An Assessment of the National Institute of Standards and Technology Programs Fiscal Year 1994 (1994)

V. Thomas Rhyne, Microelectronics and Computer Technology Corporation, Chair

B. Jayant Baliga, North Carolina State University

Gary M. Davidson, TRW

Douglas K. Finnemore, Iowa State University

James F. Freedman, Semiconductor Research Corporation

William J. Gallager, IBM T.J. Watson Research Center

H. R. Hofmann, AT&T Bell Laboratories

Roger F. Hoyt, IBM Almaden Research Center

James D. Huff, Scientific-Atlanta, Inc.

Richard I. Knight, Tektronix, Inc.

Frederick J. Leonberger, United Technologies Photonics, Inc.

George A. Maneatis, Pacific Gas & Electric Co. (retired)

Suzanne R. Nagel, AT&T Bell Laboratories

Arthur A. Oliner, Polytechnic University

Don Parker, Hughes Aircraft Company

D. Howard Phillips, Consultant, Durham, North Carolina

Thomas J. Shaffner, Texas Instruments Incorporated

Horst L. Stormer, AT&T Bell Laboratories

Hugo Vifian, Hewlett-Packard Company

Owen P. Williams, Motorola, Inc.

Invited Participant

George Gross, University of Illinois

Submitted for the panel by its Chair, V. Thomas Rhyne, this assessment of the fiscal year 1994 activities of the Electronics and Electrical Engineering Laboratory is based on site visits by individual panel members, a formal meeting of the panel on February 16-18, 1994, in Gaithersburg, Maryland, and on documents provided by the laboratory.

The formally stated mission of NIST's Electronics and Electrical Engineering Laboratory (EEEL) is as follows:

The Electronics and Electrical Engineering Laboratory improves U.S. economic competitiveness, Government operations, and health and safety by providing essential supporting technology, generic technology, and fundamental research to industry, government, and educational institutions. Key deliverables are measurement capability (for absolute accuracy and reproducibility) and materials reference data. These are realized through development of measurement methods, support theory, measurement reference standards (including the national primary standards for electricity, and materials reference standards), and calibration and other measurement services to assure measurement traceability. The deliverables are provided for electronic and electrical materials, components, equipment, and systems, operating from dc to light. These deliverables support research and development, manufacturing, marketplace exchange, and operation of electronic and electrical products.

Restated in simple terms in the laboratory's 1994 Program Plan, this mission is “to promote U.S. economic growth through improved international competitiveness, by providing measurement capability of high economic impact focused primarily on the critical needs of the U.S. electronics and electrical-equipment industries.”

EEEL strives to provide leading-edge measurement capability supportive of the major steps required to realize competitive products in the marketplace: research and development (R&D), manufacturing, marketplace exchange, and after-sales support. EEEL serves a broad spectrum of current national goals, including support for national initiatives in communications (the National Information Infrastructure, NII), transportation (Smart Cars/Smart Highways, Air Traffic Control Modernization, and Magnetic Levitation and High-Speed Rail Transportation), energy (Alternative Fuel Vehicles, Building and Industrial Conservation, and Federal Buildings Energy Efficiency), and the environment (Environmental Technology, Green Energy-Efficient Programs, and Weather Service Modernization).

The current EEEL strategy focuses on three major areas of deliverables: measurement capability, technology development, and fundamental research. EEEL efforts in these areas are tempered by the laboratory's fundamental mission of providing high-impact measurement capability. EEEL documented the need for such measurements in Measurements for Competitiveness in Electronics (NISTIR 4583, U.S. Department of Commerce, Washington, D.C., April 1993), prepared in conjunction with U.S. industry and other NIST laboratories. This assessment, and other industry-related EEEL activities such as field studies and trip reports from

technical conferences attended by EEEL staff, keep EEEL's metrology-based activities oriented to the current and future needs of U.S. industries. EEEL management stated that the laboratory is currently able to address only about 20 percent of the needs documented in its assessment.

EEEL's strategy focuses on developing measurement capabilities that are beyond the reach of individual companies involved in or affected by electronics and electrical engineering. EEEL places the highest priority on delivering absolute accuracy, followed by providing reproducible measurement capability. This may involve developing a new documented measurement method, a special measurement device for use by industries or by EEEL calibration services, an improved reference standard to assure the accuracy of a measurement method, a new method of delivery of measurement capability, or a standard reference material (SRM).

EEEL generally engages in technology development only when such development supports its measurement mission, devoting no more than 10 percent of its resources to development outside of that mission in cases where unusually high impact is forecast. An example of such an outreach project is the development of standard data structures for describing electronic components as part of the international effort in electronic product data exchange and electronic commerce.

EEEL also conducts directed fundamental research as an integral part of its measurement development projects to nucleate pathbreaking measurement capabilities in all of the broad program areas within the laboratory. Fundamental research projects are selected based on likely benefits to measurement development for U.S. industry.

EEEL disseminates its results through communications activities (publications, visits, technical meeting presentations, and so on), through joint activities (participation in standards organizations, Cooperative Research and Development Agreements [CRADAs], and work with on-site guest scientists), and through paid services (custom measurement development, SRMs, calibration services, and training courses). These deliverables serve the U.S. electronics industry; the electrical equipment industry; a variety of federal, state, and local government agencies; and educational institutions.

EEEL's operating budget for fiscal year 1994 is $45.4 million, up $3.3 million (8 percent) from EEEL operational expenditures during fiscal year 1993. Of this, $28.4 million is congressionally appropriated funding for Scientific and Technical Research and Services (STRS) or core funding, an increase of $4.4 million (18 percent). STRS is the most stable source of EEEL funding. For fiscal year 1994, STRS funds provide 63 percent of the EEEL budget, up from 57 percent in fiscal year 1993.

Other sources of fiscal year 1994 funding include $2.5 million from the Advanced Technology Program (ATP), $2.2 million from the performance of calibration services, and $12.3 million from external sources, primarily other federal agencies (other agency, or OA, funding) such as the Department of Defense (DOD). ATP funding has increased $500,000 (25 percent) over fiscal year 1993. Calibration services funding has increased $100,000 (5 percent). OA funding decreased $1.7 million (12 percent). Calibration income generated by EEEL represents 42 percent of NIST's total calibration income.

At the start of 1993, EEEL staff members totaled 354, including 238 technical professionals of whom 121 hold the PhD degree. EEEL hosted 46 guest researchers in 1993. At the start of 1994, staffing decreased by 11 to 343, including 235 professional staff members. The number of PhD staff members is planned to increase to 126 during 1994, largely through EEEL efforts to recruit recent PhD graduates. Plans for guest researchers in 1994 show an increase to 59.

The stated goals for EEEL resource development are, by fiscal year 1996, to eliminate dependence on outside funding for EEEL research and development activities, to maintain current staff size, and to train and select staff to increase competence and suitability for the changing technical and managerial requirements faced by EEEL.

The panel endorses EEEL's strategic plan but notes a mismatch between available resources and the plan's breadth. Increases in STRS funds and ATP funds available to EEEL appear likely to mitigate this situation, and long-range STRS budget projections that shift from OA to stable STRS funding should also improve EEEL's coverage of industry's known metrology needs. Should those increases not be realized, however, EEEL's capacity to meet its mission is questionable.

EEEL's efforts to maintain close ties to related industries are especially noteworthy, as are efforts to add new staff with strong technical credentials.

The panel judges that EEEL's goals for resource development are appropriate, especially as those goals move EEEL away from using OA funds to support its basic mission. The decrease of linkages to outside funding agencies, many of which have their own technical agendas, may provide opportunities for EEEL to expand activities into new areas with high potential benefit to EEEL customers. The shift away from OA funding, however, must be handled with care to assure that contact with customers and links to industry's and external agencies' needs are maintained. The transition must also be made carefully to minimize morale problems when EEEL staff face the cessation of long-term OA projects.

EEEL seems to respond best to those sectors of industry that are the more organized and vocal; however, other sectors of the industry critically important to U.S. competitiveness have unmet metrology needs. EEEL could be more proactive in identifying these industrial sectors and assisting them in identifying and meeting their metrology needs.

The panel judges the split of the Electromagnetic Technology Division to be appropriate, allowing the staff of the resulting divisions to concentrate on well-defined technical missions.

The panel found the technical programs of EEEL to be generally effective and competent. However, given EEEL's inability to address a major portion of identified industrial needs, selection of programs against current industrial and government needs requires constant attention and review. Specific comments on programs are provided in the divisional assessments below.

The following are the panel's recommendations for EEEL as a whole.

• EEEL technical management should conduct periodic reviews of fundamental research activities to verify that projects have a strong likelihood of benefiting the laboratory's mission and to identify new areas in which fundamental research is critically needed to provide sufficient measurement capabilities for advancing calibration and repeatability requirements. Given limited resources, EEEL must assure that its work addresses areas forecast to have the highest impact.

• Procedures for assigning technical work to divisions should be reviewed to assure that work is assigned most appropriately. Efforts to align the division's strengths more closely with the global EEEL mission should continue.

• EEEL should identify less organized sectors of the U.S. electronics and electrical engineering industry, which seem of critical importance to U.S. competitiveness, and provide support or leadership in identifying their metrology needs.

• As NIST's resources increase, EEEL should be proactive in obtaining resources for work in basic measurements and R&D of critical importance to the U.S. electronics and electrical engineering industries.

