GENERAL CONCLUSIONS AND RECOMMENDATIONS
The general conclusions and recommendations contained in this chapter, relating to sensor research and development, are the result of the committee's analysis of the illustrative examples and discussion contained in parts I and II of this report. Specific sensor materials development opportunities derived from the illustrative examples of Part II are summarized in Chapter 7.
Trends in Sensor Technology
Current sensor development is tending toward increased complexity in sensor systems. The greater flexibility and lower production cost associated with advanced, integrated electronic technology allows computer processing that once required large and sophisticated signal processing systems to be reduced to a microelectronic chip; for example, smart sensors have transduction, signal amplification, filtering, and other processing on a single substrate. However, from the perspective of the end user, the sensor system now appears simpler even with its increased functionality and internal complexity.
The principal technical drivers for sensor development may come from enabling/supporting technologies other than materials technology. Most recent advances in sensors have not originated from the synthesis of new transduction materials (except perhaps for chemical sensors) but from innovations in low-cost, large-scale manufacturing of interconnections, microelectronics, and micromachining. Many advanced sensor techniques, such as photon-scattering and laser acoustic technologies, require materials developments to support particular implementations, not the sensor transducer.
Networking of large sensor systems can provide improved spatial and temporal sampling in low-cost, low-maintenance systems. A network of sensors distributed throughout a large structure can provide data to a central processor that monitors performance or aids in locating and characterizing structural defects. In other cases, such as chemical sensing, the individual outputs from an array of sensitive but only moderately selective transducers can provide a composite indication that is both sensitive and selective with regard to a target chemical species.
Sensor research and development lends itself to dual use and commercialization efforts. Sensors are an enabling technology, applicable to a wide spectrum of uses. To be effective, it requires identification of potential uses and assessment of the degree of suitability. For example, sensor systems developed for structural health monitoring of an aging military aircraft or for other vehicle monitoring applications can be exploited in some form by the commercial aircraft and automotive industries. Chemical sensors used for detection of chemical warfare agents have numerous possible non-DoD applications in areas such as environmental and health monitoring. Also, infrared sensors, traditionally developed for military applications such as reconnaissance,
are finding uses in materials manufacture, intrusion detection, and chemical detection systems as they become affordable.
Few programs have existed to develop sensors solely for the sake of advancing sensor technology. Historically, sensor research and development efforts have been funded as an adjunct to large application programs that required sensors. A concentrated effort to support the advancement of sensors and the development of new or improved sensor materials will require the implementation of an effective research planning process that addresses the needs of a broad set of users with related applications.
There is a need for a generally accepted framework to describe both sensor application requirements and sensor performance capabilities. The establishment of a common set of descriptors for use by sensor users and suppliers and for researchers in the diverse disciplines associated with sensor development was identified by the committee as the most important step in facilitating the identification of sensor materials R&D opportunities and in accelerating the development and use of advanced sensor technologies.
Experience in establishing centers of excellence for sensor development provides useful guidelines for improving sensor R&D strategy. Four characteristics appear to be essential: a multidisciplinary approach with emphasis on teamwork; capabilities ranging from an initial proof of concept, exploratory and developmental research through engineering prototypes; focus on selected sensor technologies for a broadly defined range of applications in line with the core competencies of the organization (i.e., not attempting to cover the entire field of sensor technology and the associated diversity of sensor materials); and strong linkage to industry to guide the general relevance of the research.
Opportunities in Materials R&D
Sensor materials R&D can be divided into two main categories: the development of new materials, and materials engineering associated with implementation constraints for particular applications. These two categories frequently have very different approaches to materials R&D. New sensor materials are often targeted at very innovative, high payoff applications for which the requirements are ill defined. In contrast, materials engineering issues, such as longevity, resistance to a hostile environment, and incorporation in a host structure, are associated with sensor implementations in defined applications with relatively well-known requirements.
High-payoff opportunities for new sensor materials in the near term will come primarily from R&D on existing materials rather than synthesis of new compositions of matter. The committee identified fundamental research on new compositions of matter as the highest risk element of sensor materials R&D programs, although this approach also offers the highest potential payoff. Experience has indicated that the introduction of new materials has created many new and exciting opportunities for sensors. Nonetheless, the development time for a commercially available sensor based on a new composition of matter may well be greater than ten years, indicating that, although the opportunities are enormous, the associated costs will likely be high.
