PREVIOUS ADVICE


Dear GrantDoctor,My area of research is VLSI [very large scale of integration, the process of placing more than 100,000 transistors on one chip]. What is the situation regarding research grants and research projects from industry as far as my field of interest is concerned? Can you tell me what research areas in electrical engineering departments have good grants?Thank you.Nitin

Dear Nitin,

In 1965 the physical chemist Gordon Moore of Fairchild Semiconductor published a paper in which he observed that the density of transistors on integrated circuits roughly doubles every couple of years, and that the trend was likely to continue. His observation, later dubbed Moore's law, has proved remarkably reliable for 38 years, but some formidable barriers are rapidly approaching. As Moore observed in a presentation earlier this year, the exponential increase in device performance can't go on forever, but "forever can be delayed indefinitely." That pretty much sums up the state of modern VLSI research.

As devices get smaller and faster, interconnects get more problematic, due to interconnect delay and cross-talk noise. This, apparently, is one of the most important current performance limitations for VLSI circuits. If the astonishing rate of improvement in computer performance, measured in both speed and density, is to continue, these issues will need to be addressed. One approach is to reduce delays and noise; another is to increase the tolerance of circuits to noise and errors. Both approaches are being investigated--and funded--energetically. This, of course, is just one (or maybe two) important area(s) of research; there are many others. If you want more details you'll need to find another source, since the GrantDoctor is a general practitioner, not a specialist. You could, for instance, search the National Science Foundation's (NSF's) recent awards on Fastlane. Be sure to specify a date range; otherwise you may end up studying abstracts that are several years old--ancient history in VLSI terms.

You ask what areas of research are getting "good grants" (they're all good, aren't they?): Major U.S. federal supporters of research for single-PI research grants in microelectronics include the Electronics and Communications Systems division of NSF's Computer and Information Science and Engineering (CISE) directorate, especially via their Electronics, Photonics, and Device Technologies program. NSF's Computer Communications Research division of CISE also supports microelectronics research. The Mathematical and Physical Sciences directorate supports research on microelectronic materials and materials processing via their Division of Materials Research.

Another important federal player is the Department of Defense, including the Defense Advanced Research Projects Agency ( DARPA), the Army Research Office, the Air Force Office of Scientific Research, and the Office of Naval Research. Microelectronics research at DARPA is sprinkled among several different programs. If you intend to apply to DARPA, however, be forewarned: These days DARPA isn't interested in meek and mild science. In an effort to recover some of their lost glory, they're talkin' 'bout revolutions. Here's how they put it: "Specifically excluded is research that primarily results in evolutionary improvements to the existing state of practice." Although it's not directly related to your question, I'll mention that one area DARPA is interested in is what they call "macroelectronics"--circuits that are integrated, yet distributed widely over a large area, preferably flexible substrates ... an interesting twist on microelectronics research. (Reading about DARPA is always fun.)

Your question gives me an excuse to write about how microelectronics research is getting done these days. It's an interesting subject, because this is an area where industry-university partnerships are flourishing, especially overseas. Although there are still plenty of university-based researchers doing VLSI work--in departments of physics, engineering, chemistry, and materials science--this is one research sector where cooperative ventures between research and industry work very well, especially in those locations and fields where security concerns don't inhibit the sharing of ideas.

Cooperation between industry and academic science can be very effective, but in fields such as biomedical research it can also raise ethical questions. There have been some very effective university-industry collaborations in the biomedical sciences, but some of them have been controversial. In an area like microelectronics research there are fewer ethical concerns, but in the United States, at least, national security concerns have made some sensitive defense-related research incompatible with the open structure of academic research. Perhaps that's why many of the most interesting examples of academic-industrial cooperation in microelectronics are in Europe.

Chief among these is Europe's largest research center for microelectronics research, the Interuniversity Microelectronics Research Center ( IMEC) in Leuven, Belgium. IMEC is a fascinating operation, with a forward-looking "business model" that blends elements of industry and academia. They cooperate intensively with local university researchers, but they get most of their funding from industry. Their goal is to generate new technologies that will become mature commercial products 3 to 10 years down the road. When, in their view, a new product is ready for commercialization, they first seek an industrial partner and then, if they can't find one that's suitable, they "spin off" a new company. One of the business-model innovations that contributes to IMEC's success is its model for sharing intellectual property (IP) among industrial affiliates; affiliate companies get access to IMEC background information, as well as to certain categories of IP developed not just by IMEC but by its other industrial affiliates.

IMEC's financial performance is impressive; in a terrible business climate for the computer industry, the center has kept growing, last year increasing total revenues by 15% and (corporate) contract revenues by 25%. Last year IMEC also filed 43 patents, and its scientists published more than 1150 papers in journals and conference proceedings. IMEC receives only about a quarter of its 138-million-euro budget from government, the balance coming from contracts with industry, including regional companies (30.5%) and international companies (48.5%).

One interesting consequence--and evidence--of IMEC's unique approach can be seen its employee profile: While 85% of its operations are R&D related, only 10% of IMEC employees have PhDs.

Be Well,

The GrantDoctor

Due to the high volume of questions received, The GrantDoctor cannot answer all queries on an individual basis. Look for an answer to your question published in this column soon! Thank you!