• EEEL staff should review current policies regarding support of non-U.S.-owned industries, since the technical needs of foreign-owned companies may be of critical importance to industrial developments in the United States and to agreement between U.S. standards and measurements and those of international trading partners and competitors.

• EEEL technical management should shorten the cycle of project initiation and development in areas of critical metrology needs. The accelerating pace of technological advance brings new needs for accuracy and repeatability, and in many areas industry has already matched current EEEL measurement capabilities.

• Opportunities exist for more complete utilization of the Baldrige Award criteria in laboratory planning and execution. The panel found no consistent use of such a system within NIST, although some efforts have begun; EEEL has no strategic plan to put such a system in place.

• EEEL should leverage resources through targeted participation in ATP proposals. This effort can also help organize industrial participation in key areas of technology development.

• EEEL management and NIST human resources personnel must maintain the strength of the technical staff as the role of EEEL shifts away from some long-term OA customers and their specialized technical requirements. Appropriate retraining opportunities should be provided for displaced staff, and staff with new capabilities must be recruited.

Given below is the panel's major fiscal year 1993 recommendation for EEEL as a whole (quoted from the fiscal year 1993 assessment), with EEEL's response.

“As well as refine its process for selecting and funding new projects, EEEL should plan to phase out or terminate projects” (p. 32). EEEL management and staff have established top-down and bottom-up processes to optimize EEEL's project portfolio, and they continue to examine, evaluate, and improve OA processes. Projects are agreed to by staff and management and are reviewed at the division level quarterly; any necessary modifications are documented in the Quarterly Management Report. In addition, a management review of each project during the previous year assured that management and staff had consistent expectations for the projects. NIST also undertook two activities to improve the selection and evaluation of projects. It published procedures for selecting and evaluating projects so that staff and anyone inside or outside of government could understand processes and propose improvements, and it sent each staff member a summary of how internal funding is allocated to individual projects and solicited their evaluation of the process and suggestions for improvement.

The Electricity Division maintains and improves the national standards of electrical measurement and develops stable standards for the dissemination of the units of electrical measure. The division realizes the electrical units in terms of the International System of Units and determines the fundamental constants related to electrical units. The division is responsible for providing calibration services and for developing and improving the measurement methods and services needed to support electrical materials, components, instruments, and systems used for generation, transmission, and detection of conducted electrical power. In addition, members of the division apply their expertise to selected scientific and technological problems in other areas of NIST research.

The Electricity Division's strategy involves projects in six areas: national standards, low frequency, video, power, automated electronic manufacturing, and semiconductors.

The strategy in national standards is to achieve higher levels of accuracy in electrical measurements and standards and easier replication for direct use by other organizations and to support research and development, manufacturing, quality control, and marketplace exchange. These goals will be achieved through exploiting quantum phenomena.

In the Low Frequency project, the division seeks to support measurement of the values of passive components, such as resistors, capacitors, and inductors, and to provide methods for

characterizing active components, such as integrated circuits, and circuit assemblies, but with a special focus on products used for measurement. The strategy is to pursue the two key directions of achieving higher accuracy at higher frequencies for evaluation of components, circuit assemblies, and equipment and of advancing measurement efficiency. Work on high-leverage requirements for testing of analog, digital, and mixed-signal circuits is also planned, including development of models and software to improve the efficiency of characterizing mixed-signal circuits, development of a theoretical basis for fundamental advances in self-calibration of measurement equipment, and evaluation of the need for improved methods of “built-in tests.”

In the Video project, the division has identified several focus areas where progress is required for the success of advanced video systems. These areas include development of measurement methods for evaluating video quality and flat-panel display performance and possible development of measurement methods for video cameras.

In power research, the division seeks to provide measurements needed to support the electrical utility industry and to do research in dielectrics.

In the Automated Electronic Manufacturing (AEM) Program, the division seeks to develop and demonstrate specifications and supporting technologies for computer-interpretable product data exchange, with a focus on electrical and electronic products.

In semiconductor research, the division seeks to address plasma processing.

The resources available to the Electricity Division in 1993 included 80 staff members (56 technical professionals, 24 PhDs) and $9.6 million in funding ($5 million from STRS, $1.3 million from ATP, $1.1 million from calibration services income, and $2.2 million from other sources). During 1993 the division chief retired and was replaced from within.

Staffing for 1994 is planned at essentially the same level. The fiscal year 1994 budget shows a modest increase to $10.3 million ($5.7 million from STRS, $0.8 million from ATP, $1.3 million from calibration services income, and $2.5 million from other sources). The panel found the staff within this division competent to perform the divisional mission, and enthusiasm for carrying out that mission was high. Overall morale seems to be suffering because of uncertainties about future funding and changes of administration.

Although the overall strategy of the Electricity Division is consistent with industry requirements, the panel found several significant areas needing more effective planning and program improvement.

In the Electrical Reference Standards Group, the panel is concerned with staff downsizing and increases in the technical scope of the work. Single-point dependence on staff members continues; if a staff member leaves or is reassigned, technical expertise in that person 's area becomes limited or nonexistent, perhaps threatening the group 's ability to maintain the national standards for which it is responsible. The panel is also concerned about NIST's ability to provide adequate calibration services during reconstruction of this group's facilities.

New standards are being developed every day, and industrial use of existing intrinsic standards is expanding. This trend will reduce EEEL's calibration workload, but not relieve NIST and EEEL of maintaining U.S. basic standards. Since these intrinsic standards are complex and when not properly operated may yield false measurement values, NIST and EEEL must develop a measurement assurance system to preclude degradation of the U.S. measurement system.

The National Calibration Laboratory Accreditation Program will be announced during calendar year 1994; this division will be responsible for ensuring the technical adequacy of this program's electrical measurements. To be of significant value, the NIST accreditation program must be accepted internationally; this program will require identification and resolution of technical differences with other national programs, such as specific methods of developing traceability through scaling, and formal coordination in basic standards. NIST must also attain compliance with the International Organization for Standardization (ISO) Guide 25.

Registrations to the ISO 9000 series quality standards (or the U.S. Q90 series equivalents) are increasing. The calibration requirements of these standards are more stringent than generally practiced by U.S. industry, and compliance will affect NIST calibration services workload and worldwide recognition. NIST was not yet in compliance with ISO 9000 at the time of this assessment.

For direct current (dc) voltage measurements, industrial measurements in the range between 10 V dc and 10 kV dc are generally accomplished through scaling. In this voltage range, confidence of measurements is limited by effects of leakage, and scaling techniques may not detect leakage problems. The division should consider new work in this area.

In the Low Frequency Group, the technical staff is addressing higher accuracy at higher frequencies and appears to have the specialized skill required to work effectively in this area. However, some of the topics being considered in the area of measurements of complex electronic systems will require additional skills and a larger staff than currently available. Studies planned for the near future should assist in setting a focus and allocating resources for this area. The current strategy is in alignment with the division mission and will deliver the intended benefits by enhancing the efficiency of the industrial R&D and manufacturing processes for a specific but broadly used type of circuitry. Achieving these benefits requires that work be done in very close collaboration with intended users, both to validate the economic benefits and to provide a context for future priority setting.

The current strategy of the Video Group is aligned with the division mission and will deliver its first benefits to users of flat-panel displays. This work uses common technology elements but addresses quite different users within industry. It is critical that this work be done in close coordination with expected user groups. The goals established have value only if achieved within a window of need set by industry.

Staff and equipment in the Power Group are more than adequate to support all current projects, and the available facilities are of world-class caliber. In activities such as partial discharge measurement, impulse measurements, and lighting, however, facilities appear underutilized. These resources can be more effectively exploited by increasing the scope and range of impact of activities in the power sector.

The division's strategy should be more closely aligned with current and future changes in the electric utility industry. The advent of new players and stakeholders in the industry— independent power generators, energy services companies, public interest groups—calls for a broader base of customers than just utilities. This division may be able to spur the development of

a roadmap for the electric utility industry similar to that for the semiconductor industry. As this industry enters an environment of increasing competition on a national and international basis, the division should carve out a role in fostering its improved competitiveness. This division has the resources and skills to provide unique value-added services to the “new” utility industry.

The AEM Program is doing an excellent job of targeting high-impact work in product data. Its strategy of leveraging its resources through close cooperation with other NIST laboratories, other standards bodies, and key industries appears to be quite effective. The objective of facilitating harmonization among major electrical data transfer standards, though ambitious, is much needed. At present this project appears to be critically understaffed in relation to its technical objectives and industry's needs. Loss of 1994 initiative funding has not allowed planned expansion. Projected funding increases for 1995, if fully realized, will help to alleviate this shortfall.

The panel found planned work in plasma processing was well structured though only loosely connected to the work of the EEEL Semiconductor Electronics Division.

In national standards, duplicating the National Physical Laboratory (U.K.) system for resistors in the tera-ohm range and for alternating current (ac) resistance is prudent compared with previous pursuit of other technologies.

Impressive work is continuing in the Electricity Division on ac/dc thermal voltage converters, and new work in Josephson junction arrays promises development of an intrinsic ac voltage standard. Both technologies have a place in the foreseeable future, but the short- and long-term effects of each are different. Strategic planning is required to properly merge the two technologies.

High-speed pulse generation, measurement, and characterization constitute one of the frontiers in high-speed instrumentation that is continually pushed by advances in materials, devices, and system technology. The division is currently applying a highly effective combination of in-house technology, cooperation with vendors, and characterization of commercial equipment to provide capabilities that keep abreast of changing needs. The incorporation of “round-robin” tests and support of standards development are effective tools for supporting industry's metrology needs. A very high leverage component of these efforts is ensuring that precise, independently verifiable pulse sources are available for characterization of both current and future waveform acquisition systems.

