Semiconductor device manufacturing: process – Packaging or treatment of packaged semiconductor
Reexamination Certificate
1999-09-13
2001-08-28
Niebling, John F. (Department: 2812)
Semiconductor device manufacturing: process
Packaging or treatment of packaged semiconductor
C438S108000
Reexamination Certificate
active
06281039
ABSTRACT:
TECHNOLOGICAL FIELD AND PRIOR ART
This invention relates to the field of electronics and to the field of assembling two electronic components coming from different sources (different substrates).
It also relates to the generation of electrically active functions (of the diode type, for example) by an assembly operation.
Possible examples of applications are those of infra-red wave detectors, but other types of components can be constructed by this technique (for example: detectors in the visible range, electroluminescent diodes, etc.).
Two families of hybrid opto-electronic devices involving the transfer of epitaxial layers are currently being studied, using two assembly techniques:
the Epitaxial Lift-Off(ELO) technique,
the hybridization by bead (FLIP-CHIP) technique.
French patent FR-2 715 002 shows how an adapted FLIP-CHIP technique allows one to transfer and inter-connect epitaxiated layers onto a silicon reading circuit.
These techniques are aimed at carrying out the assembly of epitaxiated layers with a silicon circuit. For example, one technology is produced on a slice of silicon while one technology is produced on a slice of another material, for example AsGa.
The epitaxy produced on the AsGa substrate is transferred onto the silicon substrate. The inter-connection of these two components is provided by bringing into contact metal coatings created on each of the components;
In the case of the ELO technology, this electrical inter-connection is provided by an extra technological operation after the Si assembly operation. In
FIG. 1
, a reading circuit
2
made of silicon supports a thin epitaxiated layer
4
, made of another material. Reference numbers
6
,
10
designate the metal contacts. The contact pick-up
8
is made through the upper face of the epitaxy.
In
FIG. 2
, the electrical connection is provided by metal contacts
12
,
14
. In other words, this connection is provided by a contact and a metal-metal annealing operation on the surfaces facing one another, a-if the contact pick-up is through the surface facing the silicon circuit.
In the case of FLIP-CHIP technology, inter-connection is provided by beads
20
of the tin-lead or indium type that connect the metal contacts
22
,
24
(FIG.
3
).
In all cases, the components must be assembled with extreme precision in such a way that the technologies on the components to be joined coincide with the technologies present on the receiving components. In particular, as illustrated in
FIG. 4
, the spacings dx, dy (in two directions) between two metal contacts
28
,
30
, must be less than 5 &mgr;m, or better.
The U.S. Pat. No. 5,536,680 proposes the production of an infra-red detector by creating diodes at the moment of assembly: This technology is called SAB or Soft Aligned Bonding.
An epitaxiated layer is placed facing a silicon circuit
2
covered with adhesive
32
and fitted with micro-points
34
(FIG.
5
). A pressure is exerted between the two circuits
2
,
4
and the micro-points create a zone N in the material P opposite.
As already mentioned above, the two technologies ELO and FLIP-CHIP require the components to be perfectly aligned during assembly. The specifications generally required are accuracies better than +/−5 &mgr;m and can even reach +/−1 &mgr;m. This assembly operation necessitates expensive equipment.
SAB technology is a self-aligning technique.
The two technologies ELO and FLIP-CHIP require the carrying out of technologies on both components to be assembled, which are specific to the assembly operation:
Production of contact metal coatings on both sides, in all cases (see
FIGS. 1
,
2
,
4
)
Production of a technology, after the assembly operation in the case of ELO technology (FIG.
1
).
Production of a technology of solder beads in the case of FLIP-CHIP technology.
SAB technology does not require any detector side technology. It requires a technology on the reading circuit side (micro-points).
Finally, the two technologies do not easily allow the production of components of large size or of a small pitch.
ELO technology requires the handling of layers of a few micrometers thickness which are therefore stretchable. It is impossible therefore to guarantee an alignment better than a micrometer over a length of several centimeters.
FLIP-CHIP technology requires the production, for example, of bead technology within a pitch of 10 &mgr;m or less, which implies variations in flatness of the components facing one another which are extremely low.
On the other hand, for a given pitch and a given number of points, differences in the expansion coefficients of the components to be assembled mean that it is impossible to have components face to face at the hybridization temperature: the contact areas cross at the hybridization temperature and are no longer facing one another for the soldering.
As an example one may take the production of an infra-red detector of 2000×2000 points with a pitch of 20 &mgr;m. This leads to components of a size 40×40 mm over the photo-detection area.
Over such a size, the hybridization techniques mean that thermal expansion between detection circuits (for example a CdTe substrate) and reading circuits (for example made of silicon), at the end of a diagonal, is greater than 18 &mgr;m at the hybridization temperature, which forbids the use of this technique for the assembly.
Similarly, it is impossible to align a thin stretchable layer with sufficient precision on components of this size (ELO technology).
As for SAB technology, it enables the problems of alignment and large size to be resolved but there are other disadvantages as follows:
The micro-points must all have the same simultaneous pressure for the creation of diodes. This implies extreme flatness for the components and complex fittings for parallelism during application of the pressure.
This technique is only suitable for materials where an area of stress leads to the creation of a zone of opposite type to that where it is exerted (N on P or P on N).
This is true for compounds of the CdHgTe (CMT) type but it is not true for all the materials where a zone of stress does not generally correspond to the creation of a diode.
In the case of CdHgTe, with this technique, it is known not to use a P type ohmic contact, the micro-point automatically creating an N type zone.
DESCRIPTION OF THE INVENTION
The subject of the invention is a method of assembling first and second electronic components, the first of the components being fitted with conductive connection contacts comprising at least one material capable of diffusing into the second component, the method comprising:
a step of assembling two components, and
the production of at least one electrically active zone in the second component, by diffusion of said material for connection contacts from the first component to the second component.
The technique according to the invention allows one:
to avoid the carrying out of an extra diffusion or implantation operation into the second component,
to avoid any problem of alignment during assembly of the components,
to avoid the flatness constraints (FLIP-CHIP method),
to produce components of very large size and of very small pitch because of the lack of constraints on flatness and alignment,
to reduce the manufacturing costs in a radical way,
to create all kinds of diodes in relation to the choice of material for the connection contacts.
In particular, the connection contacts can be made of a doped metal, the doping agent diffusing into the second component.
This doping agent can be, for example, indium and/or arsenic and/or antimony and/or mercury.
It is possible to produce the electrically active zone and to carry out the assembly step simultaneously.
According to one example, the first and second components can be respectively, a reading circuit and an infra-red detection circuit, for example made of CdHgTe.
Another subject of the invention is a hybrid device comprising first and second assembled components, the first component being fitted with at least one connection contact, an electr
Commissariat A l'Energie Atomique
Jones Josetta I.
Niebling John F.
Oblon & Spivak, McClelland, Maier & Neustadt P.C.
LandOfFree
Hybrid device and a method of producing electrically active... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Hybrid device and a method of producing electrically active..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Hybrid device and a method of producing electrically active... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2522284