Design Feature: August 18, 1994

Spurred by cutbacks in defense spending and heavy overseas competition, in 1989 Congress initiated a bold step in technology partnerships between the US government (through its National Laboratory System) and private industry. The National Competitiveness Technology Transfer Act enabled Government-Owned, Contractor-Operated (GOCO) laboratories within the Department of Energy (DOE) to enter into the same legal agreements that the Government-Owned, Government-Operated (GOGO) labs within the departments of Commerce and Defense have enjoyed since the mid-1980s.

One of the instruments of technology transfer to come out of the National Competitiveness Technology Transfer Act was the Cooperative Research and Development Agreement, or CRADA. This instrument allows GOCO national laboratories to en- ter into technological partnerships with private companies and/or universities to enhance US economic competitiveness.

The DOE and Congress have created several avenues for the creation of CRADAs. First, Congress has allocated a portion of defense-program funds for a Technology Transfer Initiative (TTI) program. This program funds proposals after an announced competition and requires matching of funds from partners, either as direct or as in-kind (equivalent research) support. Second, if a specific proposal is in direct concert with existing DOE defense-program objectives, it is possible to form a CRADA as a Weapons Support Agreement. Third, it is possible to form a CRADA based on 100% funding from an industrial partner.

In addition to the standard CRADA structure, the DOE has established a Small-Business Technology Transfer Initiative at three of its national laboratories--Sandia, Lawrence Livermore, and Los Alamos--and one at the Y-12 plant. This program makes the capabilities of these four facilities available to American businesses having 500 or fewer employees. These labs work with small-business-development centers, cooperative extension services, chambers of commerce, state economic-development agencies, and vocational-education institutions to identify potential technology partners.

DOE CRADAs cover six areas of technological interest: advanced manufacturing and precision engineering, computer architecture and application, energy and environment, materials and processes for manufacturing, microelectronics and photonics, and health-care technologies.

Each CRADA is an agreement focused on a specific goal or set of goals, and each program is administered separately. Under the terms of the CRADA, the government laboratory can provide research, facilities, and personnel, while the industry partner can provide research, facilities, personnel, and funding. The CRADA agreement spells out all terms and conditions, giving all parties involved a clear indication of costs before the program begins.

Sandia National Laboratories

Established in the 1940s as an outgrowth of the Manhattan Project, Sandia National Laboratories today is recognized as one of the country's most important technological assets. With facilities in New Mexico and northern California, and test facilities in Nevada and Hawaii, Sandia is a multiprogram laboratory with major research and development responsibilities for nuclear weapons, arms control, energy, the environment, economic competitiveness, and other areas of national importance. Sandia's Technology Transfer Program has been focused on technologies in manufacturing, electronics, information, energy, environment, transportation, and health care. Sandia has initiated over 200 CRADAs with a total value of over $600 million.

The mechanics of a CRADA

Since 1991, the first year of the DOE's legislative mandate, Congress has an- nually allocated a portion of defense-program funds for TTI-based CRADAs. Proposals are submitted by the four DOE national labs, and CRADAs are awarded on a competitive basis.

For the 1993 call, Sandia National Laboratories--a multiprogram DOE laboratory with a core competency in microelectronics and photonics--and TriQuint Semiconductor--a company that specializes in manufacturing high-performance analog and mixed-signal GaAs ICs--submitted a TTI proposal targeting hardware for high-speed data communications. Realizing that the current 2.4-Gbps limit for fiber-optic data transmission would be inadequate to handle the volume of traffic that the Information Superhighway would be carrying, the DOE funded the joint project. TriQuint and Sandia would develop semiconductor devices to handle the ultrahigh frequencies that a 10-Gbps data rate would dictate.

In an era when most companies are trying to minimize their contact with government, why would a high-technology company want to get involved with a CRADA? In TriQuint's case, R&D funds were an issue: The company knew that it needed to work on a 10-Gbit telecommunications process, but the resources required for small companies to participate in this type of leading-edge, state-of-the-art project are prodigious.

At the time the TTI call for proposals was made, TriQuint was a vendor of Sandia Laboratories for radiation-hardened devices used in the nation's various defense-related programs. As a result of widely publicized Sandia ad- vances in strained-layer semiconductor technology and in the course of doing business, the subject of the telecommunications CRADA arose, and TriQuint decided to pursue an agreement. Negotiations ensued, and in January 1994, CRADA No. 1191 was signed.

The TriQuint/Sandia CRADA has a 36-month duration and a total value of $11,530,000. The technology under development will benefit American industry in general and TriQuint in particular by providing high-speed ICs for current applications, such as optical communications, and future applications, including high-speed optical computer backplanes, high-speed optical commuications for advanced computer architectures, optical switching, and intelligent vehicle highway systems.

Sandia and the federal government will benefit from high-speed semiconductors that will enhance computing speed for the simulation of nuclear weapons and nuclear-weapons systems, semiconductor technology for use in radar applications, and support in microelectronics and photonics core competency. The technology will also provide high-speed amplification, distribution, and switching devices for the Information Superhighway infrastructure and will provide a commercial path through which these parts can ultimately be made available to the consumers in the private sector as well as to DOE defense systems.