The technical content of the Low Frequency project builds effectively on staff expertise in waveform acquisition devices and standards. However, special attention will be needed to ensure that this work is closely coupled with that of industrial users. The ultimate success of this project will be determined as much by successful technology transfer as by the extension of current analytic techniques.

The panel found work in ac/dc thermal transfer standards impressive. In addition to applications within NIST, these devices have a high probability of commercial use. The division has three CRADAs relating to these devices and a fourth potential CRADA.

Although progress is being made, it appears to the panel that the division's impedance laboratories will require another 3 years of work before they are at parity with industry. Use of phase and voltage methods is a dramatic improvement in technology.

Video projects are all directed at industrial measurement requirements and standards. Because of the rapid change in the technology base for flat panels and video compression, the division faces the challenge of delivering useful measurement methods in a timely manner. Current work lags behind users' requirements, as this project is still in the start-up phase. Establishing initial high-leverage milestones is essential; these are in place for flat-panel testing, where the initial focus on flat-panel display measurements seems logical and practical. The most difficult project to define and address is metrics for video quality. The rapid shift to compressed digital video both increases the importance of metrics and complicates the task. It would be a significant fiscal year 1994 achievement to define representative industry requirements for metrics and to determine a best-case technical approach to pursue. It is possible that this may accelerate the need for additional computational capability to ensure that software developed during the project is created in a form transferable to broad segments of industry.

The AEM Program's work in facilitating the development and demonstration of electronic commerce of component information technologies and standards has been of great benefit to that national effort and holds promise for having a significant impact on the U.S. electronics component industry. Work to promulgate data exchange via World Wide Web also appears to be of great value, especially for the expanding NII.

The following are the panel's recommendations for the Electricity Division.

• The Electricity Division's mission statement should mention the fundamental goals of its work and, since it does not currently include the entire scope of the division's responsibilities, it should be written with a wider focus.

• Although relatively small in dollar terms, the power industry influences the competitiveness of most U.S. industries. NIST support of the power industry is modest, and even this may be lost as skills and facilities are redeployed to areas with more organized and vocal customers. Changes in the structure of the electric utility industry and its research arm, the Electric Power Research Institute (EPRI), provide a window of opportunity enabling the division to serve this industry's needs by (1) broadening activities to provide services to new players and stakeholders in the power industry through industry groups such as the Edison Electric Institute, the American Public Power Association, the National Rural Electric Cooperative Association, and the National Institute for Electric Power; (2) improving technology transfer to the industry (for example, SF₆ measurement technology has not been transferred for use); (3) widening the use of collaborative arrangements with the industry; and (4) developing new customers for metrology services such as independent power producers and large industrial customers.

• The panel recommends that non-U. S.-owned equipment manufacturers have access to research results, since such research would be used to develop, aid, or improve technology in U.S. plants.

• The division should establish programs in areas in which it can play a valuable role, such as (1) development of measurement and evaluation tools and metrics for energy efficiency programs, including conservation and demand-side management; (2) increased research and metrology activities in electromagnetic fields; in particular, some focused activity associated with risks to health should be considered; (3) metrology and calibration services in power quality, an area of increasing concern to electricity users; (4) development of sensors for monitoring distribution and transmission systems; (5) development of sensors and metrology for environmental impacts of the utility industry; and (6) assessment of best practices in the industry to set benchmarking standards.

• Techniques being developed by the division for measurements in complex electronics systems are ready for practical application and ready to be assessed for obstacles to implementation and economic and technical benefits. The division should consider technology transfer to a target group of early users and should conduct a technical and economic analysis of the results. Implementation experience will provide an invaluable perspective for directing future work.

• At this early stage, work on metrics for video quality can benefit from defining the problems that this project will address. Particular attention should be given to quality metrics required for successful creation and operation of video for the NII and how the proposed metrics will be employed by industry.

• Resources allow the division an expanded role in measurement applications. This will require closer linkage to industrial applications and partners in order to identify requirements and manage the transfer of technology, faster time frames in order to meet industry requirements and match the rate of change of industrial technology, and establishment of new technical competencies as required by targeted measurement applications.

• EEEL should ensure that the Automated Electronic Manufacturing Program receives the 1995 resources needed to accomplish its strategic plans.

• The division should continue its thrust in high-speed instrumentation, establishing ultraprecise and ultrahigh-speed pulse measurement capability. Optoelectronic/electro-optic technology is the best choice to provide absolute performance beyond commercial systems. The division should pursue further development in this area with the goal of characterizing the best commercial and developmental waveform sources and acquisition systems.

Given below are some of the panel's fiscal year 1993 recommendations for the Electricity Division (quoted from the fiscal year 1993 assessment), with the division's responses.

• “As the panel recommended in its fiscal year 1991 report, a Fundamental Constants Committee should be formed to coordinate fundamental constants work NIST-wide. Committee membership should include staff from the Fundamental Electrical Measurements Group, the Cryogenic Metrology Group, and the Time and Frequency Division of the Physics Laboratory, as well as from NIST's Standard Reference Data Program and the division 's Precision Measurement Grant Program, and should establish a working subgroup for maintenance of the fundamental constants. . . .” (p. 39). The responsibility for the adjustment of the fundamental constants and the precision measurement grants resides with the Physics Laboratory. Research and development that contributes to our knowledge of the fundamental constants is carried out in the Physics Laboratory, the EEEL, the Manufacturing Engineering Laboratory, and the Chemical Science and Technology Laboratory. EEEL sees no barrier to communication and coordination with the other laboratories and does not feel that a committee would significantly improve this process.

• “EEEL should become a node on the Electric Power Research Institute 's EPRINet to facilitate the work of the Applied Electrical Measurements Group” (p. 32). During a visit to EPRI by three EEEL managers, a NIST connection to the EPRINet was requested. This request was accepted politely, but with limited enthusiasm, by EPRI management. No connection was established, and because of the good working relationship between EPRI and NIST, EEEL felt it would be imprudent to put undue emphasis on this matter.

• “The Electronic Instrumentation and Metrology Group should expand its dissemination of the techniques developed for measuring complex electronic systems, e.g., by initiating college and professional tutorial courses in the measurement of complex electronic systems. . . .” (p. 41). During the past year, a cooperative program with the Computing and Applied Mathematics Laboratory has begun to generalize the approach while maintaining the necessary theoretical rigor. This work has attracted the attention of the applied statistical community, with an American Statistical Association Senior Research Fellowship awarded to a university professor to study the statistical aspects of the NIST testing strategies and to formalize the method. Work is under way to develop a MATLAB “Toolbox” software package that can implement the NIST testing strategies approach. Also, a third NIST workshop, “Testing Strategies for Analog and Mixed-Signal Devices,” was planned at NIST for April 26-28, 1994.

• “The Electrical Instrumentation and Metrology Group should consider a new initiative on time measurement and generation with the goal of improving accuracy, linearity, and time jitter by a factor of 10 to 100. These capacities are needed to surmount barriers in a variety of areas ranging from digital signal processing applications to instrumentation for testing of next-generation digital devices. The need is critical within both NIST and U.S. industry” (p. 42). The primary responsibility for time measurement within NIST rests with the Time and Frequency Division in the Physics Laboratory, whereas the primary responsibility for meeting the measurement needs of the electronics industry rests with EEEL when such needs are not otherwise effectively addressed. EEEL will discuss this opportunity with the Time and Frequency Division to determine if a joint project should be established.

The mission of the Semiconductor Electronics Division is to provide technological leadership to semiconductor manufacturers by providing, through development and evaluations, measurement methods, data reference artifacts, models and theory, and associated (measurement) technology to allow cost-effective manufacturing and performance/quality assurance of semiconductor devices and integrated circuits; to conduct research in semiconductor materials and processes, devices, and integrated circuits to provide, through both experimental and theoretical work, the necessary basis for understanding measurement-related requirements in semiconductor technology; and to disseminate and foster application of technical results and assist in the development of standardized test methods and standard reference materials to enhance the quality, performance, and reliability of semiconductor materials, devices, and integrated circuits to aid in manufacturing productivity and in the development, transfer, and exploitation of semiconductor technology for public benefit.

The evolution of silicon integrated circuits has been based on “shrink” technology, i.e., the ability to fabricate smaller device dimensions, thus increasing the density per square centimeter with associated reductions in cost and improvements in performance. State-of-the-art processing is already below 1.0-µm minimum dimensions, and devices are in development at 0.5-µm minimum dimensions. Semiconductor Technology: Workshop Conclusions (Semiconductor Industry Association, Washington, D.C., 1993), commonly known as the SIA Roadmap, projects that shrink technology will continue to 0.12 µm by the year 2007, resulting in 16-gigabit memory chips and 100-megabyte transistor microprocessors. Experimental data on single devices indicate that field-effect transistors can function at 0.05-µm minimum dimensions. Manufacturing cost-effectiveness, not device physics, will determine the extent of integration of integrated circuits. Metrology and in situ and ex situ characterization are essential to cost-effectiveness. The Semiconductor Electronics Division 's strategy is to provide industry the measurement capability needed to meet these challenges. Industry is encouraging this effort through formation of a National Center for Semiconductor Metrology at NIST to provide the metrology requirements defined by the SIA Roadmap. Funding is planned at $25 million per year, focusing on measurement support for silicon complementary metal-oxide semiconductor digital integrated circuits. The establishment of this center is assumed herein as a given.

