T 0826/10 11-03-2014

Reasons for the Decision

1. The appeal is admissible.

2. Article 123(2) EPC: Main Request

2.1 The subject-matter of claim 1 of the main request is a combination of the subject-matter of claims 1 and 10 as originally filed, plus features disclosed in the passage of the description from page 6, line 30 to page 7, line 1 (see also original claim 4 in this regard). Dependent claims 2 to 6 are based on claims 2, 3, 5, 7 and 8 as originally filed.

The Board is therefore satisfied that the main request fulfils the requirements of Article 123(2) EPC.

3. Clarity of the term "pillar"

3.1 Under the title, "Additional Remarks", the Examining Division expressed reservations about the use of the term "pillar" in claim 1, and provided an explanation of the manner in which this term had been interpreted for the purposes of assessing novelty and inventive step.

Although the appellant referred, in the statement of grounds of appeal, to "the objection of lack of clarity", the Examining Division did not in fact raise any specific objection under Article 84 EPC 1973 or any other provision of the EPC in this regard. The Board also sees no reason to raise any such objection.

4. Inventive Step: Main Request

4.1 In the contested decision document D2 was cited as the closest prior art; the Board also considers it to be a suitable starting point for evaluating inventive step.

In determining which of the features of claim 1 of the main request may be identified in document D2, it should be noted that the terms "nonsolderable" and "solderable" used in the present application have essentially the same meaning as the terms "non-wetting" and "wettable" used in document D2 (compare the definitions of "nonsolderable" and "solderable" in the description of the present application, page 7, lines 2 to 6 with the definitions of "non-wetting" and "wettable" in document D2, column 3, lines 32–41 and 50–53).

Furthermore, document D2 discloses that the solder balls are sited on pads which may be on the substrate or on the integrated circuit. For the purpose of comparison with claim 1, the embodiment of D2 in which the solder balls are on the pads of the substrate will be considered.

4.2 Document D2 discloses (in figures 2 and 5–7, and the associated text) a method comprising the sequential steps of:

- forming a nonsolderable conductor (26 or the element having surface 72) formed of a nonsolderable material on a substrate (21,71) so as to define a conductor pattern;

- forming a solderable conductor pillar (27 or the element having surface 73, there being no reason why this element may not be referred to as a "pillar") formed of a solderable material on the nonsolderable conductor, the solderable conductor pillar constituting a limited portion of the nonsolderable conductor (see figure 1 etc.);

- registering a surface mount device with the conductor pattern (see column 1, lines 11–19 and column 2, line 60 to column 3, line 19); and

- heating the solder bump so as to reflow the solder bump and form a solder bump connection that bonds the surface mount component to the solderable conductor pillar.

4.3 The principal feature distinguishing the subject-matter of claim 1 from that of document D2 is therefore as follows:

- "registering a surface mount device with the conductor pattern, wherein a solder bump (16) on the surface mount device contacts the solderable conductor pillar" (emphasis added by the Board).

By contrast, in document D2 the interconnect ball 74 (i.e. the "solder bump") is already mounted on the surface 73 (i.e. the "solderable conductor pillar") prior to registration. At the registration stage the interconnect ball contacts a pad on the surface mount device (integrated circuit) and reflow takes place.

4.4 A further difference between the subject-matter of claim 1 and that of document D2 is the following:

- "the solderable conductor pillar (114) having a shape chosen from the group consisting of square, circular and shapes that are elongate in a longitudinal direction of the conductor (112)".

4.5 In the description of the present application (for example, page 4, lines 8–15 and page 7, line 27 to page 8, line 24), the problem solved by the claimed method is said to be to provide accurate control of the height, shape and distribution of the solder connection between a surface mount device and a substrate following the reflow step by which the surface mount device and the substrate are mechanically and electrically connected. The Board is satisfied that this represents a technical problem which is plausibly solved by the principal distinguishing feature of claim 1 of the main request.

This being the case, no justification can be seen for reformulating this problem to the more general one of "mounting a surface mount device to the substrate", as stated in the contested decision. It is established case law that "the correct procedure for formulating the problem is to choose a problem based on the technical effect of exactly those features distinguishing the claim from the prior art that is as specific as possible without containing elements or pointers to the solution" (T 1019/99, point 3.3 of the Reasons). The general formulation "mounting a surface mount device to the substrate" is therefore inappropriate as it does not take into account the specific technical contribution of the principal distinguishing feature of claim 1 of the main request.

The Board therefore considers that the objective problem solved by the principal distinguishing feature of claim 1 of the main request is to provide accurate control of the height, shape and distribution of the solder connection between a surface mount device and a substrate following the reflow step.

