VIEWPOINT:
Technology Roadmaps for Compound Semiconductors: Knowledge Based Investments for the 21st Century
Published in IEEE Spectrum, January 1999, pp. 54-55
Imagine a world without desk-top laser printers; without cellular telephones and pagers; without global positioning; without CD players CD-ROMs, and DVD players; without laser fax machines; and without today's combinations of information technologies that together comprise Information Superhighways. Compound semiconductors such as gallium arsenide (GaAs) and indium phosphide (InP) have brought about or enabled all of these microelectronic, optoelectronics, and wireless marvels.
Compound semiconductors are critical to the success of many technologies that have opened new markets. Whereas, the total compound semiconductor market a few years ago was measured in hundreds of millions of dollars, today it is measured in billions of dollars. Some of the most common compound semiconductor devices are based on GaAs substrates for such applications as high speed digital electronics, high frequency analog electronics, lasers, light emitting diodes, and power transistors. Other commonly used compound semiconductor devices are based on InP substrates that are used in optical communications systems. Today's markets that have high-growth rates for using compound semiconductors in opto-electronic and wireless systems for civilian applications are relatively new phenomena. They are explained, in part, by the technologies becoming more mature.
In the United States, much of the prior development of compound semiconductor materials was to meet demands for defense applications. The recent U.S. decrease in R&D support for defense applications of compound semiconductor materials has encouraged some companies to pursue commercial markets more aggressively than they had in the past. Their actions also may have contributed to the high-growth markets mentioned in the first two paragraphs.
Today, compound semiconductor companies compete on technology, fabrication, and design. This mode of competition among scaled silicon CMOS manufacturers is changing because the research and development costs associated with advanced larger wafer sizes and smaller linewidths for CMOS are too great for any one company or country to accept. The competitiveness among CMOS manufacturers is shifting from an emphasis on technology and fabrication to a much greater emphasis on product design and architecture. A similar shift may occur for one or two major applications of compound semiconductors; particularly, in applications for which compound semiconductors and elemental silicon co-exist.
We can expect many more innovations based upon compound semiconductors in the coming years. Compared to the field of elemental semiconductors, which is dominated by scaled silicon complementary metal oxide semiconductors (CMOS), the field of compound semiconductors offers greater opportunities for innovations by combining microelectronic, optoelectronic, and microwave or millimeter-wave technologies and by permitting many more engineering choices to exploit new effects in advanced heterostructures, double-heterojunction bipolar transistors, psuedo-morphic high electron mobility transistors, and the like. Expected high-growth markets for devices and systems based on compound semiconductors include: (1) satellite communications systems that are fundamentally dependent on the availability of compound semiconductor microelectronics; (2) the introduction of powerful and more-efficient semiconductor laser replacements for conventional solid-state lasers; (3) future deployment of wireless real-time, digital video information networks based in part on mobile digital receivers that contain high resolution analog-to-digital converters made from compound semiconductor circuits; (4) intelligent cruise control of cars and trucks that rely on compound semiconductors in radars; (5) electronic distribution of "movies;" (6) replacement of many incandescent and fluorescent lights, particularly those in locations that are difficult and expensive to access; and (7) high resolution displays based on LEDs and compound semiconductor lasers.
The future trends in microelectronics suggest that the infrastructure for consensus-based planning in the compound semiconductor industry needs to be strengthened. Such coordinated planning involves many stake holders concerned with materials, processes, devices, circuits, interconnects, packages, and systems and is expected to increase the probability for successful, knowledge-based investments of resources that will become more limited. As is now the case for scaled elemental silicon CMOS, and should be the case for selected applications of compound semiconductors, international technology plans emphasize technical barriers and needs. The organizations that participate in such planning are free to develop their individual internal plans and priorities to address potential solutions as they may desire.
The International Technology Roadmap for Semiconductors [ITRS, formerly, the National Technology Roadmap for Semiconductors (NTRS)] treats mainly silicon CMOS. It offers a template for the compound semiconductor industry to utilize, if it chooses, on how to proceed in developing its international infrastructure for consensus-based planning. Because there are many diverse markets for compound semiconductors, any technology roadmap for compound semiconductors has to be selective. For example, a portion of the compound semiconductor industry may want to do consensus-based planning for those markets in which compound semiconductors and silicon co-exist and thereby compete for market share. One example would be in systems that use radio frequency (RF) integrated circuits (ICs) operating at or above 1GHz. Synergisms, similar to those that exist today between the ITRS/NTRS and silicon based RF ICs, would benefit the RF IC compound semiconductor industry if it had a technology roadmap. Another example would be systems that use high-speed digital circuits.
Possible initial action items for progress towards a technology roadmap for compound semiconductors include:
- Assess which applications are most amenable for developing an international technology roadmap for compound semiconductors (ITRCS) to support an enhanced global infrastructure.
- Identify which organizations will sponsor the development and maintenance of the proposed ITRCS for the one or a few major applications of compound semiconductors and therefore speak for that portion of the compound semiconductor industry.
- Perform economic assessments of the benefits and costs associated both with and without developing roadmaps for those few selected applications.
- Convene workshops that identify the key system applications, technical and commercial problem areas (e.g., cost reduction for epi-layer applications, millimeter-wave circuits, thermal management, and packaging and mounting of radio frequency devices), and the technology performance gaps between what is or will be available and what the intended market application requires.
*******
Herbert S. Bennett, Ph.D., (F) is a NIST Fellow at the National Institute of Standards and Technology, Gaithersburg, Maryland. He is Chairman of the Compound Semiconductor Integrated Circuits Technical Committee of the IEEE Electron Devices Society. This viewpoint is based partly on discussions with members of that Committee and does not necessarily represent the views or policies of any of the supporting host institutions or of the IEEE. A contribution from the U.S. National Institute of Standards and Technology, not subject to copyright.