The division's strategy also addresses needs not encompassed by the SIA Roadmap, namely, needs in power semiconductors, compound semiconductors for electronic and optoelectronic applications, and analog devices.

Current strategy in silicon integrated circuits includes nanoelectronic, contactless, and nondestructive optical probes for on-line and in situ processing, electrical and thermal characterization, thin film characterization, and test structure metrology for advanced manufacturing. Areas for expansion are modeling for product and processing design,

measurement immediacy during processing, contamination during processing, packaging, and lithography.

Costs and time to market dictate that computer-aided design, including process and device simulation, must be practiced in the semiconductor industry tomorrow the way circuit simulation is practiced today. In the Modeling for Product and Processing Design Program, the division plans to provide measurement methods for and reference data on materials properties, chemical reaction rates, and other factors needed as input for these models. Test methods will also be developed that will allow the evaluation of model performance by comparing process output with model predictions.

The Measurement Immediacy During Processing Program is being developed because testing and measurement represent an increasing percentage of manufacturing cost. The division's strategy is to reduce the number of quantities for which there are currently no adequate measurement methods, with special emphasis being placed on nondestructive, noncontacting methods. The effort will then be extended to replace ex situ with in situ and real-time measurements.

The Contamination During Processing Program will develop imposed measurement methods for contaminants in starting materials and new SRMs for assuring the accuracy of these methods. This division will focus on contaminants in bulk silicon, extending existing measurement methods by establishing lower limits of sensitivity. The division will also study nondestructive measurements methods that have high accuracy and high selectivity and are susceptible to miniaturization. Entirely new measurement approaches must be explored for particle contamination, since particle levels to be measured are below practical extrapolations from known methods.

Completed chips must be placed in “packages” that provide physical protection, heat dissipation, and electrical interconnections. As chips have become faster and more complex, package design has become more complex and more costly. The Packaging Program will provide reference materials data on critical categories of properties that industry can use in designing its packages. The division will also provide measurement methods that manufacturers and users of packages can employ to verify package performance.

Lithography is key to shrink technology, printing the current image and transferring the images into circuit material. Dimensions, tolerances, alignment, and reproducibility are critical. Deep submicron dimensions mandate nanodimension tolerances, and current techniques for metrology and characterization are simply not extendable. New approaches will be explored and developed.

The resources available to the Semiconductor Electronics Division in 1993 included 56 staff members (38 professionals, 24 PhDs), 6 guest researchers, and $6.9 million in funding ($5.4 million from STRS, $209,000 from ATP, $203,000 from SRM funding, and $1.1 million from other sources). Planned 1994 staffing is essentially the same. The fiscal year 1994 budget shows a 5 percent increase to $7.3 million ($5.7 million from STRS, $775,000 from ATP, $85,000 from SRM funds, and $744,000 from other sources).

The strategic plan for the Semiconductor Electronics Division is clearly targeted correctly and focuses the division on research, development, and application of methods, techniques, and standards for the demanding metrology needs of the semiconductor design and manufacturing community. The division's strategy seeks direction from the broadest semiconductor constituency possible. The strategic plan recognizes the need to make considerable improvement in the metrology capability for semiconductors and accepts the responsibility of providing a timely response to the needs expressed in the SIA Roadmap. The strategic plan accurately expresses the issues related to modeling parameter measurements, real-time metrology for process control, contamination, packaging, and lithography. Also, the division properly maintains a smaller program for those semiconductor technologies not addressed by the SIA Roadmap.

The current projects in the Semiconductor Electronics Division (Nanoelectronics, Semiconductor Characterization Technology, Electrical and Thermal Characterization, Thin Film Characterization, and Test Structure Metrology for Advanced Semiconductor Manufacturing) are all essential activities, but the panel thinks there are other critical projects, such as multichip modules, that should be considered for higher priority.

It is essential that NIST personnel take advantage of the upcoming renewal of the SIA Roadmap to drive the establishment of metrology needs within each of the technology working groups. The SIA Roadmap Coordination Group has requested that each working group provide its metrology needs in priority order; NIST representatives are participants in the key working groups and should lead in this assessment.

It is evident that much of the work of the Semiconductor Electronics Division is world class and of direct significance to the semiconductor industry. The panel was pleased with the enthusiasm of the scientists who made presentations during laboratory tours and encountered a healthy mix of new staff and experienced personnel.

The panel was particularly impressed with the Electrical Test Structure Overlay project, not only for its technical achievement (separation of tool versus mask components), but also for its alignment with the metrology mission of NIST and with the needs of industry. Likewise, the division's development of scanning probe methodologies is basic to future nanostructure measurement requirements as outlined in the SIA Roadmap. The scanning capacitance microscope and modeling effort will impact understanding and control of doping distributions at ultralarge-scale integrated circuit (ULSI) contact dimensions. Programs in ellipsometry and the optical spectroscopies (infrared and photoluminescence) continue a tradition of cutting-edge accomplishment and are well recognized within the technical community. Of note, however, is an apparent lack of comparable expertise with other ULSI characterization tools (e.g., x-ray diffraction, ion microscopy, and Rutherford backscattering). The panel believes that collaborative activities are a reasonable means of accessing these technologies; the interaction with Arizona State University for transmission electron microscopy studies is a good example of this approach to collaborative work.

The division's determination to ensure the relevance and impact of its projects is evident by its frequent interactions with industrial partners, with the Semiconductor Research Corporation, and with the consortium SEMATECH; by its publications, presentations, and promotional conferences; and by its strong alignment with the SIA Roadmap. In fiscal year 1993, the panel noticed a visible preoccupation with customers, and the panel notes this again.

The ultimate goal of NIST research must be aligned with NIST's primary product, metrology and characterization, and this division must focus on parameters key to semiconductor device manufacturing and utilization. Rapid changes in industrial needs, driven by rapid changes in technology, require the division to be flexible in its research programs, while avoiding the trap of doing “trendy” work rather than science and engineering relevant to its mission.

All research and development organizations strive to achieve both technical excellence and customer satisfaction. These two goals are not identical and typically compete in a resource-limited environment. Resources must be allocated through a strategic plan or top-down portfolio ranking of competencies and projects that align with the laboratory's mission. Otherwise, competencies and projects of less relevance appear and are often difficult to terminate. Although the mission of the division is clearly stated, the panel has noted activities that are not clearly aligned with this mission, which is an indicator of the need for stronger portfolio management at the strategic level.

The panel found the following competencies and programs to be well aligned with the metrology mission: Electron Density and Hall Mobility, Scanning Capacitance Microscopy, Photoreflectance Spectroscopy, New Electrical Test Structure for Overlay, Faster Metal-Oxide Semiconductor Oxide Reliability Characterization, SRM for Interstitial Oxygen, 14-nm-Thick Oxides Certified References, and scanning tunneling microscopy Characterizations in Air on GaAs (gallium arsenide) Junctions. Those with less emphasis on measurement and metrology include Electro-Thermal Power System Simulation, Micromachined Semiconductor Devices, X-ray Lithography Mask Design, and SIMOX (separation by implanted oxygen) material development. It is essential that the latter programs be brought quickly to the level of understanding required to determine whether they relate to the primary mission of the division.

It is commendable that the division has made an effort to align its projects with the metrology needs identified in the SIA Roadmap. The division has also established good ties with the consortium SEMATECH, as recommended by the 1993 panel. This strategy should continue into the future.

The following are the panel's recommendations for the Semiconductor Electronics Division.

• The panel reiterates its fiscal year 1993 recommendation that the Semiconductor Electronics Division take leadership in formulating with industry a national strategic plan for compound semiconductors.

• NIST, with the support of the SIA Roadmap Coordination Group, is establishing a Semiconductor Metrology Advisory Group composed of government, industry, and university

interests and equipment suppliers. NIST should use this advisory group to establish priorities for the metrology needs expressed in the renewed SIA Roadmap.

• In response to OA funding, the division often undertakes research that falls outside its primary metrology mission. The format for the division's program plan should be changed to relate each project activity clearly to its funding source, thereby facilitating a better assessment of project objectives and accomplishments in relation to the appropriate strategic mission.

• Although it may be advantageous to maintain expertise in a given technical field, projects that were supported under OA funds must be evaluated to assure that their objectives can be clearly related to the strategic metrology mission of the division before STRS funds are applied to them. Examples of projects that need reevaluation are the silicon-on-insulator effort and model development for commercial software.

• Because of rapid changes in the metrology needs of the semiconductor industry, the division must develop an approach to rapidly acquiring new expertise as such expertise becomes necessary. The panel recommends that the division acquire this expertise from industry, university laboratories, and other government laboratories by funding sabbaticals at NIST for the required experts when the need arises for their competence. To facilitate acquiring new expertise, activities in other national laboratories should be evaluated for establishing ties similar to those the division has with Sandia National Laboratories.

• The panel endorses the strategy to allocate a small fraction (less than 10 percent) of STRS funds to basic science research. However, these activities should be evaluated yearly for their contribution to the strategic metrology mission. Basic research activities that do not show relevance to the NIST metrology mission within a reasonable period (2 to 3 years) should be terminated to develop new initiatives, allowing the division's more timely response to rapid changes in the semiconductor industry.

