Active solid-state devices (e.g. – transistors – solid-state diode – Combined with electrical contact or lead – Flip chip
Reexamination Certificate
2001-01-26
2003-06-03
Clark, Jasmine J B (Department: 2815)
Active solid-state devices (e.g., transistors, solid-state diode
Combined with electrical contact or lead
Flip chip
C257S696000, C257S693000, C257S780000, C257S789000, C439S066000, C439S067000, C439S077000, C439S078000
Reexamination Certificate
active
06573609
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to the art of electronic packaging, and more specifically to components useful for mounting and/or testing semiconductor chips and related electronic components. The present invention also relates to semiconductor chip assemblies and electronic devices incorporating such components.
2. Description of the Related Art
Modern electronic devices utilize semiconductor components, commonly referred to as “integrated circuits” which incorporate numerous electronic elements. These chips are mounted on substrates that physically support the chips and electrically interconnect each chip with other elements of the circuit. The substrate may be part of a discrete chip package, such as a single chip module or a multi-chip module, or may be a circuit board. The chip module or circuit board is typically incorporated into a large circuit. An interconnection between the chip and the chip module is commonly referred to as a “first level” assembly or chip interconnection. An interconnection between the chip module and a printed circuit board or card is commonly referred to as a “second level” interconnection. In “chip on board” packaging, the chip is mounted directly on the printed circuit board. This type of interconnection has been referred to as a “1½ level” interconnection.
One relatively common packaging scheme is called a “hybrid circuit”. A hybrid circuit typically contains a semiconductor chip that has been mounted and electrically interconnected to a circuit that has been formed on a thin layer of a rigid ceramic material. The method used to electrically interconnected the chip to the circuit is generally any of the methods that are known for use in first level bonding, such as wire bonding, tab bonding and flip chip bonding. In some cases it is desirable to mount and electrically interconnect the hybrid circuit to a printed circuit board. Solder is typically used to form the interconnection. It is difficult, however, to rework a hybrid circuit that has been soldered to a printed circuit board. In order to rework the assembly, the hybrid circuit must be removed from the printed circuit board. When the hybrid circuit is separated from the printed circuit board, part of the solder mass is removed from the contacts on the hybrid circuit. Non-uniform partial solder masses remain on the hybrid circuit contacts, the printed circuit board or both. When the hybrid circuit is resoldered to the printed circuit board, the non-uniform partial solder masses can cause short circuits and alignment problems.
Another problem associated with the assembly process is testing. In a typical assembly process, each hybrid circuit is tested before it is soldered to a printed circuit board. Testing involves clamping the hybrid circuit to a socket to engage the solder balls of the hybrid circuit with the test contacts of the test assembly. When the solder balls are engaged with the test contacts, the solder tends to creep and to deform, especially if the hybrid circuit is equipped with high-lead solder. The testing process, like the rework process, can lead to short circuit and alignment problems. To overcome these problems, it is desirable to use solid core solder balls to interconnect the ceramic substrate to a printed circuit board.
In U.S. Pat. No. 3,303,393, which issued on Feb. 7, 1967, Hymes et al. disclose a semiconductor chip assembly with flip-chip connections, which incorporates copper core solder balls. One solid core solder ball is provided between each contact on the chip and each contact pad on the substrate. Although these connections work well for small devices, with larger devices, the rigid connections provided by the solid core solder balls tend to crack at the soldered junctions between the balls and the opposing surfaces. Warpage or distortion of the chip or substrate, furthermore, make it difficult to engage all of the solid core solder balls between the chip and substrate simultaneously, or to engage all of the solid core solder balls between the chip and a test fixture. Although it is desirable to use solid core solder balls to interconnect a hybrid circuit to a printed circuit board, such an interconnection would be subject to similar problems.
The electrical power that is dissipated when a microelectronic device is in operation tends to heat up that device. When the device is no longer in operation, it tends to cool down. Over a period of time, the device tends to undergo a number of heating up and cooling down cycles as the device is repeatedly turned on and off. These cycles, which cause an associated expansion and contraction of the device, are commonly referred to as “thermal cycling”.
A device in which a hybrid circuit is bonded to a printed circuit board using solid core solder balls would be subject to substantial strain, caused by thermal cycling, during operation of the device. Electrical power dissipated within the hybrid circuit during operation would tend to heat up the hybrid circuit and, to a lesser extent, the printed circuit board. The temperature of the hybrid circuit, therefore, and, to a lesser extent, the printed circuit board would rise each time the device is turned on and fall each time the device is turned off. Since the hybrid circuit and the printed circuit board are normally constructed from different materials having different coefficients of thermal expansion, the hybrid circuit and printed circuit board would normally expand and contract by different amounts. This is commonly referred to as a “thermal mismatch”. The thermal mismatch causes the electrical contacts on the hybrid circuit to move relative to the electrical contact pads on the printed circuit board as the temperature of the hybrid circuit and printed circuit board change. The relative movement would deform the electrical interconnections between the hybrid circuit and the printed circuit board and place them under mechanical stress. Since these stresses would be applied repeatedly with repeated operation of the device, they would cause breakage of the electrical interconnections. Thermal cycling stresses may occur even where the hybrid circuit and printed circuit board are formed from like materials having similar coefficients of thermal expansion. This is because the temperature of the hybrid circuit may increase more rapidly than the temperature of the printed circuit board when power is first applied to the hybrid circuit. Unfortunately, solid core solder balls are neither flexible nor strong enough to withstand the strain generated by differential rates of thermal expansion.
Commonly assigned U.S. Pat. Nos. 5,148,265; 5,148,266; 5,518,964; 5,659,952; 5,929,517; 5,679,977; 5,685,885; 5,848,467; 5,852,326; 5,950,304; 6,133,627; 5,801,441; 6,104,087; 5,798,286; 5,830,782; 5,959,354; 5,913,109; 6,080,603; and 5,688,716; and U.S. patent application Ser. No. 09/271,688, filed on Mar. 18, 1999, the specifications of which are incorporated by reference herein, provide substantial solutions to the problems of thermal stresses and component testing. Nonetheless, still further improvement is desirable.
SUMMARY OF THE INVENTION
One aspect of the present invention provides a flexible chip carrier. The flexible chip carrier of this aspect of the present invention includes a rigid interposer having first and second surfaces. The rigid interposer is preferably adapted to mount and electrically connect a semiconductor chip onto the first surface of the rigid interposer. An interconnection between the rigid interposer and a semiconductor chip is a “first level” interconnection. The rigid interposer may be adapted to interconnect a semiconductor chip using any of the known methods of creating “first level” interconnections. Some conventional “first level” interconnection methods include wire bonding, tape-automated bonding and flip-chip bonding. The second surface contains a plurality of contacts disposed in a pattern. The area encompassed by the contacts is defined as a “contact pattern area”. The rigid interposer is p
Fjelstad Joseph
Myers John
Clark Jasmine J B
Lerner David Littenberg Krumholz & Mentlik LLP
Tessera Inc.
LandOfFree
Microelectronic component with rigid interposer does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Microelectronic component with rigid interposer, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Microelectronic component with rigid interposer will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3091402