Slow but steady is the hallmark of progress in the three foundation technologies of components, interconnects, and packaging. At first glance, to the untrained eye, advances in these areas seem mundane and obvious. But the facts are that size is always shrinking, performance is constantly improving, and cost keeps going down. Furthermore, these advances apply for many components--including resistors, capacitors, inductors, circuit-protection devices, switches, and relays--and they apply in equal measure to interconnects and packaging. But while technological progress often is incremental, the effort needed to get to these levels isn't trivial. Neither are the benefits accrued from making the devices that much smaller, faster, or less expensive. When it comes to progress in these three areas, the old adage applies: The whole is greater than the sum of the parts.
That's because so many of the major advances in application development depend on the modest and not-so-modest gains made outside the IC area. The strides taken in packing more functionality onto the IC could be negated by placing the chip on a board with oversized passives.
Higher I/O counts on-chip would lead to massive ICs and pc boards if semiconductor packages, connectors, and pc boards could not be fabricated with finer pitch interconnects. The migration to finer process geometries in silicon would be halted if circuit-protection devices weren't available to stop ESD. Moreover, rising clock and bus speeds could be useless if adequate filtering and shielding weren't available to quash EMI, while maintaining signal integrity throughout the system.
These issues underlie a host of looming IC and system-level developments. Taking the circuit-protection issue as an example, it's clear that the various high-speed bus protocols like USB 2.0 will need low-capacitance ESD protection and EMI filtering in small, functionally integrated components. These parts will be critical to the reliability of the system. However, performance concerns exist with the polymer and semiconductor-based ESD protection devices that have been developed to this point. More work must be done to ensure the successful implementation of many high-speed bus designs.
Never-before-seen circuit-protection components also may pop up this year. A new category of products known as arc fault circuit breakers are just now being developed and tested. These devices will detect and locate the arcing that threatens wiring in airplanes and other vehicles. Their successful development could have a major impact on the safety and reliability of transportation.
Other component areas will reveal new "building blocks" for emerging industries. Consider the supercapacitor, also known as an ultracapacitor or an electric double-layer capacitor. Able to provide pulses of high power, supercapacitors can satisfy the high current demands that batteries alone cannot.
For example, there are 5- to 10-F parts in a 4- by 15- by 10-mm form factor. These supercapacitors can deliver 3- to 5-A peaks from a battery that's limited to 1-A peaks. This capability is being exploited in handheld devices like MP3 players. With inevitable cost reductions, these parts will find their way into more and more applications.
One of the most promising apps involves the hybrid electric vehicle, which will include cars, trucks, and buses. The supercaps needed for these applications exist but are currently too expensive. Yet that situation is changing as supercapacitor manufacturers dramatically cut component cost by ramping up supercapacitor production, while developing new materials. Within just a few years, very large supercapacitors will become viable for use in many types of hybrid electric vehicles.
Another component that's about to make a tremendous impact is the formerly humble LED. The ongoing development of high-brightness LEDs will see these components replacing incandescent signals and light sources in an ever-growing variety of applications. Before the decade is over, LED light sources should make great inroads in general lighting applications, where their high efficiencies and long operating life will change the face of lighting in ways we're just starting to contemplate.
The developments made in packaging and interconnect will be less dramatic, but they're no less important. Packaging and interconnect technologies constantly face the challenges of handling smaller yet denser ICs, complicating the task of bringing out more and more signal lines from smaller chip packages. Three-dimensional packages that feature ICs stacked atop one another and the thinning of the silicon wafer are just two notable efforts in this area. But there are also limits.
Stacking chips in the z direction is limited by the need for smaller heights on the pc boards they're stacked on. And in the lab, wafers can only be thinned (using either chemical etching or mechanical grinding) to about 10 µm before breakage and handling problems set in. Clearly, radically new packaging and interconnect methods are essential.
There's a need for another paradigm shift. New materials for more advanced array packages are being investigated to keep pace with rising pin counts in smaller spaces and deal with higher levels of heat dissipation. On-chip interconnects using light signals instead of metal wires offer another option. Some advocate so-called system-on-a-package (SoP) or advanced system-on-a-chip (SoC) devices. These approaches may all be required to meet future packaging and interconnect demands.