• The panel recommends that EEEL further develop its strategy for project selection, considering published methodologies such as the portfolio-based methodology described in Third Generation R&D (P.A. Roussel, K.N. Saad, and T.J. Erickson, A.D. Little, Inc., Cambridge, Mass., 1991).

Given below are some of the panel's fiscal year 1993 recommendations for the Semiconductor Electronics Division (quoted from the fiscal year 1993 assessment), with the division's responses.

• “Research on power devices and molecular beam epitaxy should continue to be done, but a broader strategic framework should be developed to encompass the research on HgCdTe and silicon-on-insulator materials ” (p. 46). Significant effort was made in the past year to determine

the appropriate NIST program in silicon technology in cooperation with the Office of Microelectronic Programs and other laboratories within NIST. Planning for the nonsilicon portion of the program is not as advanced but will be an EEEL priority effort in fiscal year 1994.

• “The Semiconductor Electronics Division should increase its collaboration with SEMATECH as an efficient way of increasing its collaboration with its principal customers” (p. 48). In fiscal year 1993, the division had significant interactions with SEMATECH in the areas of materials characterization, thin film characterization, and lithography, including linewidth determination.

• “The panel urges the Semiconductor Electronics Division to provide national leadership by defining an insightful strategic plan for its compound semiconductor research” (p. 48). The division is actively developing plans for appropriate support of compound semiconductor research.

• “The Semiconductor Electronics Division should extend its modeling (including modeling of thermal effects) to mainstream complementary metal oxide semiconductors and bipolar complementary oxide semiconductors ” (p. 48). As indicated in its program plan, the division plans to extend its modeling work from power devices to a broader range of devices. The development of a phasing plan for specific tasks is part of the effort to plan for the National Center for Semiconductor Metrology.

• “Due to uncertainty in the long-term application of silicon-on-insulator device technology to very-large-scale integrated circuit digital complementary metal oxide semiconductor technology, the project should be given a relatively low priority” (p. 49). EEEL considers this project part of its response to OA needs, one of NIST's responsibilities. Within the past year, there has been increased interest in this technology by industry to help in the development of very low power devices to be incorporated in portable, battery-powered, advanced electronic systems.

• “Align the power device and thermal measurements projects with the electric-car project in the Electricity Division” (p. 49). EEEL is coordinating its electric-vehicle-related activities. For example, staff from the Semiconductor Electronics Division and from other NIST laboratories worked with staff from the Electricity Division to organize and host a workshop on electric vehicle technology. In addition, the two divisions are coordinating their responses to industry efforts to involve NIST staff in activities to lower barriers to the production of effective and appealing electric vehicles.

The Electromagnetic Fields Division develops and evaluates systems, devices, and methods for measurement and analysis of radio frequency electromagnetic fields, signals, noise, interference, and properties of materials for guided and freely propagating fields, including frequency and time domain representation of electromagnetic fields and their interaction with materials and structures; provides essential measurement and calibration services, enabling industry and government to solve important national, commercial, industrial, and military problems such as evaluating the performance of microwave and millimeter systems, components,

and materials used, for example, in advanced radars, satellite and mobile communications, and automated test systems; assists other agencies with measurement-related issues such as determining levels of nonionizing radiation and solving electromagnetic interference problems; and disseminates results to industry, universities, and other government agencies to foster effective research, development, manufacturing, and marketplace equity.

The Electromagnetic Fields Division consists of the Microwave Metrology Group, the Fields and Interference Metrology Group, and the Antenna and Materials Metrology Group. Its principal program areas include microwave/millimeter wave metrology for continuous-wave transmission line measurements, noise measurements, and dielectric measurements, antenna metrology, and fields and interference metrology. These program areas have resulted in the development of substantial measurement services that account for over 20 percent of all NIST fees obtained from calibration and special test services. These services and associated standards provide a consistent base of measurements to enable multiple contractors in the defense, aerospace, communications, and related industries to assemble complex systems and perform stringent performance assessments of these systems.

Within these program areas, the division studies the interaction of electromagnetic (EM) fields with devices, components, and materials, with an emphasis on coherent radiation, applying the results of these investigations to the improvement of measurement science in frequency ranges up to 100 GHz. The division also develops theory, measurement methods, and standards for measuring radio frequency and microwave power, impedance, attenuation, S-parameters, phase, noise, antenna gain, antenna polarization, antenna pattern, EM properties of materials, EM field strength, and EM/electromagnetic compatibility (EMC) parameters. The results of these efforts are disseminated as metrology capabilities and services to industry and government; as assistance in developing standardized test methods to assess and enhance the performance, reliability, design accuracy, production quality, and safety of electronic devices, components, subsystems, and systems; and as aids in developing, transferring, and exploiting electromagnetic technology for public benefit.

The major deliverables provided by the division are primary physical standards; reference standards; calibrations and special tests; measurement methods; measurement theory, models, and computer programs; special devices and instrumentation for new measurement methods and standards; and consultation on measurement problems.

The resources available to the Electromagnetic Fields Division in fiscal year 1993 included 85 staff members (53 professionals, 21 PhDs), 4 guest researchers, and $10.8 million in funding ($3.9 million from STRS, $150,000 from ATP, $1.1 million from calibration revenue, and $5.6 million from other sources). The 1994 staffing is planned at two fewer than in 1993. The fiscal year 1994 budget shows a 3.6 percent increase to $11.2 million ($4.6 million from STRS, $330,000 from ATP, $1.0 million from calibration revenue, and $5.2 million from other sources).

The panel agrees with the Electromagnetic Fields Division's strategy to move from heavy support of metrology techniques for the DOD to the development of low-cost metrology techniques for commercial application and the consumer market. This strategy is in direct support of U.S. competitiveness in a broad spectrum of consumer electronics.

The panel also strongly supports the strategy to emphasize metrology for microwave integrated circuits and EMC for new areas such as wireless networks for computers, personal communication services, and communications and radar systems for intelligent highway and vehicle systems. It is essential for NIST to take a lead role in coordinating the development of EMC standards (for emission and immunity) for the United States and in harmonizing these with international standards. The division is commended for its survey of industry needs in EMC. This provides the division the basis for identifying new initiatives that should have widespread support. The EMC strategy does not, however, reflect the potential impact that NIST could have if it were more proactive. Because of this division's reputation, it could take a lead role in activities such as planning EMC workshops or organizing industry technology councils.

Although the division has an appropriate strategy, it should place more emphasis on developing new initiatives and funding to fully implement the strategy. This process could be facilitated with a carefully planned roadmap that identifies the technologies, capabilities, facilities, and personnel required over the next few years.

The division recognizes that as its STRS funding increases, its projects will focus on new initiatives rather than on its more mature metrologies. The challenge during this transition will be to implement new initiatives in a rapidly changing commercial environment while operating within NIST's funding response time.

The programs in the Electromagnetic Fields Division reflect that the division is in transition from major OA funding (primarily DOD) to more STRS funding, with a related increased focus on commercial and consumer electronics applications of microwave metrology. The division's move into monolithic microwave integrated circuits is particularly appropriate, and the panel found the program in this area excellent. Work in materials characterization is also good and expected to be important for the future. The EMC measurements program is important as the division moves to more commercial electronics applications.

The following are the panel's recommendations for the Electromagnetic Fields Division.

• Commercial applications based on microwave frequencies should lead to a large number of metrology issues involving microwave integrated circuits, antennas, and electromagnetic compatibility. The Electromagnetic Fields Division should be positioned to support U.S.

industry in these emerging technologies and applications, emphasizing the development of low-cost transfer standards and calibration techniques that are easily used by industry.

• The division should consult with relevant industrial organizations to assess what new measurement problems and needs for new standards are expected to arise from new microwave applications and then prepare a systematic roadmap and develop appropriate initiatives. This should be done as soon as feasible so that the division can respond rapidly to these industries. NIST management should also provide a rapid submission of such initiatives.

• Microwave integrated circuits in both monolithic microwave integrated circuit and hybrid forms will find greatly increased utilization in new commercial applications in frequency bands ranging from about 1 GHz to almost 100 GHz. Since most of these new applications may not be economically feasible without improved microwave integrated circuits, current divisional activities in this area should be expanded.

• Vehicle collision-avoidance systems operating at millimeter wavelengths and extensively utilizing microwave integrated circuits will be on the market in a few years. Since the division has the special calibration knowledge and skills required to solve issues arising from these applications, the division should survey manufacturers to determine NIST's support and develop the appropriate initiatives.

• Many different types of antennas will be required for new commercial applications, but in some cases (e.g., the new generations of automotive electronics that will provide personal communications, computer-generated maps driven by the Global Positioning System, collision-avoidance radars, guidance systems for the “intelligent” highway), a proliferation of communications and radar functions will be present simultaneously. Antennas of different types operating at different frequencies will lead to an unprecedented number of mutual coupling effects. The division should be sensitive to new technical issues arising as a result and consider initiating new measurement programs to identify and help solve such problems.

• New commercial applications require improved metrology in electromagnetic compatibility involving measurements for both immunity and emissions. The division has not developed a systematic roadmap for these issues based on its own technical judgment and surveys of industrial needs. Such a strategic planning approach should be supported by appropriate levels of funding and staff.

• The division should provide leadership in electromagnetic compatibility, since there is no industrial organization representing that area. In this context, the division should be proactive in helping to prepare U.S. industry to meet the new European Union EMC standards, including standards harmonization, EMC metrology, and EMC calibration standards.

