asahi Optical". Under the heading "Signum" the number "7150" is indicated and reference is made to "Lot 930601".

1.3.4 Document D5-L is a copy of an "Invoice" of Starck Germany addressed to Starck Japan dated "16.06.93" and referring to "Order no. 14 4641/3" (see upper right part of D5-L). The invoice concerns the shipment of 200 kg of "TA2O5 TANTALUM PENTOXIDE GRADE HPC MICRON" to Tokyo. It mentions "ASAHI OPTICAL" in the field "Delivery address". The number "7150" appears in the fields "your order" and "marking".

1.3.5 Having regard to the documents mentioned above, the board notes the following: The same lot number "930601" is referred to in D5-A, D5-J and D5-K. Hence, despite the different designations used ("LT micron" in D5-A and "Grade HPC Micron" in the other documents), there is no reason to assume (as done by appellant II) that the material described in D5-A could have been subjected to some kind of processing leading to modified BET surface area or packed bulk density (Klopfdichte "KD") values before the issuance of the certificate D5-J and the dispatch note D5-K. The invoice D5-L was issued on the same day as the dispatch note D5-K and refers to the same order numbers (14 4641 3 and 7150), to the same final customer (Asahi Optical) and to the same amount and type of product as documents D5-J and D5-K.

The board is thus convinced that the material referred to in documents D5-A, D5-J, D5-K and D5-L is actually one and the same material.

1.3.6 Considering the degree of coherence of the data contained in documents D5-A, D5-J, D5-K and D5-L, and in particular the relationship of the respective dates mentioned therein, the fact that they all refer to the same final customer ("Asahi Optical") and the same quantity ("200") of tantalum pentoxide product, the board takes the view that there cannot be any reasonable doubt that 200 kg of the material referred to in D5-A were actually shipped from Starck Germany to Starck Japan on 16 June 1993 for further distribution to the final customer Asahi Optical. This finding is not affected by the fact pointed out by appellant II that there is no express reference in the dispatch note D5-K to the inspection certificate D5-J.

1.3.7 The question of whether or not the product referred to in D5-A has already been made available to the public by virtue of the commercial transaction (invoiced price of 4.592.000,00 Yen) between Starck Germany and its distributor company in Japan need not be decided since, as will appear from the following, the ultimate sales of the material to the final customer Asahi Optical (Asahi Kogaku Kogyo K.K.) anyway represent a novelty-destroying disclosure of the claimed tantalum pentoxide material.

1.4 Documents D5-C, D5-D and D5-E to D5-H

1.4.1 These documents were cited to establish the ultimate sales of the material referred to in D5-A to the final customer Asahi Optical (Asahi Kogaku Kogyo K.K.).

1.4.2 Each of documents D5-D and D5-C is a photocopy of two "purchase orders" from Asahi Kogaku Kogyo K.K. (Asahi Optical) addressed to Starck Japan (translated as "Starck Vitec AG") in Tokyo (compare company address in D5-L and in the translations of D5-C and D5-D). The four "purchase orders" (in chronological order) are dated 29.09.1993, 04.11.1993 (see D5D and translations), 27.12.1993 and 28.01.1994 (see D5C and translations). They concern consecutive orders of 40 kg, 40 kg, 80 kg and 40 kg, respectively, of a product with the "product code 60367062" and the "order code 9614", which in each case is designated as "Tantalum oxide (t) 4N" and has a price of 26200 Yen/kg (see the translations into German). Each of the four purchase orders (see the original documents in Japanese) bears the handwritten number "7150" (which became "2150" in the translations). The latest "purchase order" of 28.01.1994 additionally bears the comment "final shipment" (see the translation into German: "Schlußlieferung"), which indicates that this order was the last order of a series of orders. Moreover, at least three of the purchase orders bear the inscription "(6759)", which is not clearly legible in the upper purchase order shown in D5-D.

1.4.3 Documents D5-H, D5-G, D5-F and D5-E are "sales invoices" respectively dated 30.10.1993, 01.12.1993, 01.02.1994 and 01.03.1994 (see originals in Japanese and their translations). The four invoices were issued by Starck Japan in Tokyo ("Starck Vitec AG" in the translations) and addressed to Asahi Kogaku Kogyo K.K. (Asahi Optical). All four bear an invoice number

"6759-" and concern four consecutive deliveries of, respectively, 40 kg, 40 kg, 80 kg and 40 kg of a product designated as "tantalum oxide (t) 4N" for a price of 26200 Yen/kg (see the translations into German). All four invoices additionally bear the inscription "(6759)" and two of them (see D5H and D5G) bear the handwritten number "7150" (which became "2150" in the translations). Each invoice refers to a delivery number and delivery date (two columns at the left).

