Ex Parte Pocas et al

Patent Trial and Appeal Board

Jul 21, 2014

10584052 (P.T.A.B. Jul. 21, 2014)

UNITED STATES PATENT AND TRADEMARK OFFICE
____________________

BEFORE THE PATENT TRIAL AND APPEAL BOARD
____________________

Ex parte STEPHANE POCAS, HUBERT MORICEAU, and
JEAN-FRANCOIS MICHAUD
____________________

Appeal 2012-005770
Application 10/584,0521
Technology Center 2800
____________________

Before ADRIENE LEPIANE HANLON, CATHERINE Q. TIMM, and
GEORGE C. BEST, Administrative Patent Judges.

TIMM, Administrative Patent Judge.

DECISION ON APPEAL

STATEMENT OF CASE
Appellants seek review of the Examiner’s decision to reject claims 27-
29, 34-36, 41-43, 45, 53, 54, and 65 under 35 U.S.C. § 103(a) over Neilson2
in view of Kakumu3 and Yamamoto.4 We have jurisdiction under 35 U.S.C.
§§ 6(b) and 134(a).

1 According to Appellants, the real party in interest is Commissariat A
L’Energie Atomique. (Appeal Br. 2).
2 Neilson et al., US 6,054,369, patented Apr. 25, 2000.
3 Kakumu, US 5,654,241, patented Aug. 5, 1997.
4 Yamamoto et al., US 4,577,396, patented Mar. 25, 1986.
Appeal 2012-005770
Application 10/584,052

2
We REVERSE.

The claims are directed to a method of forming a metallic ohmic
contact at the interface between two semiconductor wafers. Claim 27 is
illustrative:
Claim 27. A method of sealing a first wafer and a second
wafer each made of semiconducting materials, comprising:
implanting a metallic species in at least the first wafer at
a dose above 1016 species/cm2,
assembling the first wafer and the second wafer by
molecular bonding, and
after the molecular bonding, forming a metallic ohmic
contact including alloys formed between the implanted metallic
species and the semiconducting materials of the first wafer and
the second wafer, said metallic ohmic contact being formed at
an assembly interface between the first wafer and the second
wafer,
wherein the forming includes causing the implanted
metallic species to diffuse towards the interface between the
first wafer with the second wafer and beyond the interface.
(Claims App. at Br. 17.)
OPINION
While Neilson teaches implanting a metallic species (dopant) into a
wafer, assembling two wafers by molecular bonding, and causing the
metallic species (dopant) to diffuse, Neilson does not do so to form an ohmic
contact. Instead, Neilson performs the method to form a particular density
of recombination centers in a buffer layer 24 (Neilson, col. 3, ll. 47-57; col.
4, ll. 36-55; see also, e.g., Fig. 2b (showing recombination centers (x), e.g.,
dopants, at a wafer-to-wafer interface 20 concentrated in buffer layer 24)).
Appeal 2012-005770
Application 10/584,052

3
Specifically, Neilson discloses doping and high temperature treating to
distribute the dopants in buffer layer 24 at a concentration in the range of
1014 species/cm3 to 1019 species/cm3 (Neilson, col. 4, ll. 50-52).
The Examiner acknowledges that Neilson neither discloses implanting
the metallic dopant at a dose above 1016 species/cm2 or forming a metallic
ohmic contact including alloys formed between the dopants and
semiconducting materials of the wafers (Ans. 6-8). However, according to
the Examiner, it would have been obvious to use the implant dosage of
Kakumu “to form areas of reduced resistance to carrier flow” and to use the
implant species and dosage of Yamamoto “to form alloyed or silicide areas
of reduced resistance to electrical contacts” (Ans. 6-7).
The problem is that a reason to dose the buffer layer 24 of Nielson to
form alloys that function as ohmic contacts at the region of the interface
between the wafers, i.e., within the buffer layer, has not been established.
Kakumu forms titanium-silicide layers 18 in the areas of the source and
drain, and on the gate electrodes of a transistor (Kakumu, col. 3, ll. 28-38;
Fig. 2D). Yamamoto dopes a surface region of a silicon substrate so that a
silicide or silicon alloy is interposed between an electrode or wiring and the
silicon substrate in order to reduce peeling of the electrode or wiring from
the substrate (Yamamoto, col. 2, ll. 25-33). Neither Kakumu nor Yamamoto
suggests forming ohmic contacts in a buffer layer, and the Examiner
provides no convincing reasoning based on knowledge within the prior art
indicating that forming ohmic contacts within Neilson’s buffer layer would
have been contemplated by one of ordinary skill in the art at the time of the
invention. Therefore, we agree with Appellants that the Examiner’s
conclusion of obviousness is based upon impermissible hindsight.
Appeal 2012-005770
Application 10/584,052

4
CONCLUSION
We do not sustain the Examiner’s rejection.
DECISION
The Examiner’s decision is reversed.

REVERSED

cdc