Although the CRADA has a very specific statement of work (SOW) defining each participant's contribution, a CRADA partnership of this type is a dynamic program, not a static one. In traditional government contracts, the award is for a specific program with a definite product in mind. Whether this product is a completed number of aircraft carriers or a $10 component for those aircraft carriers, the process is pretty well-fixed. The CRADA represents an almost total departure from the traditional government-contract structure be- cause it is very flexible. If, at some time during the research, the partners decide that goals must change, either due to some change in the customer's requirement or because of unforeseen technical problems, the changes within the guidelines of the SOW can be implemented by simple modifications to the CRADA legal document by the in- volved parties under the guidance of the DOE. Straightforward mechanisms are in place if more extensive adjustments are required.

The role of the sponsor, that is, the department within the federal government for which the CRADA was established, remains fairly fixed. The roles of the national laboratory and its private-sector partner can change as the program progresses. In the TriQuint/Sandia example, the participants see the roles of each partner changing as the research proceeds in a mutual leader-follower type of relationship. At one stage of the research, one partner might lead, but at the next phase of the program, the other partner might take control, depending on which partner possesses the greatest expertise in each phase of the program.

For all partners involved, the marketplace is the real customer, despite the fact that the DOE is the formal sponsor that provides the matching funds that allow Sandia to participate in the CRADA. This is an implicit admission that the production volumes for many defense-related components are no longer sufficient to keep dedicated producers in business. The beauty of this type of partnership is that the DOE funds the national laboratory's participation, but the result is commercially viable technology for the open market.

TriQuint Semiconductor Inc

TriQuint Semiconductor was formed in 1985 as a wholly owned subsidiary of Tektronix to continue the investigation of GaAs IC technology started by Tektronix and to manufacture those devices for Tektronix equipment. In 1991, the original TriQuint merged with GigaBit Logic and Gazelle to form the present company. Triquint supplies high-performance digital, analog, and mixed-signal GaAs semiconductor components to the telecommunications, wireless, and computer/network markets. The company has entered into joint-development and technology-transfer agreements with partners such as AT&T and Sandia National Laboratories to develop future GaAs manufacturing processes. TriQuint has also signed an agreement with AT&T Inc to manufacture wafers using some of these new processes.

Because of Sandia's status as a national laboratory that is government-owned and contractor-operated (in this case, the contractor is Martin Marietta), it cannot be associated with the development, marketing, or volume production of actual commercial products. Sandia's charter is to codevelop technology that can be used by TriQuint to produce marketable products, but this same technology can be utilized by the DoD and DOE for use in any appropriate federal-government program.

Because a CRADA is the formal vehicle for a congressionally mandated program of technology transfer from the public to the private sector, three critical requirements must be met by all applicants:

(1) The agreement must target major national technology goals that are market-driven and industry-defined, with direct benefit for the ultimate systems needs of the sponsoring agency (here, the DOE defense systems).

(2) The agreement must lead to a high degree of collaboration among industry, universities, and federally supported national laboratories, utilizing federal/industrial cost-sharing arrangements to the fullest extent possible.

(3) Fairness of opportunity must be demonstrated, and significant benefit must be demonstrated to problems of national importance and the DOE defense interests.

Satisfying these re-quirements isn't as difficult as it might look. Recently, the Semiconductor Industry Association issued a technology plan that would serve the US microelectronics industry through the year 2013. Assembled by 179 top US semiconductor experts from industry, government, and universities, the plan identifies many opportunities for DOE/national- laboratory partnerships with private industry. Similarly, the DOE has developed a critical-technologies list of numerous enabling technologies. Development of these critical technologies will result from advances made by following industry road maps.

The actual CRADA itself can be initiated through a couple of channels. The first step is to contact either the DOE or one of the national laboratories to find out the current status of the TTI program. If the timing or funding of a current TTI is not advantageous, other mechanisms are available for CRADA activity. For work that benefits existing defense-program activities directly, it is possible to form, at any time, either a CRADA with matching DOE program funds or a CRADA in which industry funds the national laboratories directly. Since many technology goals have been previously defined, the DOE is actively seeking partners in the private sector to begin work on these programs. In these cases, it is necessary only to contact the appropriate national laboratory or laboratories and formulate a proposal.

In some cases, private industry has specific goals that meet criteria No. 1 (above) but of which the DOE might not be aware. In this case, it is necessary to advise the DOE of this fact. The DOE will then review the inquiry and either approve or disapprove it. The negotiation stage of the CRADA varies from agreement to agreement and depends heavily on such factors as the type of technology involved, the private companies involved, previous proprietary technologies or agreements by either partner, as well as the specific goals of the parties involved. If a CRADA is awarded, the document outlines the purpose of the partnership, the benefits to be derived from it, the duration of the partnership, and the financial impact on all parties involved. The CRADA usually takes effect as soon as all parties have signed the agreement.

Clearly, the CRADA is a bold departure for the US government from the types of procurements formalized in the Federal Acquisition Register. In helping US industry stay competitive and, in fact, in many areas, leap far ahead of foreign competition, the government is making the kind of commitment to the future of this nation that many feel should have been its role all along. Certainly, this type of helping hand will be warmly welcomed by private industry in the United States.