The answer is yes. First, we need to define ‘packaging.’ I define packaging as every technology required between the IC and system. I often refer to packaging as the bridge between ICs and systems. I don’t view packaging as simply ‘interconnections,’ as we did in the past. It is a lot more than that. It is packaging the IC, without which the IC cannot be guaranteed to be good or used in systems. It also includes components. ICs and components are assembled onto system-level boards – requiring PCBs and assembly technologies such as SMT.
With this perspective then, the total packaging market is as big as semiconductors – about $150 billion. Some will argue that most of the current ICs are a commodity, and that the real value is in packaging, which includes not only IC/package co-design, but also miniaturized proprietary packages with ultra-small form factors and electrical test.
In addition, there is an emerging concept called SoP, system-on-package, being pioneered by Georgia Tech. In this concept, the value of ‘package’ becomes even more important. The basis of this concept is simply that while CMOS is good for certain functions such as logic and memory, it is not good for functions such as filters, antennas, capacitors, resistors, power amplifiers, switches, baluns, optical waveguides, surface acoustic, and bulk acoustic components.
The SoP concept eliminates the current set of problems such as bulkiness, lower performance, and lower reliability at systems level. The SoP, with its focus on miniaturization of components in thin film form at the microscale in the short term, and nanoscale in the long term, is expected to lead to digital convergence of computing, communication, and consumer products all in one package the size of a watch or a PDA.
Another reason for packaging to be an academic subject is the emerging digital convergence, taking packaging to another platform – a disruptive technology platform that requires integration of RF, optical, sensor, and digital by nano and MEMS technologies.
Why hasn’t packaging become an academic subject, while VLSI and microelectronics have throughout the world? The reasons have to do with the view that packaging isn’t important or science-based.
Companies like IBM viewed packaging as a critical strategic technology going back to the ‘70s, and most of their leading-edge computers had to depend on package integration of ICs by multichip modules to end up with the most powerful computers. This was also recognized by Japanese computer companies Hitachi, Fujitsu, and NEC in the ‘80s. At about this time, Semiconductor Research Corp. (SRC) was started and began to fund research in packaging at project levels at universities like Arizona, Cornell, MIT, and Standford. Subsequent major centers started at the University of Maryland, Arkansas, and Colorado, to name a few.
Then came Georgia Tech’s Packaging Research Center (PRC). With unparalleled 10-year funding from the National Science Foundation for a national engineering research center, the PRC has changed the academic landscape. The PRC pioneered the SoP paradigm and developed a comprehensive set of courses and curricula, leading to an educational track called ‘Devices, Integration, and Packaging,’ which involves 30 faculty and 250 students in four engineering disciplines: Electrical and Computer Engineering, Mechanical Engineering, Chemical Engineering, and Materials Science and Engineering. The nine courses at the undergraduate level and the 14 courses at the graduate level encompass systems packaging, electrical, and mechanical design, materials, processes, RF, optical, thermal, assembly, reliability, and electrical test. These led to packaging curricula in all of these disciplines. The number of students taking classes in these courses has averaged about 600 per year during the past 5 years.
Fundamentals of Microsystems Packaging is an undergraduate textbook that conveys the new packaging landscape. The book is being used by 51 universities around the world. Extensive research and hands-on educational facilities had to be set up to educate students in this new, package-centric view. So far, the PRC has graduated a total of 525 students with specialization in packaging. More than 75 universities around the globe offer research and educational programs, including in China, India, Singapore, Malaysia, Taiwan, Korea, Hong Kong, Japan, Germany, U.K., Sweden, France, and Netherlands, among others. The number of U.S. universities with packaging programs is also extensive.
I see packaging as an important technology to enable the expected digital convergence. In time, it will become a pervasive academic subject at all leading universities around the globe.
RAO TUMMALA, Ph.D., Pettit Chair Professor in Microsystems Packaging, may be contacted at Georgia Institute of Technology, 813 Ferst Drive, NW, Atlanta, GA 30332-0560; (404) 894-9097; e-mail: firstname.lastname@example.org.