1.4.4 Concerning the data indicated in the two series of documents D5-C/D5-D and D5-E to D5-H, the board inter alia notes the following:

i) the two companies (Starck Japan and Asahi Kogaku Kogyo) involved in the documented transactions are the same in each case;

ii) the dates specified in the two series of documents and their order in time correspond to each other respectively (although according to D5H the first delivery was earlier than foreseen in D5D);

iii) the product designation, price and amounts also match in the two series of documents; and

iv) the inscription "(6759)" is almost omnipresent.

In view of the high degree of coherence of the data displayed in the two series of documents D5-C/D5-D and D5-E to D5-H, the board considers it established beyond reasonable doubt that in 1993/1994, i.e. before the priority date of the patent in suit, Asahi Kogaku Kogyo K.K. bought and received a material designated as "Tantalum oxide (t) 4N" in four batches with a total amount of 200 kg.

1.4.5 In the absence of evidence to the contrary, the board considers that the sales in question were normal business transactions in the sense that they were not subject to any kind of confidentiality agreement between the two companies involved.

1.5 The product sold to Asahi Kogaku Kogyo K.K. was the Ta2O5 material referred to in D5-A.

1.5.1 As evidenced by documents D5-J, D5-K and D5-L, the marking "7150" is the order number under which Starck Japan "internally" ordered and received from Starck Germany 200 kg of Ta2O5 out of the lot no. 930601 referred to in D5-A, this material being ultimately destined for the customer Asahi Optical in Japan. The number "7150" also appears in handwriting on all four purchase orders originating from this customer (see D5-C and D5-D) and on two of the invoices addressed to this customer (D5-G and D5-H). In the board's view, this cannot possibly be a coincidence, in particular when taking into account the coherence of the other data referred to in, on the one hand, D5-A and D5-J to D5-L, and in, on the other hand, D5-C/D5-D and D5-E to D5-H. These coherent data include in particular the dates specified and their order, the companies involved in the transaction, the nature and total amount of material involved. Under theses circumstance, the board excludes the possibility that the number "7150" could have been applied to classify the documents according to another aspect of the transactions, e.g. by a customer number, as suggested by appellant II.

1.5.2 The board thus concludes that there is no reasonable doubt that the material sold and delivered to the ultimate customer Asahi Kogaku Kogyo K.K. in four consecutive batches of 40, 40, 80 and 40 kg was indeed the material referred to in D5-A, ordered and received "internally" by Starck Japan under the order number "7150".

1.6 Summarising, there is a convincing uninterrupted chain of proof corroborating the allegation of appellant I that the specific Ta2O5 powder referred to in D5-A has been made available to the public before the priority date of the patent in suit by virtue of sales of substantial amounts of this material to the company Asahi Optical (Asahi Kogaku Kogyo K.K.) in Japan. A skilled person can determine the BET surface area and packed bulk density values of this particular powder by well known analytical methods and reproduce it without undue burden by adapting known preparation methods so as to obtain these values. No arguments to the contrary have been submitted by appellant II. Therefore, in accordance with the principles laid down in opinion G 0001/92 (OJ EPO 1993, 277, Conclusion), the sales to Asahi Optical made available to the public a tantalum powder with a BET surface area and a packed bulk density as required by claim 6 as granted.

1.7 The board also considered the further aspects which led appellant II to argue that the evidence cited was not sufficient to prove convincingly that tantalum oxide according to claim 6 became publicly available before the priority date of the patent in suit. However, the board considers that for the reasons given below, these aspects do not give rise to reasonable doubts strong enough to break the chain of proof set out above.

1.7.1 In the board's view, the reasons for which "7150" became "2150" in the translations of documents D5-C, D5-D, let alone in the translations of documents of

D5-G and D5-H, are not apparent. The handwritten figure "7" appearing on the original documents in Japanese does not resemble a "2" at all. An example of a handwritten "2" can be found in document D5-C (see lower purchase order, delivery date "2/4", left bottom corner).

1.7.2 The board has no reason whatsoever to assume that the number "7150" was hand-written on the purchase orders and sales invoices at a much later stage, remote from the time of the business transaction concerned. Moreover, the said inscriptions establish the identity of the material ordered by and delivered to the ultimate client and the material shipped from Starck Germany to Starck Japan, irrespective of exact date on which they were written on the four purchase orders shown in D5-C and D5-D.