Exploiting materials developed for purposes other than sensing can lead to rapid sensor technology advancements at relatively low cost and risk. As a case in point: fiberoptic sensors are now commonly employed in many applications. Optical fiber development was supported by the communications industry; performance was greatly improved and cost decreased rapidly as usage increased. The sensor community very creatively took advantage of these materials without the need to support a major development activity of the fibers.
Materials processing science will be the foundation for developing affordable sensor materials. Materials synthesis and processing will facilitate the transfer of innovations in materials science to commercially viable products. Numerous existing materials could be incorporated in commercial sensors if the material could be produced at low cost in the needed configurations and quantities.
Universities and federal research laboratories play a critical role in conducting frontier research. Frontier research can be defined as leading-edge research that is conducted without a particular application
in mind or does not expect to be commercialized within the near or intermediate term (e.g., within 10 years). While industrial research centers are having increasing difficulty in justifying such research, universities are well positioned to conduct such research and to use such programs as vehicles to educate students. Federal research laboratories, including the military laboratories, generally sponsor frontier research and conduct a portion of the research in-house to keep abreast of the leading-edge technologies. Long-term commitment to such research is essential to maintain a stable effort.
In the view of the committee, focused programs in which sensors are treated as a separate field of endeavor, as opposed to an adjunct to larger programs, will contribute significantly toward accelerating the development and use of advanced sensors. The primary R&D approach to exploit high-payoff opportunities in sensor technology should be the multidisciplinary integration of existing technologies (e.g., transducer materials, signal processing, and packaging) for specific or generic applications. This will allow for aggressive investigation of new sensor technologies from a system perspective. Frontier research activities should therefore be closely coordinated with in these applied R&D programs.
The committee recommends use of a communication tool to facilitate interdisciplinary cooperation and the identification of research opportunities and needs for sensor systems and technologies. A standard terminology is needed to describe sensing requirements and technology attributes. The communication tool and descriptors should form the basis for an evolutionary methodology that can be augmented and refined for use by specific research groups and applications.
The committee recommends that the communication tool be used in conjunction with additional information on technical risk1 and potential payoff in developing a decision-making methodology to guide sensor materials R&D. Guidelines to facilitate such R&D planning could include the following steps:
a systematic approach to identifying research opportunities based on a comparison of requirements to available technologies;
realistic assessment of technical risk and challenges;
an estimate of potential benefits; and
effective communication of sensing needs and capabilities across the diverse technical disciplines involved in sensor development.
Organizations undertaking sensor R&D programs should maintain a broad research base with critical core competencies. Because the emergence of sensors as a field is the result of the convergence of activities from physics, chemistry, materials science, and engineering, a multidisciplinary approach to R&D is particularly important, both in identifying opportunities for future development and in conducting R&D in the field of sensor technology.2 Thus, organizations undertaking R&D programs in sensors should possess, or have access to, expertise from all technical disciplines involved in sensor technology. To give a sense of relevance and urgency to any applied R&D program, a customer or end user with a specific implementation need should be identified and charged with demonstrating the potential payoff in a joint effort with the developer.
R&D programs that develop sensor materials should focus on selected classes of materials. In view of the diverse range of sensor materials and the high costs of fabrication facilities for many advanced sensor materials, the committee recommends that specific R&D programs focus on selected classes of materials, such as semiconductors or ceramic-based materials, rather than attempting to encompass a very broad range of endeavor.
Priorities in Materials R&D
The committee recommends that sensor materials R&D be pursued in three main areas. In order of decreasing priority these are:
development of processing techniques for existing sensor materials;
assessment and development of sensing capabilities
in existing materials that have properties not yet exploited for sensor applications; and
fundamental investigation of novel sensing approaches, such as using multiple physical responses to a sensing phenomenon and new compositions of matter.
The committee recommends that this prioritization be used as a guide in allocating resources, with the highest priority category receiving the largest share in aggregate. The lowest priority would also receive resources, although in aggregate not at the same levels as the categories above it.
NRC (National Research Council). 1986. New Horizons in Electrochemical Science and Technology. NMAB-438-1. National Materials Advisory Board, NRC. Washington, D.C.: National Academy Press.