4.6 Document D2 is concerned with a different technical problem, relating to the formation of the solder ball itself. This problem is to provide a method which "allows an interconnect ball of increased height to be formed on a pad" (column 6, lines 41 to 46). It should be noted that although the procedure of forming a solder ball also involves a reflow step (see column 5, line 58 to column 6, line 29 and figures 6,7), this is a quite different reflow step to that which occurs in relation to fixing the integrated circuit to the substrate.

A brief mention is made in document D2 of the registration and reflow steps whereby the integrated circuit and the substrate are mechanically and electrically connected (column 1, lines 11–19 and column 2, line 60 to column 3, line 19). However, nowhere in document D2 is there any reference or allusion to providing accurate control of the height, shape or distribution of the solder connection between the integrated circuit and the substrate following the reflow step.

Document D2 was the only document cited in the contested decision. It follows that no document cited in support of the finding of lack of inventive step mentions – or even hints at – the objective problem solved by the principal distinguishing feature of claim 1. This fact alone is considered to call into serious question the finding that the subject-matter of claim 1 does not involve an inventive step.

4.7 The Board's understanding of the argument of the Examining Division that the principal distinguishing feature of claim 1 is obvious is as follows:

According to the electrical interconnect method taught in document D2, the non-wetting/wettable (i.e. nonsolderable/solderable) pad may be formed either on the integrated circuit or on the substrate (column 2, lines 57–67), with the interconnect ball (or "solder bump") then being formed on the pad (see e.g. figure 7).

However, document D2 also discloses details of prior art interconnect methods according to which "solder balls are formed on pads of either the integrated circuit, the substrate, or both" (column 1, lines 14–15). The final possibility ("or both") means that both the integrated circuit and the substrate have solder balls, and that the solder balls on one surface are brought into contact with the solder balls on the opposite surface during the registration step.

It would be obvious for the skilled person to adopt this prior art technique (providing solder balls on both surfaces) to the interconnect method taught in document D2 (having nonsolderable/solderable pads). The result would be an arrangement in which both the integrated circuit and the substrate have nonsolderable/solderable pads and associated solder balls, such that during the registration step each solder ball on the integrated circuit (i.e. the "surface mounted device") would contact a solder ball on the substrate. A solder ball on the substrate may be referred to as a "solderable conductor pillar", and hence, the skilled person would arrive in this manner at the principal distinguishing feature of claim 1 of the main request.

4.8 The Board is not convinced by this argument. According to the passage cited above describing the prior art (column 1, lines 14–15), "In general, solder balls are formed on pads of either the integrated circuit, the substrate, or both". There is, however, no explicit disclosure in this passage of two solder balls coming into contact at the registration step, and it is notable that in the more detailed description of the techniques of the prior art provided in Document D2 (column 3, lines 1–11), the only possibility referred to involves solder balls contacting pads on the opposite substrate and being bonded to them by reflow.

Hence, the position of the Examining Division appears to be based on a speculative interpretation of the cited passage, rather than on the actual subject-matter disclosed.

4.9 Moreover, even if this interpretation were accepted, it would still remain to be demonstrated why a skilled person would modify the teaching of document D2 by adopting this particular prior-art arrangement.

The reference to solder balls of the prior art being formed on both the integrated circuit and the substrate merely appears as one possibility among others in the context of a very brief summary of the prior art. There is nothing in this passage or elsewhere in the document indicating any particular advantage associated with such an arrangement, or any reason why a skilled person would be motivated to select it.

4.10 The technical advantage actually disclosed in document D2 is that the nonsolderable/solderable pads allow solder balls "of increased height to be formed" (column 6, lines 41 to 46).

The Board can accept, therefore, that if a skilled person were to consider putting solder balls on both surfaces, it would be obvious to also provide nonsolderable/solderable pads on both surfaces, as argued in the contested decision.

However, this observation is of no relevance for answering the question posed above, namely, why would a skilled person provide solder balls on both surfaces?

4.11 In view of the above considerations, no convincing reason can be seen why, starting from the electrical interconnect method taught in document D2, it would be obvious to a a skilled person to add the principal distinguishing feature of claim 1 of the main request, either on the basis of the objective problem mentioned above or for any other reason.

The Board therefore judges that the subject-matter of claim 1 of the main request involves an inventive step within the meaning of Article 56 EPC 1973.

4.12 In the light of this conclusion it is not necessary for the Board to discuss the other feature in which claim 1 differs from document D2 (i.e. the shape of the pillar).

5. As the Board is satisfied that the other requirements of the EPC are complied with, a patent may be granted on the basis of the appellant's main request.