Comment on this story

The UT/ORNL Joint Institutes could vault Tennessee into the ranks of the nation’s top research universities

by Joe Sullivan

A recently completed, three-story red brick building commands a central location on the Oak Ridge National Laboratory campus; but it doesn’t belong to ORNL. Rather, the $10 million new facility was built by and belongs to the University of Tennessee. Although it’s fully furnished, this Joint Institute for Computational Sciences, JICS as it is known, is mostly uninhabited.

The challenge facing the university now is to fill the new facility with top scientists to conduct research that takes advantage of what ORNL brings to the undertaking: namely, the world’s most powerful computer. It’s being assembled in an adjacent new building. Already searches are underway for eight newly-created UT faculty positions toward a goal of 20 or more over the next three years. Each of them is expected to bring along teams of doctoral students and what are known as “post-docs” to support their work so that JICS will have total complement of some 100 researchers, spanning an array of scientific disciplines. Anticipated federal grants will cover much, but by no means all, of the cost of heretofore-incalculable work on everything from gaining a better understanding of the earth’s climate to harnessing new sources of energy.

Launching JICS alone will represent a major leap forward for UT as a research university—an area in which the university presently lacks renown (although it has a number of faculty members who are highly regarded in their fields). But JICS is not alone. Indeed, it is only one of three joint institutes of comparable size on which UT and ORNL are collaborating. Ground is due to break this fall on an $8 million building that will house a Joint Institute for Biological Sciences. Another $8 million is docketed for a Joint Institute for Neutron Sciences that will avail itself of ORNL’s piece de resistance: the $1.4 billion Spallation Neutron Source that is due to be completed in 2006.

Although they’ve been slow in coming to fruition, all three are elements of an ambitious joint venture set in motion when UT gained a role in ORNL’s management in 1999. Collectively, these undertakings represent more than just a leap—let’s call it a quantum jump—toward vaulting UT into the upper echelon of American research universities.

Chancellor Loren Crabtree reckons it will take compensation on the order of $200,000 each to attract the 50 or more top academicians envisioned. A goodly number of them are expected to be members of the National Academies of Science and Engineering—a prestigious designation that also inures to the university. (UT has only two national academy members at present.) But money alone won’t bring them here. It also takes the allure of the SNS and of ORNL’s 250-teraflop computer, which will have six times the capacity of the most powerful computer extant, which is in Japan. (A teraflop is a mind-boggling trillion calculations per second.)

When the joint institutes all hit full stride, Crabtree foresees them generating upwards of $100 million a year in federal research expenditures. That would more than double the $77 million in such funding that UT garnered last year—counting the sum of $44 million on the Knoxville campus and $33 million at the Health Sciences Center in Memphis.

A Marriage of Convenience

With more than 1,500 scientists and an annual budget that already exceeds $1 billion, ORNL would continue to dwarf UT as a research entity. And there’s still a tendency in some quarters in Oak Ridge to look down the nose at a university that’s not nearly as prestigious or prolific. But a collaboration with UT serves ORNL’s purposes because, as a creature of the U.S. Department of Energy, the national lab isn’t eligible as a rule to get funding from other federal agencies.

The joint institutes are structured so as to qualify for grants from the two behemoths of federally funded research: The National Institute of Health with its $28 billion annual budget and the National Science Foundation which presently awards close to $5 billion in research grants each year.

UT positioned itself for ORNL collaboration in the late 1990s by forming a joint venture with Battelle to seek the ORNL management contract then held by Lockheed Martin. Battelle, which is based in Columbus, Ohio, has an exemplary record of managing other federal research facilities, and DOE awarded UT-Battelle the ORNL management contract in 1999. One inducement was a $26 million commitment by the state to build the three facilities that will house the joint institutes.

Former UT President Joe Johnson had championed the UT-Battelle partnership, but according to the UT-Battelle officials in Oak Ridge, the university’s commitment to the joint venture languished after Wade Gilley succeeded Johnson as UT’s president in 1999. “Gilley was a member of the UT-Battelle board of directors, but he didn’t even attend many of its meetings,” says one.