• The division should determine the support it can provide to the information superhighway on issues such as EMC, antennas, microwave integrated circuits, and other technologies.

• The division should consider retraining existing staff and hiring new staff, particularly younger staff, to address new applications. There should be some overlap of new staff with retirees to provide continuity in key technologies.

• In anticipation of new increases in STRS funding and OA funding decreases, the division should systematically examine which programs will continue to increase and which should be phased out or reduced in scope. Areas that are expected to retain continuing industrial interest, though not prominent in new applications, should not be totally eliminated; such areas include near-field metrology, radar-cross-section metrology, and six-port measurements.

Given below are some of the panel's fiscal year 1993 recommendations for the Electromagnetic Fields Division (quoted from the fiscal year 1993 assessment), with the division's responses.

“The Electromagnetic Fields Division should take the following steps:

• Develop a strategic plan . . . for the transition to the new industrial environment.

• Identify emerging technical requirements in industrial growth areas . . . and propose the necessary services to maintain NIST's and the division's worldwide leadership in microwave measurements and standards.

• Make in-house capability . . . readily available to commercial customers.

• Shift the division's electromagnetic compatibility and electromagnetic interference activities from Department of Defense to commercial applications.

• Extend the near-field calibration capability to higher frequencies and electrically large antennas. . . .

• Explore radar cross-section measurements for commercial applications ” (pp. 51-52).

In addition to contributions to updating Measurements for Competitiveness in Electronics (NISTIR 4583) and the new strategic plan, the division staff conducted a special study of commercial needs for improved measurements in the area of electromagnetic compatibility during fiscal year 1993 and expects these activities to expand with growth in NIST funding. Through CRADAs and guest worker agreements, industry has access to NIST with very few restrictions. The division has conducted a planning study to identify commercial need for support in the area of electromagnetic compatibility and interference. In addition, the division has funded commercial work in this area. As defense needs decrease and commercial needs become more apparent, EEEL has had difficulty expanding nondefense funding rapidly enough to maintain a productive program.

As of January 1994, the Optical Electronic Metrology Group, the Cryogenic Metrology Group, and the Superconductor and Magnetic Metrology Group constituted the Electromagnetic Technology Division. During 1994, the Optical Electronic Metrology Group will have completed

the process of becoming a separate division, the Optical Electronics Division, leaving the other two groups to form a second division. Since this reorganization was under way at the time of the assessment, the panel chose to assess the programs according to the new division structure.

The mission of the new Optical Electronics Division is to enhance the competitiveness of the U.S. optical electronics industry, primarily by providing it with comprehensive and technically advanced measurement capabilities, primary standards, and traceability.

The overall strategy of the Optical Electronics Division is targeted at five key segments of optical electronics: optical fiber communications, optical fiber sensors, optical information storage, optical signal processing and computing, and lasers.

Division activities to support the measurement needs of optical fiber communications are primarily directed at the component level, where the strategy is to develop and evaluate measurement techniques, develop and disseminate reference data, develop standard reference materials and components, provide measurement services, participate in industry-wide efforts toward measurement standardization, and provide technical support to government and university activities. These activities are directed at fiber optic and optoelectronic components. The division will also survey industry on its metrology needs for process control and identify areas of high impact for attention.

Optical fiber sensors have promise in such measurement applications as displacement, rotation, temperature, flow, pressure, fluid level, chemical and biological quantities, and magnetic and electrical quantities. Work in this area involves developing measurement methods and standards for components used in sensors and developing next-generation sensing technology, primarily focused on sensors for electromagnetic quantities with improved selectivity.

Optical and magnetic information storage have promise for improved storage density. The division strategy targets specific metrology needs where extensions of existing capability in other division programs can be used.

Optical signal processing and computing represent a research area that has promise for powerful new capabilities based on optical technology. The division's strategy in this area is to exploit the present programs in optical fiber communications to support specific measurement needs in this emerging field.

The laser program within the division is focused on a limited number of measurement needs related to beam energy and power for spatial beam profiling and temporal beam quality for high-power gas and solid-state lasers.

The resources available to the Optical Electronics Group in fiscal year 1993 included $6.6 million in funding ($4.7 million from STRS, $130,000 from ATP, and $1.7 million from other sources). Fiscal year 1994 funding for the Optical Electronics Division is anticipated at $6.6 million ($5.3 million from STRS, $220,000 from ATP, and $1.1 million from other sources). The division has 40 staff members (35 professionals, 21 PhDs) and 6 guest researchers.

The mission of the new Optical Electronics Division is very broad in scope in a technological area that has experienced explosive growth over the past decade and holds promise for an even wider set of industrial applications. The division strategy by necessity is more limited in serving the measurement, metrology, and standards needs of this market. The panel supports plans within the division to reexamine current projects in terms of the new division's mission statement and related customer needs and priorities. This reassessment is both timely and appropriate and may suggest a different organizational design to leverage the core capabilities of the division in better ways.

The new division is already a recognized international leader in optical fiber measurement systems and standards. The core program in this area is strategically linked to critical customer needs and requirements, both in optical communications and optical sensors. Building on this capability is strategically and tactically sound and provides needed improvements in measurement accuracy and standardization.

Strategy in the source and detector area is targeted at critical measurement techniques, calibrations, and standards for frequency response, high-accuracy spectral responsivity, linearity, and laser noise. These targets are all important for next-generation systems and represent an excellent strategic use of division resources.

As integrated transmitters, detectors, and optical amplifiers emerge as critical components or subassemblies, a division strategy must be defined to address associated measurement needs. The division is uniquely positioned to contribute to metrology and standards for optical amplifiers at both the component and subassembly levels and needs a targeted, integrated strategy. In other integrated optoelectronic devices, the division's efforts are a mix of device demonstrations and measurements to support integration, materials characterization, and process control. Although this work supports the general mission of the new division, its strategy must be explicitly developed. In particular, further movement into metrology for process control must be closely associated with best-in-class industrial processes to have the most meaningful impact. Division strategy in this area must be well linked to the division's primary mission, and resources must be balanced against other priorities.

Similarly, strategy in the optical fiber sensor area must be further refined. Characterization of key sensor components is strategically aligned with the division mission, and it benefits from synergy with work in other areas. However, development of next-generation sensing technology, with a focus on electromagnetic quantities, requires a strategy to create pull from both industry and other NIST divisions. The panel recognizes the core competence in this area and suggests

clear definition of specific customer needs to demonstrate the importance of this investment to industrial competitiveness.

The laser power and energy measurement strategy appears responsive to specific industrial requests and appropriate for the limited and focused capability in this area. However, given the breadth of technologies associated with laser applications, it is critical that all appropriate NIST laboratories cooperate to assure that NIST provides programs that meet the needs of this area.

Given the double-digit growth of the optical electronics industry and the importance of the fiber optics infrastructure for global multimedia communications, the lack of any funding increase for this new division in fiscal year 1994 seems inappropriate. Flat funding does not allow growth or help reduce project cycle time, and standards and traceability for many basic parameters have yet to be established. These demands cannot be addressed currently unless funds are directed away from other important projects.

Only the Integrated Optoelectronics project, which has included device work and process metrology and which now also includes thrusts in optical interconnects, received a substantial funding increase. Integrated optoelectronics process metrology is not a unique competence for NIST and is not necessarily a high industry priority compared with optoelectronic device measurement metrology, erbium-doped fiber amplifier (EDFA) characterization methodology, and fiber dispersion measurements and standards (in particular, polarization mode and chromatic dispersion). It is not clear from project allocations that funding is prioritized relative to critical needs.

The Optical Electronics Division must learn how to move into new areas while managing its existing capability. In many cases it is appropriate that external financial support be developed in order to continue services in mature areas. By refining a strategy for external funding, new initiatives could be undertaken in areas such as optical data storage. The panel is concerned that timely responsiveness to these emerging areas of optical electronics is not being achieved because of financial constraints and project scope. The panel encourages development of a refined and integrated strategy during fiscal year 1994 to address these issues.

The panel was impressed by the high quality of the Optical Electronics Division's work; in particular, work on fiber measurement and standards, fiber sensors, and integrated optoelectronic devices has made outstanding progress.

The panel emphasizes the need for the new division to refine the focus of its technical activities relative to U.S. industrial needs. The panel observed some synergies among the projects and highlights them here, as they could lead to a more customer-focused organization.

For example, the panel has concluded that combining fiber measurement activity and the characterization portion of the fiber sensor work might be advantageous, given commonality in basic measurement needs. Characterization of fiber components for EDFAs could also be included in this area.

The Integrated Optoelectronics project is a well-intended effort to demonstrate new integrated optics devices and to contribute to in-process metrology for III-V structure growth. However, work on integrated optics devices should be more focused on device metrology, since process metrology work is very capital intensive. Also, very tight coupling with high-volume

state-of-the-art industrial processes is instrumental to making the most valuable contribution in this area. The benefit of in-process metrology and its consistence with the division's mission are not clear to the outside observer.

The program for measurement of sources and detectors has continued to show excellent results and represents one of the core competencies of the division. The panel encourages more proactive work in source and detector standards. The need for future frequency response measurements beyond 60 GHz has to be monitored, but the need for more accurate power measurements, wavelength standards, and noise measurements is very real and needs to be addressed now. To leverage resources better, the EDFA characterization effort at the subassembly level could be integrated into this program, a step that would also better align the division's capability with the components manufacturing industry.