1.7.3 Appellant II also emphasised the following aspects:

- The order number "7150" used in the "internal" transaction between Starck Germany and Starck Japan only appears in hand-written form on the documents relating to the business transactions in Japan between Starck Japan and Asahi Optical; and

- the "product code", "order code", "order number" and product name used in the latter transactions differed from the ones used "internally" between Starck Germany and Starck Japan.

In the board's view, this particular handling and classification of orders received from customers and the subsequent invoicing is not something so peculiar that it would justify reasonable doubts with regard to the identity of the product referred to in D5-A and the product sold and delivered to Asahi Optical.

1.7.4 Considering the overall coherence of the (printed) data in the "purchase orders" and "sales invoices", with regard to the chronology of the orders and deliveries, the amounts concerned, the consistency in customer name and invoice numbering, and the corresponding customer's order numbers/delivery numbers (although only partially visible in the Japanese original documents D5-E to D5-H), the fact that the number "7150" was only handwrit-ten on two (D5-G and D5-H) out of four sales invoices does not mean that it would be impossible to allocate these documents to the corresponding purchase orders.

1.7.5 In view of the totality of the evidence on file in the present case and of its coherence, the fact that no proof of payment and no acknowledgement of receipt from the purchasing company have been filed is not sufficient to call into question the delivery chain of the material referred to in D5-A from Starck Germany to the ultimate customer Asahi Kogaku Kogyo K.K. In the present case, such specific evidence is not required to prove beyond reasonable doubt that the products ordered by the customer and billed (with reference to delivery dates) by the seller were indeed delivered to the customer company.

1.7.6 The answers to the questions of whether or not the handwritten mark "10/5" on D5-H concerns an amended delivery date and why only the latter document is marked with a paraph have no bearing on the board's foregoing findings.

1.7.7 For the board, there are no gaps in the chain of commercial transactions, so that it is established beyond any reasonable doubt that a product according to claim 6 as granted was ultimately sold to Asahi Kogaku Kogyo K.K.

1.8 The board therefore concludes that the subject-matter of claim 6, alternative (b), lacks novelty over the tantalum pentoxide material made available to the public by virtue of the sales evidenced by documents D5A, D5-J to D5-L and D5-C to D5-H (Article 52(1) and 54(1)(2) EPC).

Auxiliary requests - Admissibility

2. The amendments in the claims according to the auxiliary requests 1 to 15 of appellant II restrict the ambit of the claims. Under the circumstances of the present case, the increased number of independent claims of the same category according to some of the auxiliary requests is not objectionable (see e.g. T 0263/05, OJ EPO 6/2008, 329, point 5. of the reasons and Headnote I, II.1 an II.2). The said amendments are thus not objectionable under Rule 80 EPC.

Some of the amendments in the claims according to the auxiliary requests 4 to 15 filed at the oral proceedings constitute an attempt to address objections under Articles 123(2)(3) and 84 EPC raised in writing by appellant I in its last written submission of 14 September 2009. Despite the number of requests filed, the actual amendments made are not particularly complex.

Therefore, the board took all the auxiliary requests of appellant II into consideration.

Auxiliary requests - Allowability of the amendments

3. Each of the auxiliary requests 1, 3, 5, 7, 9, 11, 13 and 15 comprises an amended independent claim directed, in one of the claimed alternatives, to a "niobium pentoxide powder characterized as ... (b) having a BET surface area of greater than 2m**(2)/g and a packed bulk density of less than 1 g/cc" (emphasis added). The claims concerned are

- in auxiliary request 1: claim 5;

- in auxiliary request 3: claim 5;

- in auxiliary request 5: claim 4;

- in auxiliary request 7: claim 4;

- in auxiliary request 9: claim 5;

- in auxiliary request 11: claim 5;

- in auxiliary request 13: claim 1; and

- in auxiliary request 15: claim 1.

3.1 In each of the aforementioned independent claims, the upper limit of the range for the packed bulk density value has been lowered from 1.8 (see alternative (b) according to claim 5 as granted and claim 5 of the application as originally filed) to 1.