That neglect is paradoxical, to say the least, because one of Gilley’s primary goals was to make UT one of the top 25 public research universities in the country. But he chose to place his emphasis on the creation of an array of new Research Centers of Excellence both on the Knoxville campus and at the Health Sciences Center in Memphis.

With an investment of more than $30 million in their startup, several of these centers have fared well. A Center for Information Technology Research headed by computer-science professor Jack Dongarra has garnered more than $30 million in research grants during its first three years of existence. A Tennessee Advanced Materials Laboratory headed by physics professor Ward Plummer brought in a like amount. But their growth has been stunted by the state’s failure to make good on prior commitment of $7.5 million for expansion. “We’re carrying on at a bare bones level, but we could have accomplished a whole lot more if we’d gotten the additional funding,” says Dongarra.

Following Gilley’s demise in 2001, equally ill-fated former President John Shumaker paid a lot more lip service to the ORNL connection during his one year on the job. But while they won’t say so for publication, ORNL officials continued to chafe over what they consider a lack of uptake on UT’s part in furthering the joint venture.

“The lab knows where it wants to go over the next decade. We’re committed to being among the best in the world in neutron science, computational science and biology. The joint institutes can be an important part of that and represent a dramatic turning point for UT,” says ORNL’s director of communication, Billy Stair. “The challenge is to match ORNL’s scientific agenda with UT’s.”

In part, the friction between the two entities is a product of what both sides acknowledge as a culture clash. ORNL is run much like a business, with missions and goals established by top management. By contrast, universities are steeped in a tradition of shared governance in which important decisions don’t get made without a lot of faculty involvement. “Universities typically move slowly because of shared governance and the need for state approvals,” acknowledges UT’s vice-chancellor for research, Clif Woods, “whereas ORNL has goals to meet and can’t afford to sit around and ponder too much.”

At the same time, UT has been shackled by severe budgetary constraints. Through it all, the one person who all parties agree has done as much as anyone in his position could to keep the joint institutes moving toward fruition is Chancellor Crabtree.

Two years ago, Crabtree recruited Thom Dunning from the University of North Carolina to serve as director of JICS. Dunning, who is a molecular chemist with a DOE background, is now heading the search to fill the institute’s first four senior-scientist positions—positions that Crabtree had to sweat blood to keep in this year’s budget. At the same time, a search is underway for a full-time director of JINS which is being headed on an interim basis by Takeshi Egami, a highly regarded physicist whom UT lured away from the University of Pennsylvania last year. Crabtree and Woods acknowledge that formative work on JIBS is lagging behind the other two. “We haven’t done as good a job as we should have of defining what we want it to be,” Woods candidly concedes.

JICS, JINS, and JBIS; get used to the acronyms, because they figure to become UT’s biggest new dimension over the next decade. As esoteric as their missions may seem to laymen, including this journalist, it’s worth the effort to try to understand what they are all about and what the world may have to show for them.

JICS

At first blush, the quest to build the world’s most powerful computer at a cost of $250 million may sound like an ego trip or status symbol for the edification of a few computer scientists. But the ruddy, bearded Dunning assures that the 250-teraflop machine now being assembled at ORNL’s Center for Computational Sciences is none of the above.

Actually, its not a single machine, but rather some 4000 processors that will be linked for multiprocessing in stages by a development team consisting of Cray, IBM and Silicon Graphics. By next year, they will surpass the 40-teraflop capacity of the computer in Japan that’s now the world’s largest. The 250-teraflop goal is due to be attained in 2007.

But what on earth are the practical applications of being able to make 250 trillion calculations per second? Dunning says there are plenty, and they aren’t primarily the work of computer scientists, but rather a regalia of scientific disciplines often collaborating in multi-disciplinary teams.

His list of applications includes:

• Gaining a better understanding of the earth’s climate from much higher resolution models than are now attainable. These models can be used in planning for or anticipating climate change, understanding and then perhaps controlling water flows, and what-iffing myriad scenarios.

