The Department of Energy is one of the principal Federal sponsors of basic research in the physical sciences. The Department has the responsibility to support fundamental science and engineering in areas relevant to its broad missions in national security, energy, and environment. Exercise of this responsibility requires a long-term and broad view of the areas of science that should be developed, as well as support for the education of the future scientific and engineering workforce of the United States in these fields. Historically, the DOE and its predecessor agencies have provided important financial resources for science and engineering education. They have also supported, as a national trust, certain areas of science involving the development, design and operation of large, complex user research facilities run for the benefit of the science community at large. These scientific, technological and educational activities have contributed both essential human resources and technologies to DOE's national security, environmental protection, and energy technology missions, and have also served the national interest in other domains.
DOE's spending on basic research is heavily weighted towards its own contractor-operated laboratories - including those considered in this study. These laboratories perform 67% of DOE-sponsored basic research, while the universities carry out 20%.,14
For the laboratories reviewed here, basic research accounts for varying fractions of the overall research and development effort, ranging from 56% at Brookhaven National Laboratory to only 0.7% at Sandia National Laboratory. Overall, the laboratories devote about 19% of their combined R&D budget to basic research.
Most laboratory directors believe that basic research is an essential part of their portfolio of activities. We agree with their arguments that basic research is necessary to support their core missions, that it is necessary to attract the most talented people to the laboratories, and that a strong in-house community of experimental and theoretical researchers is necessary to keep major experimental facilities healthy and organic. (As a rule of thumb, one laboratory director estimates that an internal community of first-rate experimentalists capable of successfully competing for 10-20% of the available machine time is necessary for this last function.)
Basic research at the DOE laboratories is under increasing pressure. In part this is the result of the same forces that are being brought to bear throughout the nation's R&D system. But there are several additional factors that are specific to the DOE laboratories: (1) a decline in the status of basic science within DOE; (2) an increasing trend towards micro-management of research by DOE program officers; (3) the potential diversion of funds away from basic research towards the new technology transfer activities; (4) the decline in defense R&D funds, which historically have been the traditional source of support for much of the basic science at the weapons laboratories; (5) inadequate up-front budgetary allowances for operation and maintenance of large user facilities; and (6) increasingly burdensome compliance requirements, especially at large scientific facilities, which are driving up laboratory operating budgets and reducing the amount of funding available for research.
The Task Force is concerned about what appears to have been a significant decline in DOE funding for fundamental research over the past three years, with the prospect of still deeper cuts to follow. The Task Force further notes that while these reductions have been occurring, overall support for basic research in the federal government as a whole has remained roughly constant or even increased somewhat.
The Task Force is also concerned that the distribution of DOE's basic research effort has failed to keep pace with recent changes in its mission. The Department is aware of the problem, and has instituted steps to address it. However, bureaucratic walls between program offices which contribute to the problem persist. As indicated in Section IV, there is a particular need for long-term, basic research in disciplines related to environmental cleanup. The activity that does exist in this area is poorly integrated into the cleanup program. By focusing too heavily on near-term demonstrations of cleanup technology, DOE is not using its laboratory capabilities effectively. The cleanup of DOE sites will not be accomplished in the near-term in any case, and in most situations there is no proximate danger to human life. Adopting a science-based approach that includes supporting development of technologies and expertise in universities as well as the laboratories could lead both to reduced cleanup costs and smaller environmental impacts at existing sites and to the development of a scientific foundation for advances in environmental technologies (see Section IV).
Compared with the universities, the main strength of the DOE laboratories in basic research has generally been in building and overseeing the operation of large-scale, complex scientific facilities (light sources, neutron sources, accelerators, reactors, etc.). However, it should also be noted that not all facilities of this type have been located at the multiprogram laboratories, and that some , including some very large ones, have instead been placed at program-dedicated laboratories such as the Fermi National Accelerator Laboratory (FNAL), Stanford Linear Accelerator Center (SLAC), and the Princeton Plasma Physics Laboratory (PPPL).
The laboratories are also capable of forming large, interdisciplinary research teams needed for certain types of `big science' problems even where large facilities are not involved. Universities are not generally as well equipped to assemble teams to conduct closely coordinated, multidisciplinary research over an extended period.
Compared with the government laboratories, the main advantages of conducting basic research at the universities lie in: (1) their primary role in educating the next generation of scientists and engineers; (2) the rejuvenating effects of the constant flow of students and post-docs through university labs; (3) the absence of national-security-related barriers to the flow of knowledge, (4) the lower price charged by the universities to sponsors for comparable research in many cases; and (5) the quality benefits provided by the systematic application of the principle of merit-based competition. Merit review is applied to basic research at the laboratories in a variety of ways, and sometimes unevenly.
The advantages enjoyed by the universities are most pronounced in `small science.' However, the DOE Office of Basic Energy Sciences also supports some individual investigator research within the laboratories on a competitive basis. To varying degrees the laboratories themselves also support such research out of their operating budgets. This type of small science research at the national laboratories is generally of a high standard, and in some cases matches the best university work in the field. On the other hand, the research culture at many of the laboratories has been influenced by their relative physical and intellectual isolation and by a sense of entitlement to research funds, and this has contributed to a loss of vitality in some research areas.
The third major category of basic research institutions, corporate scientific research laboratories, are being downsized, redirected towards activities with shorter-term commercial payoff, and in some cases dismantled altogether. Historically, research at these industrial laboratories has had a different character from either university or national laboratory research. While long-term and often fundamental in nature, it has also been shaped by knowledge of markets and motivated by expectations of commercial opportunity. And as two of our colleagues have pointed out elsewhere, `the best of the industrial laboratories achieved a high degree of coherence of purpose and of organizational cohesion, which enhanced their productivity. And the larger enterprises could operate on a scale sufficient to employ experts across a broad range of relevant topics, facilitating work on many fronts.'
The nation presently faces the challenge of finding an institutional substitute for the corporate central research laboratories. However, we do not think that the national laboratories provide a good institutional basis for a general solution. Partly the problem is the one discussed in more detail in Section VI. If an organization is to serve as an effective laboratory for an industry, that industry must have a large say in allocation decisions and in evaluation. This would be awkward at best, and probably politically unviable, in a government-owned laboratory. But in addition, for a laboratory to perform the central research function effectively a considerable flow of personnel between the laboratory and the industry is required. University-affiliated facilities seem better adapted to meet portions of this need than government laboratories. In some cases, DOE laboratories may effectively be able to perform a function analogous to corporate central research. But these are most likely to be in areas where a DOE mission - like advancing superconductivity technology - and industry interests have a strong overlap.
To Table of Contents