3.2 Concerning the basis for this amendment, appellant II referred to page 13, second paragraph of the application as filed (published as WO 97/13724 A1), which reads:

"A second embodiment of valve metal pentoxide may be characterized as high surface area, low bulk density, calcined, niobium pentoxide particles having a BET surface area greater than 2 m**(2)/g, preferably greater than 2.5 m**(2)/g, more preferably greater than 3 m**(2)/g, even more preferably greater than 4 m**(2)/g, and in some cases yet even more preferably greater than 6 m**(2)/g, and a packed bulk density of less than 1.8 g/cc, preferably less than 1 g/cc and in some cases more preferably less than 0.75 g/cc."

3.3 However, the quoted paragraph does not expressly refer to Nb2O5 having both a BET surface area greater than 2 m**(2)/g and a packed bulk density of less than 1.8 g/cm**(3). In other words, this paragraph does not refer to Nb2O5 having a BET surface area in the most general range specified in the application as filed, namely a BET surface area which may be (almost) as low as 2 m**(2)/g and, at the same time, a relatively low packed bulk density in the preferred range of less than 1 g/cm**(3).

3.4 It is noted that whereas Table 1 of the application as filed refers to a sample of Nb2O5 (No.3) that falls within the ranges according to alternative (a) of the claims in question, Table 1 contains no example of Nb2O5 meeting the double criterion defined in alternative (b) of said claims, i.e. which has both a packed density of less than 1 g/cm**(3) and a BET surface area greater than 2 m**(2)/g.

3.5 On the other hand, Table 1 refers to two examples (Nos. 1 and 4) having surface areas (2.82 and 2.18 m**(2)/g, respectively) relatively close to but greater than the value of 2 m**(2)/g. However, the packed densities of these two samples (1.71 and 1.62 g/cm**(3), respectively) are substantially higher than the upper limit of 1 g/cm**(3) specified in alternative (b) of the amended claims.

3.6 Example 5 of the application as filed relates to Nb2O5 having a packed density of 1.02 g/cm**(3), and constitutes the only example of Nb2O5 actually having a packed density value close to (but nevertheless slightly outside and higher than) the claimed range of "less than 1 g/cc". However, the corresponding surface area of this Nb2O5 sample is not indicated in the table, and it cannot be derived from the application as filed, not even from the data presented in Table 1 or the passage quoted by appellant II, that this powder would implicitly have a BET surface area of more than 2 m**(2)/g.

3.7 More generally speaking, a clear and precise, explicit or implicit correlation between these two properties can neither be inferred from the description passage invoked by appellant II, nor from the data displayed in Table 1 or any other part of the application as filed, in particular not for powders having BET surface area values close to but greater than 2 m**(2)/g and a packed bulk density of less than 1 g/cm**(3).

3.8 Under these circumstances, the board concludes that nothing in the application as filed points towards a Nb2O5 powder with the specific combination of properties according to alternative (b) as now claimed. Such powders having a BET surface area which may be almost as low as 2 m**(2)/g and, at the same time, a packed bulk density of less than 1 g/cm**(3) are thus not directly and unambiguously disclosed in the application as filed, neither implicitly nor explicitly.

3.9 The amended claims concerned, although restricted in scope, are thus directed to subject-matter generated by the amendment and extending beyond the content of the patent application as filed, contrary to the requirements of Article 123(2) EPC.

3.10 Therefore, auxiliary requests 1, 3, 5, 7, 9, 11, 13 and 15 must be rejected.

Auxiliary requests 2, 4, 6, 8, 10, 12, 14 - Novelty

4. Each of the sets of claims according to the auxiliary requests 2, 4, 6, 8, 10, 12 and 14 comprises an independent claim having the same wording, except for the numbering, as claim 6 as granted. The claims concerned are

- in auxiliary request 2: claim 6;

- in auxiliary request 4: claim 5;

- in auxiliary request 6: claim 5;

- in auxiliary request 8: claim 6;

- in auxiliary request 10: claim 6;

- in auxiliary request 12: claim 2; and

- in auxiliary request 14: claim 2.

4.1 The subject-matter of claim 6 as granted lacks novelty for the reasons given under points 1 to 1.8 hereinabove.

4.2 The reasons for rejecting the main request apply mutatis mutandis to these auxiliary requests which each contain a claim with the same wording as claim 6 as granted.

4.3 Consequently, auxiliary requests 2, 4, 6, 8, 10, 12 and 14 must also be rejected.

5. Summarising, none of the requests of appellant II is allowable. The further objections raised against the process claims and other independent product claims according to the various requests of the appellant II, which were also discussed at the oral proceedings, have no bearing on the board's above conclusion and need not, therefore, be dealt with in the present decision.