A method of fabricating flexible assemblies comprising flexible integrated circuits bonded onto or into flexible membranes has been developed. The method provides for bonding of thinned (more specifically, thin enough to be flexible) integrated-circuit chips to the membranes and for electrical connection of the integrated circuits to other circuitry on or in the membranes. The method is expected to enable the further development of a variety of membrane-based flexible, lightweight electronic systems and assemblies — for example, phasedarray antenna assemblies comprising integrated-circuit transmitting/ receiving (T/R) modules further integrated with arrays of transmission lines and antenna radiator elements.

The conventional method of integrating T/R modules and other electronic circuitry into phasedarray radar antenna assemblies involves bolting or epoxying the modules to rigid boards and making the electrical connections by soldering the package leads to contact pads on the boards. Recently, flexible membrane materials have been used to make lighter-weight antenna assemblies. In the present method, in order to further reduce the weight of a given antenna assembly, the T/R circuitry is miniaturized and integrated into one thinned device or a few thinned devices and, instead of packaging the integrated circuits and soldering the packages to contact pads on the membrane, one utilizes the so-called flipchip- on-flex technology to directly integrate the thinned dice with the membrane and the circuitry on or in the membrane. Flip-chip-on-flex technology is not new, but the application of it to thinned integrated circuits is new.

A Flexible Integrated-Circuit Device is electrically and mechanically bonded to copper contact pads on theopposite side of a flexible membrane in either of two variations of the method described in the text.
The method admits of two variations (see figure). In the first variation, each flexible integrated circuit is placed on one side of a flexible polyimide membrane with solder bumps on its contact pads intruding into vias in the membrane. This arrangement helps to minimize the overall thickness of the completed assembly. Through reflow of the solder bumps, the die is electrically and mechanically bonded to copper contact pads that are parts of circuitry on the opposite side of the membrane. A fixture is needed to maintain the die and membrane flat during reflow. Reflow is followed by dispensing and curing of an underfill (a layer of polymer to fill any remaining gap between the membrane and the die). The underfill-dispensing process must be carefully controlled to prevent flow of the underfill material onto the outer surface of the die.

In the second variation, the bumps on the contact pads on the die are made of gold (instead of solder) and flexible membrane is made of a liquid-crystal polymer (instead of a polyimide). As in the first variation, the die is placed on one side of the membrane with its bumps intruding into vias in the membrane. Then in a thermocompression process, the die becomes bonded to the membrane and the bumps become bonded to copper contact pads on the opposite side of the membrane. In this case, no underfill is needed.

The method as described thus far can be extended to enable embedment of a die between two flexible membranes to form a laminated flexible membrane containing flexible electronic circuitry.

This work was done by Alina Moussessian and Linda Del Castillo of Caltech and R. Wayne Johnson of Auburn University for NASA's Jet Propulsion Laboratory. For more information, contact This email address is being protected from spambots. You need JavaScript enabled to view it..