• Understanding how to control fusion energy. “After 50 years of studying it, we don’t know how to recreate it in a controlled environment,” Dunning says. “By modeling on the big machine we may be able to find a way to do so.” DOE has already invested billions of dollars in trying to harness fusion energy, which could provide an almost limitless supply of power without any pollutants or waste materials.

• Pursuing the creation of a whole new class of stronger and lighter materials through nanoscience studies of tiny particles and nanoscale processes.

• Realizing the benefits of the genomics revolution that will require enormous amounts of computational power to handle the data coming out of sequencing studies.

“We’ll have teams in all these areas,” Dunning says. “The first measure of our success is getting the right people, established people in their fields who will be bringing research dollars and graduate students with them. That’s my job right now.”

JINS

The ballyhoo surrounding the Spallation Neutron Source makes it sound like the biggest thing to happen in Oak Ridge since the original Manhattan Project.

The SNS’ capacity for generating brighter, more intense streams of neutrons for probing the properties of the materials at which they are aimed is expected to yield more insight into the nature of matter and its behavior under varying conditions. The results can yield dividends in a number of fields. The fruits of neutron research cited on SNS’ website (sns.gov) include credit cards, pocket calculators, compact discs, shatterproof windshields, adjustable seats and automatic window openers in cars. Numerous medical and industrial applications are also cited.

JIBS

While the definition of JIBS’ mission is still in a formative stage, it, too, is expected to have a full-time director and a complement of top scientists in the fields of genomics and microbiology as well as multidisciplinary pursuits.

“When you talk to big biologists, they need the sort of analytical tools for the study of complex organisms that ORNL can provide,” says Otto Shwartz, director of UT’s division of biology: Among them: High throughput computational and nuclear magnetic resonance capabilities. A major thrust is expected to be identifying and sequencing organisms that can contribute to pollution abatement and energy creation. ORNL also boasts the world’s most powerful electron microscope, as well as its fabled Mouse House, that may figure into the JIBS equation.

Big science that begets big discoveries obviously contributes to the betterment of mankind. And assembling the teams of distinguished scientists to avail themselves of everything that ORNL has to offer will obviously contribute mightily to UT’s prestige as a research university. Beyond prestige, though, what will Tennessee have to show for the huge investment needed to make these joint institutes flourish?

Despite the budgetary constraints that Governor Phil Bredesen has imposed on UT these past two years, he’s clear that a major investment in research is an engine for driving the state’s economic growth. The research-driven enterprises that have suffused North Carolina’s Research Triangle and the environs of the University of Texas at Austin are held out as prime examples of the kinds of returns that Bredesen wants to realize in Tennessee. And the Knoxville area stands to be a prime beneficiary of the investment Bredesen is now saying he’s prepared to make.

Getting inventions from the laboratory into the marketplace is no small feat, and not even ORNL has excelled at commercialization of its research in the past. But the UT-Battelle management team that’s now in place claims strong capabilities for doing so, and is eagerly looking forward to harvesting rich fruit from the seeds sown by the joint institutes.

“I get excited because as we assemble more of the world’s brightest scientists here, more inventions are going to occur, and then we will do everything we can to seed business opportunities in Tennessee,” says UT-Battelle’s director of technology transfer and economic development, Alex Fischer. Already, since he came on board after having served as the state’s commissioner of economic development, Fischer claims that 43 companies have been spun out of the lab. Their employment now exceeds 200 mostly high-paying jobs, and Fischer foresees these totals growing rapidly. Battelle has a $150 million venture capital fund for investing in enterprises spun out of the labs it manages, but as yet none of that money has come ORNL’s way.

Another source of return on research investment is licensing fees on patented inventions, which are the intellectual property of the institution that conducts the research. This is another area in which UT has languished, with less than $1 million in licensing income in 2002 compared to $4 million to the University of Texas and more than $10 million for the three universities that form North Carolina’s research triangle (Duke, North Carolina and North Carolina State).