The total scope of the Laser Power and Energy Measurement project is limited relative to the laser industry's needs. In the communications bands (850, 1300, and 1500 nm), there is good synergy with other division activities. Given limited resources, more general work needs to be very responsive to specific customer requests. Expansion in this area should be based on an external funding strategy.

In the Optical Fiber Sensor project, fiber measurement activity is very synergistic with the Optical Fiber Measurement Systems and Standards project. As pointed out in the fiscal year 1993 assessment (p. 62), there seems to be enough synergy to consolidate these two efforts. Providing sensors for other laboratories and groups to enhance measurement capabilities is a very important task that is closely coupled to potential users and is also a basis for providing an opportunity for specific customer feedback. This aspect of the work warrants a separate group activity.

The explosion in electronics and multimedia communication will provide a tremendous challenge and opportunity for EEEL, but particularly for the new Optical Electronics Division. One measure of the division 's effectiveness will be how well it observes new technology, invents measurement solutions for emerging applications, and provides standards and traceability for products in the manufacturing phase. The new division can play a major role in meeting this challenge.

The following are the panel's recommendations for the Optical Electronics Division.

• The Optical Electronics Division should consider potential needs related to displays, optical computation/interconnect, ultrahigh-speed (tera-bit) technology, and soliton measurements.

• The division should support a moderate effort on emerging technologies with fundamental needs for new measurement concepts, including high-speed technology up to 60 GHz, measurement needs for wavelength division multiplexing, and optical storage.

• The division should maintain or expand significant efforts on technologies already in the manufacturing phase: (1) fiber measurements (expanded effort in dispersion measurements and standards, EDFA component characterization, and nonlinear effects in fibers [EDFA; wavelength-division multiplexers, WDM]); (2) integrated optoelectronics (modulators

[electro-optical converters], polarizers/depolarizers, WDM filters, and frequency standards in the optical communications bands); (3) measurements for sources, detectors, and amplifiers (expanded effort for noise measurements, EDFA characterization); (4) power measurements (maintain progress in improved power meter accuracy and standards); and (5) optical sensors (build cooperative efforts with other NIST laboratories).

• The division should continue to assess customer needs by maintaining direct contact with all major optical electronics industries and participation in standards groups.

• The division should develop an external funding strategy for mature programs in order to use internal funds to expand support for critical areas, particularly those entering the manufacturing stage.

• The division should clearly delineate the EEEL role relative to that of the Physics Laboratory in detector characterization and narrow-line laser diode characterization. Industrial pull on the existing capability should be a major consideration in resolving issues of organizational overlap.

• NIST requires a strategy for meeting the needs of the laser industry in industrial and medical applications. The Optical Electronics Division alone does not have the resources to support needs ranging from short-wavelength lithography to medical welding and cutting applications. The division role should be defined in the context of the entire NIST effort.

Given below are some of the panel's fiscal year 1993 recommendations concerning the programs in the new Optical Electronics Division (quoted from the fiscal year 1993 assessment), with the division's responses.

• “The Laser Power and Energy Measurement project's activities in the definition of parameters to characterize non-Gaussian laser beam profiles should continue, and efforts in the optical-fiber power meter area should continue to investigate issues associated with reflections from fiber connectors. The transmittance measurements on samples of laser-safety eye wear are also significant. It is important that the excimer calorimetry system and the Nd:YAG laser calibration system come on-line as expeditiously as possible, to be responsive to industry requests” (p. 60). The division is continuing a laser-beam-profile round-robin with industry to help industry improve its measurement accuracy. Work on optical fiber power meters continues, and during fiscal year 1993, a NIST-developed system was installed at the Air Force's Primary Standards Laboratory. The division plans to complete a round-robin in this measurement area within 2 years. A round-robin on connectors is being conducted in cooperation with the Telecommunication Industry Association and is expected to be completed within fiscal year 1994. Excimer laser and the Nd:YAG systems were scheduled for completion in fiscal year 1994. NIST has completed the work that it needs to do in laser-safety eyewear and will continue to monitor the situation to see if additional work is necessary.

• “Specially grown glasses have produced promising results, and a significant level of effort should continue. The lasing properties of these glass guides should be investigated, with a significant focus on understanding the implications for amplifiers and addressing material and measurement issues” (p. 62). The division has continued work on promising devices, with particular attention to their operation at telecommunications wavelengths. Since the 1993 assessment, a laser operating at 1557 nm, 1350 nm, and 900 nm was demonstrated.

• “Because of the potential importance of LiTaO3 devices, the Optoelectronics Metrology Group should collaborate with integrated optic manufacturers, users, and material suppliers to develop characterization methods to qualify and improve wafers. As a result of the embryonic state of guided-wave devices, the group has a good opportunity to be involved in the development of the technology from the beginning” (p. 62). EEEL finds a strong consensus in industry that better measurements and data on LiTaO3 are needed. Interest in LiTaO3 devices, including work on visible sources, also seems to be increasing. The division is responding by initiating a materials characterization effort in fiscal year 1994.

• “To examine sensors beyond those that are intensity-based, such as frequency-based sensors, expand the new approach to remote self-calibrating systems. Supplement outside funding with more internal funding directed toward some new sensor applications such as distributed, environmental, or chemical sensing. Also, strengthen theoretical work to provide new impetus to experimental ideas” (p. 63). In optoelectronics, EEEL has given optical telecommunications the highest priority, followed by optical enhancements of computing and optical sensors. Internal funding for generic sensor development and evaluation will be limited for the foreseeable future. Electro-optic sensor development is generally done by the organizational unit that has responsibility for the quantity of interest; thus, the Chemical Science and Technology Laboratory is developing electro-optic chemical sensors. The Optical Electronics Division's project focuses on generic sensor data and characterization techniques and serves as a resource to other NIST programs, including specific sensor development responding to unmet needs for electrical measurements appropriate to other EEEL programs.

The Cryoelectronic Metrology Group and the Superconductor and Magnetic Measurements Group, which were being reorganized into a new division at the time of this assessment, invent, develop, use, and transfer to others standards and measurement techniques based on cryogenic phenomena and magnetism that can be used broadly throughout industry.

The strategy of the new Cryoelectronic Metrology and Superconductivity and Magnetics Division is based on the recognition that both superconductivity and magnetics will be sources of new advances in many areas of electrical engineering and electronics. The division has established a program to develop measurement techniques in these areas and to provide basic design data

describing properties of magnetic materials and practical superconductors being developed by industry for large magnets, electrical machines, and magnetic recording. Another program applies the unique properties of tunnel junctions and superconductors, and particularly the Josephson effect, to the advancement of NIST's basic mission of providing fundamental standards and physical measurements to create new fundamental physical standards of quantities such as voltage, current, and microwave power. The magnetics strategy also includes development of high-spatial-resolution measurements of magnetic fields, topography of components in storage devices, and exploratory work on sensors for readback in magnetic recording based on giant magnetoresistance. Expansion of work into a broader range of topics for the storage industry was proposed in the fiscal year 1992 magnetics initiative but was not carried out.

The new Cryoelectronic Metrology and Superconductivity and Magnetics Division consists of the Cryoelectronic Metrology Group and the Superconductor and Magnetic Measurements Group. Forty-nine total staff members are being carried forward into the new division, representing 39.6 full-time equivalent staff. The fiscal year 1993 funding for the two groups was $6.9 million ($4.3 million from STRS funds, $379,000 from ATP funds, and $2.2 million from other sources). For fiscal year 1994 the new division will receive a 4 percent increase in funds to $7.1 million ($4.9 million from STRS funds, $475,000 from ATP funds, and $1.8 million from other sources).

The panel was presented with varying statements of the Cryoelectronic Metrology and Superconductivity and Magnetics Division's superconductivity mission and strategy and consolidated them (see above). One concise statement that is both descriptive and prescriptive should be developed and communicated. The panel believes that such a statement will help focus activities on the core mission, a particularly important step in a time of constrained resources.

The emerging division is planning no growth in staff in the area of superconductivity. The panel concurs with this decision because it supports NIST's intention to increase activities in strategic areas, such as magnetic storage technology, for which there is an enormously large U.S. industry. The panel cautions, however, that as internal resources are shifted, superconducting activities must remain at a level sufficient to meet the core mission in that area of technology.

Supporting and extending the Josephson voltage standard, developing new quantum-based current and capacitance standards, developing techniques, instruments, and standards for wires and tapes to be used in large-scale superconducting magnets, and developing ultrasensitive radiation detectors are activities that are at the core mission of this emerging division. These areas were world class when part of the Electromagnetic Technology Division; they should be a continuing emphasis as the division is split.

The panel believes that both basic research in materials and generic support of integrated superconducting thin film technology are also vital for this division, most particularly as that research supports and extends divisional work in standards, materials, and detectors or is closely

and meaningfully coupled with major application problems. Resources dictate that other research problems must be very carefully selected and pursued. For example, the panel questions the strategic value of work on tunable ferroelectrics and further work on superconductor-normal and superconductor-semiconductor contacts, in particular thin film geometries. Although the panel recognizes that the division is capable of excellent scientific work in these areas, limited resources could be better deployed to support the core mission. Examples of higher-impact research problems that are more in keeping with the division's mission include such areas as research to enable high-current, low-resistance contacts to high-critical-temperature (high-Tc) wires and superconducting splices between high-T_c wires.

