Active solid-state devices (e.g. – transistors – solid-state diode – Encapsulated – With specified encapsulant
Reexamination Certificate
2000-06-14
2004-01-06
Thomas, Tom (Department: 2811)
Active solid-state devices (e.g., transistors, solid-state diode
Encapsulated
With specified encapsulant
C257S787000, C257S788000, C257S795000, C257S778000, C257S782000, C257S737000, C257S738000, C257S780000
Reexamination Certificate
active
06674178
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a semiconductor device having a structure whereby a direct connection is made between an interconnection board and a semiconductor pellet.
2. Background of the Invention
To improve their portability, there is a desire to make compact electronic circuit devices for video cameras, notebook-size personal computers and the like smaller and lighter, and the achievement of smaller electronic components or electronic components of the same dimensions but a higher level of integration used therein contributes to the achievement of more compact and lightweight electronic circuit devices.
In semiconductor devices, while there has been reduction of the size of a semiconductor pellet and an increase in the level of integration, efforts are not limited to these improvements, and there have been direct connections made from a semiconductor pellet to an interconnection board, thereby improving the mounting density.
An example of this is illustrated in FIG.
9
. In this drawing, the reference numeral
1
denotes a semiconductor pellet, which has a large number of bump electrodes
3
formed on a main surface of a semiconductor substrate
2
onto which are large number of electronic elements (not shown in the drawing) are formed. These bump electrodes
3
are generally formed by solder or gold.
The reference numeral
4
denotes an interconnection board, formed by a conductive pattern (not shown in the drawing) of copper formed on a surface of a insulating substrate
5
, the conductive pattern being then covered by a photoresist film (not shown in the drawing). Windows are then formed in the photoresist film at positions opposite the bump electrodes
3
, so as to expose part of the conductive pattern, thereby forming pad electrodes
6
at the windows. These pad electrodes
6
are generally made of a copper foil of thickness 12 to 18 &mgr;m, over which a nickel plating is covered to a thickness of 3 to 5 &mgr;m, and further over which gold is laminated to a thickness of 0.03 to 1 &mgr;m. In the drawing, however, this is shown as a single layer.
The reference numeral
7
denotes a resin that makes a mechanical connection between the semiconductor pellet
1
and the interconnection board
5
, and which protects the interconnection pattern (not shown in the drawing) on the semiconductor pellet
1
from external corrosive gases.
The above-described structure is widely known, and is disclosed in Japanese Unexamined Patent Application publication S63-241955 (prior art 1), U.S. Pat. No. 5,795,818 (prior art 2), Japanese Unexamined Patent Application publication S60-262430 (prior art 3), and Japanese Unexamined Patent Application publication H9-97816 (prior art 4).
Of the above-noted prior art examples, in prior art 1 the bump electrode
3
is formed by solder, and the semiconductor pellet
1
is flip-chip connected to the interconnection board
4
.
In the above-noted prior art 2, there is language describing the application of a load of approximately 20 g per bump electrode
3
to the pad electrodes
6
, the semiconductor pellet
1
being heated to 240° C., and the interconnection board
4
being heated to 190° C., with ultrasonic vibration also being applied to the electrodes
3
and
6
so as to make a metal-to-metal joint, and another example noted is that in which a load of approximately 10 g per bump electrode
3
is applied, the semiconductor pellet
1
being heated to 180° C. and the interconnection board
4
being heated to 190° C., with ultrasonic vibration also being applied to the electrodes
3
and
6
so as to make a metal-to-metal joint.
In both prior art examples 1 and 2, after making a mechanical and electrical joint between the semiconductor pellet
1
and the interconnection board
4
, a resin
7
is supplied between the semiconductor pellet
1
and the interconnection board
4
as an adhesive. Because there is a tiny spacing of approximately 100 &mgr;m between the semiconductor pellet
1
and the interconnection board
4
, however, it is difficult for the resin
7
to enter therebetween, and even if it does enter therebetween, there is the problem of a tendency for air bubbles to remain.
Prior art 1 is such that air bubbles do not remain in the resin
7
, but the work involved is troublesome.
In prior art 2, while it can be envisioned that there would be a tendency for air bubbles to remain in the resin
7
, there is absolutely no language with regard to how these air bubbles are removed.
In contrast to the above, in prior art examples 3 and 4, a resin
7
is supplied to a region that includes the pad electrodes
6
on the interconnection board
4
beforehand, a semiconductor pellet
1
being supplied to the top of this resin
7
and pressure applied thereto, so that the pad electrodes
6
and the bump electrodes
3
are superposed, thereby forcing the resin
7
from between the bump electrodes
3
and the pad electrodes
6
and making an electrical connection between the electrodes
3
and
6
. The applied pressure is maintained in this condition, and resin
7
is cured. After it hardens sufficiently, even if the applied pressure is released, the resin
7
between the semiconductor pellet
1
and the interconnection board
4
makes a mechanical connection, and the pressure between the electrodes maintains the electrical connections.
In the above-noted semiconductor device, because the resin
7
supplied to the top of the interconnection board
4
beforehand is pressed and spread out, it is difficult for air bubble to remain within the resin
7
, thereby solving the problem that remained with the prior art examples 1 and 2.
In the prior art example 3, there is language to the effect that when a light-curable resin is used there is absolutely no heat applied to the semiconductor pellet
1
and the interconnection board
4
, and even when using a thermally curable resin, the curing temperature is raised to no more than 150° C., the result being that it is possible to reduce the thermal distortion of constituent materials, and possible to achieve a connection having high reliability.
In the prior art example 4 as well, there is language to the effect that, although a thermally curable resin
7
is supplied as pressure is applied to join the bump electrodes
3
and the pad electrodes
6
, by using a resin
7
that has a curing rate of contraction that is larger than the coefficient of thermal expansion, even in a high-temperature environment because the curing rate of contraction exceeds the coefficient of thermal expansion, force does not act to peel the bump electrodes from the pad electrodes, so that the connection does not become unstable, and language to the effect that, because the end of the bump electrode is in point contact with the pad electrode
6
with pressure applied therebetween, there is further broadening so as to achieve a surface contact, the result being that the resin
7
between the electrodes is driven out from the contacting parts, thereby enabling the achievement of a reliable contact, with no included impurities.
In this manner, in a semiconductor device according to the prior art examples 3 and 4, even after compression deformation is caused within the elastic limit of the bump electrodes
3
superposed with the pad electrodes
6
as the resin
7
is cured, the pressurized contact between the electrodes
3
and
6
is maintained.
The coefficients of thermal expansion of the semiconductor device elements such as semiconductor substrate
2
, bump electrodes
3
, insulating substrate
5
, and pad electrodes
6
, in a semiconductor pellet
1
which is based on silicon and an epoxy resin based interconnection board
4
, are 2.4 PPM/° C., 15 PPM/° C., 16 PPM/° C., and 20 PPM/° C., respectively, and because the coefficients of thermal expansion of the insulating substrate
5
, the pad electrodes
6
and the bump electrodes
3
are similar, differences in length caused by thermal expansion do not become a problem.
Although there is a large difference between the coefficients of therm
NEC Electronics Corporation
Parekh Nitin
Thomas Tom
LandOfFree
Semiconductor device having dispersed filler between electrodes does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Semiconductor device having dispersed filler between electrodes, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Semiconductor device having dispersed filler between electrodes will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3246917