Fischer says UT and ORNL will “share equitably” in licensing fees derived from the joint institutes. At the same time, UT has formed a research foundation to push patenting and licensing of inventions resulting from research conducted under the university’s own auspices.

To an academic like Crabtree, augmented research capabilities are also a vital stepping stone toward augmenting the stature of the university as a whole. “Looking out 10 years, our clear goal is to have UT knocking on the door of membership in the Association of American Universities,” Crabtree says. Presently, there are 61 members of this widely recognized top tier of universities, of which 35 are state schools (six of them in California alone). Membership is by invitation only.

Of the four southern state universities that are members of the AAU (Florida, North Carolina, Texas, and Virginia) all have at least 16 faculty members who belong to National Academies of Science and Engineering. As another key measure of their strength, all were the recipients of federal research expenditures well in excess of $100 million compared to $77 million for UT when the flagship Knoxville campus and the Health Sciences Center in Memphis are combined.

Crabtree believes the joint institutes can go a long way toward enabling UT to match these AAU universities (though they are upwardly moving targets). But he stresses that high standing in the sciences and engineering alone are not enough to gain UT admission to the ranks of the elite. “You’ve got to be a well-rounded university, so we need to bring in distinguished people in the arts, humanities and social sciences as well,” he says.

The stature that he seeks is by no means for stature’s sake alone. “If we can make our university more prestigious, then we’re going to keep our best people in state and we’re going to get the business that Tennessee needs to grow in a knowledge economy,” the chancellor says with fervor.

It remains unclear, however, that UT can get the additional revenues needed to achieve or even realistically aspire to lofty goals. Crabtree places the annual cost to the university of supporting the joint institutes alone at $10 million to $20 million a year. “It’s the rare federal research project that covers its full cost,” he says.

At the same time, UT is hurting for more funding in many other ways. Two successive years of state appropriation cuts have sliced $20 million from the Knoxville campus’ operating budget by Crabtree’s way of reckoning. While faculty positions have been preserved and salaries even raised a bit, funding for things like administrative support, building maintenance and laboratory operation have suffered. “We badly need to get that money back,” says Crabtree.

In addition, there’s a $22 million collective shortfall in faculty salaries compared to peer institutions that Crabtree hopes to narrow over time. Beyond that, his plans for growing the UT-Knoxville student body from 26,000 to close to 30,000 over the next decade call for 75 additional faculty positions (exclusive of the joint institutes) at a cost of $6 million. And all of that is before getting to the strengthening of several programs in the liberal arts and social sciences in which Crabtree believes UT can and should excel.

The sum of all these quests is over $60 million, which would represent close to a 20 percent increase in the Knoxville campus’ $354 million budget for the current year. While annual tuition increases have been averaging close to 10 percent over the past several years, there’s a limit to how much further they can rise without impairing UT’s affordability. So a reversal of the recent decline in state funding is imperative for the university to achieve its goals.

If state revenues continue to grow at the robust $500 million now projected for the current year, and if Bredesen succeeds in his efforts to contain the growth of TennCare outlays, then there may finally be room in the state budget for substantial higher-education funding increases for the first time in more than a decade.

UT’s new president John Petersen hails Bredesen as “the best governor who’s been in any state I’ve been in terms of his understanding of higher education and what it can do for economic development and quality of life.”

Petersen shares Crabtree’s vision for UT. Both at Bredesen’s behest and based on his own background as a scientist, the new president may place even more emphasis on strengthening the university’s ties with ORNL. “I’ve had numerous talks with the people at Oak Ridge and will continue to have more, including taking my role on the UT-Battelle board very seriously,” Petersen avows.

UT and ORNL have long been viewed as two pillars of the Knoxville area’s economy, and collaboration between the two should provide an economic boost as well. After all, the sums of 250 trillion calculations per second can add up pretty fast.

August 26, 2004 • Vol. 14, No. 35 © 2004 Metro Pulse