The current projects and measurement capabilities associated with magnetics provide a good base for characterization of bulk and thin film magnetic samples. The division has documented opportunities to contribute to information storage technology in Measurements for Competitiveness in Electronics (NISTIR 4583). Work focused on storage applications of magnetics should be reexamined for its ultimate impact on industrial competitiveness.

Since the panel's fiscal year 1993 assessment, members of the Cryoelectronic Metrology Group made several major strides towards quantum-based current and capacitance standards, including 0.5-ppm counting accuracy in a five-junction electron pump, as verified by counting with a single-electron interferometer. Key noise sources were identified and eliminated to reach this stage. The next step to achieve metrological accuracy is a seven-junction pump.

The Single Electron project, the result of a competence-building effort now nearing the end of funding, has established this division as a world leader in this area of metrological applications. The potential is clearly present for EEEL—and ultimately for U.S. industry —to take the world lead in applying these quantum standards, as is done for the series-array voltage standard. It is absolutely vital that funding and staffing of this project be stabilized as the competence-building funding ends. Unfortunately, it appears that important NIST-wide funding initiatives in quantum standards aimed at building up more broadly based expertise in this area have only peripherally included this project, and the project's funding and staffing appear to be insecure.

The division's support and advancement of the Josephson array voltage standard had excellent accomplishments along a broad front since the fiscal year 1993 assessment. There is now a commercial supplier for 1-V standard chips, which NIST certifies and is making a standard reference material. Commercial attempts to fabricate 10-V chips have not yet yielded working parts, but EEEL has fabricated enough 10-V chips now to maintain a comfortable inventory for supporting worldwide use of the 10-V standard. This division is still the only laboratory in the world producing such chips. Excellent support has also been provided to users of the NIST standard, which now number more than 35 national, commercial, and military standards laboratories. U.S. companies now supply complete voltage standard systems traceable to NIST, a profitable business.

Promising new research results in the division include the successful demonstration of a rapidly programmable Josephson voltage standard that may eventually serve as a high-accuracy root-mean-square voltage standard, possibly replacing thermal converter standards. Another

important accomplishment since the panel's fiscal year 1993 assessment was the development and delivery to several companies of supercurrent simulation circuits. These circuits will enable companies to make measurements that ensure meaningful and useful product specifications for superconducting wires and tapes. Worldwide round-robin tests have been conducted as part of a quality assurance program, and an EEEL scientist was chosen to chair the Technical Committee on Superconductivity of the International Electrotechnical Commission.

Valuable progress and accomplishments were also observed in the development of kinetic inductance radiometer devices for ionizing radiation detectors and spectrometers, the achievement of extremely low contact resistance and high current densities in high-T_c superconductor-normal metal thin film contacts, and the production of high-T_c devices incorporating junctions and ground planes as required for high-frequency applications. Particularly noteworthy is the stabilization of the low-T_c junction and integrated-circuit processes that are reliable for the low-T_c electronics projects as well as for the single electron tunneling project.

Technical work under way in magnetics reflects the excellent skills and talents of the individuals involved and the group's management. The group's scanned probe microscopy work is of high caliber, and efforts have been made to provide support and service to the storage industry near the locale of the Boulder Laboratories in Colorado. Planned expansion to cover the industry more broadly will enhance the group's visibility and vitality.

The group's imaging of head fields and domain structures is on a par with that of other research groups. Work within EEEL should be focused, however, toward a goal that will solve a particular problem relevant to magnetics or storage, such as Barkhausen noise in thin film heads, or understanding of noise spectra of thin film media from domain wall structures.

The group's work on giant magnetoresistive materials and structures is very good. However, there may be issues more relevant to magnetoresistive heads in the near term, such as biasing, stability, and electrical damage.

The group is uniquely positioned to be a center of competence for all storage-related work in the federal government. This would provide important technical leadership to the other NIST laboratories and a focal point for other agencies involved with the storage community such as the National Science Foundation, the Department of Energy, the Advanced Research Projects Agency, and DOD. It may also reduce some duplication of effort and provide better utilization of resources. Many of the support or “fire-fighting” tasks the group has performed for government and industry are of great value. Better documentation of these specific projects in the group's annual report and plans would be beneficial.

The following are the panel's recommendations for the Cryoelectronic Metrology and Superconductivity and Magnetics Division.

• The new Cryoelectronic Metrology and Superconductivity and Magnetics Division should have one mission and strategy statement that is well understood by all staff. This should focus work on fewer, key projects and lead to stable staffing for those projects. Resources in superconductivity should be limited, though the group's many technical successes in that area tend to spawn ever-spreading activities.

• The single-electron tunneling activity needs to switch from competence-building funding to stable core funding. Funding and staffing for this project should be stabilized at an adequate level and this project made a core component of any NIST-wide activity in this area.

• With the development of superconducting materials and instruments that operate without a liquid cryogen, such as the Nb₃Sn magnetic resonance imaging device recently announced by a major U.S. corporation, there is an enormous need for techniques and procedures to give quality assurance in the 10 to 77 K range. New protocols and standards will be needed as more instruments are manufactured to operate in this temperature range. It is important that EEEL anticipate and adequately support the needs of the new division in this area.

• Work on high-T_c aterials and processing should be focused on specific processes and device structures related to the division's core mission. More genetic activity should be deemphasized, since it is less likely to have an impact and will greatly dilute resources.

• The division should obtain a sample preparation apparatus, or convert an existing one, for dedication to magnetics.

• The division should decide if storage-oriented work should be expanded according to its fiscal year 1992 plans; if not, it should be discontinued. If continued, this program should (1) become a recognized center of competence in magnetic recording, following the example of the superconductivity group; (2) pick one area in storage, perhaps heads, and become intimately familiar with process steps, metrology requirements, and technical challenges (Barkhausen noise detection and measurements, electrostatic damage in magnetic recording heads, biasing, and so on); (3) establish a regular seminar series in magnetics with recognized experts as speakers; (4) become a member of and interact strongly with industrial storage standards committees such as the American National Standards Institute to become more familiar with metrology requirements in the industry; and (5) vigorously pursue participation in ATPs in data storage.

Given below are some of the panel's fiscal year 1993 recommendations for the program in the new Cryoelectronic Metrology and Superconductivity and Magnetics Division (quoted from the fiscal year 1993 assessment), with the division's responses.

• “The atomic-force microscopy research should be located close to the magnetism and sample preparation facilities” (p. 57). NIST has been given funding to construct the buildings needed to maintain its present program into the next century. Until the new facilities are completed, management and staff must work together to use inadequate space in the best ways possible.

• “The Cryoelectronic Metrology Group should continue its correct competence building project under core funding. A second permanent staff position should be added, and the two

postdoctoral fellowships should be continued” (p. 56). The presumption of the NIST competence program is that successful competence projects are continued by the division using its available funding. Thus, a competence project carries with it the requirement that the division grow or reprogram its work and is part of NIST's system to assure turnover of projects. The division intends to continue this successful program. Because the competence project requires continued funding from the division, EEEL generally attempts to transfer staff from other projects to complete the work rather than hiring additional staff.

• “The Cryogelectronic Metrology Group should emphasize demonstrations and applications of cryoelectronics designed for optimum performance at liquid-nitrogen temperature. However, because of staffing constraints, the group should probably set priorities for the use of the group 's unique talents” (p. 56). The division is actively developing applications of high-temperature superconductivity. Advances since the fiscal year 1993 assessment include the fabrication of step-edge Josephson junctions, demonstration of a tunable 5-GHz resonator, and demonstration of the phase locking of junctions at 1.06 THz. During fiscal year 1994, the division expects to develop additional microwave components, microbolometers, and arrays of Josephson junctions.

• “The Superconductor and Magnetic Measurements Group should focus on quantitative measurements of stray magnetic fields arising from bit recording and magnetic recording heads. Furthermore, the group should acquire a spinstand, i.e., a magnetic recording system, that can record any desired bit pattern ranging from isolated bits to complicated patterns” (p. 58). The division is continuing its work in the development of scanned probe microscopy to measure magnetic fields on the nanometer scale. During fiscal year 1993, the staff worked with a local company to evaluate their spinstands. Preliminary results suggest that commercial spinstands do not have sufficient versatility to be effective in the NIST program. The division has initiated a program to develop and construct a more versatile instrument based on the spinstand concept.

The panel found the Office of Microelectronics Programs (OMP) to be strategically linked to related industries through ties to the SIA, especially in support of SIA Roadmap. OMP is very effective in placing its funds across NIST laboratories in support of key industry needs. Significant increases in planned funding are in keeping with national need. Projects currently under way with OMP participation or sponsorship are well chosen, and the National Center for Semiconductor Metrology (funded at the $25 million level) appears to be well structured and appropriate.

The panel is concerned that a smooth transition be made from the present level of activity to the expanded levels anticipated for the future. Given the past performance of OMP and its close working relationships with industry and all NIST laboratories, the panel judged that the likelihood of success will be high as work expands.

The Office of Law Enforcement Standards is closely linked to the law enforcement customers it serves, and the recent execution of a stable, long-term relationship with the National Institute of Justice (NIJ) has provided needed assurance of ongoing support.

The staff of the office does good work in the variety of areas in which customers have sought support. The office continues to serve as a unique leader in many key areas and serves as a national asset to the law enforcement community.

The previous leader's retirement has been handled smoothly, although the elevation of a key technical contributor into the management position has resulted in loss of technical competence that needs to be replaced as soon as possible.

Now that uncertainties about NIJ funding have been resolved, there is a need for funding for maintenance and renovation of the office 's test and storage facilities.