We now examine some of the difficulties for the application of biosensors online. A workshop on the use of biosensors for on-line analysis (Scheper et al., 1994) emphasized that the biosensor should be considered as an integral part of the bioprocess as a whole. The workshop indicated that much work needs to be done before biosensors can be used on-line, especially with regard to the "harsh conditions" to which some of these biosensors may be subjected. The workshop also pointed out some of the drawbacks that arise when biosensors are used in situ: (1) the biosensors cannot be sterilized; (2) they function within a limited range of analyte concentrations; (3) if an enzyme is used in the detection process, its pH optimum is different from the pH optimum of the process. The workshop further emphasizes that since information from a wide variety of disciplines is required and the process itself is very time consuming, it may be advisable to design a biosensor in a modular fashion and then tailor it to meet specific analytical protocols. Also indicated was the need for the incorporation of "intelligent" software to run the analytical systems.
In a guest editorial, Weetall (1996) indicates that, in general, biosensors suffer from quite a few problems that need to be overcome before they can be used on a large scale. In essence, the biosensor should be inexpensive, user friendly, sufficiently sensitive and accurate, and easily manufactured with high select rates. Weetall emphasizes that the biosensor technique will be competing with the more established clinical technologies that are already inexpensive, rapid, automated, multianalytic, accurate, and sensitive. Furthermore, this author indicates that the true test for biosensor competitiveness is the cost per test; the more labor required, the greater the cost.
Weetall further indicates that electrochemical sensors can measure analytes in the 10~6M concentration range, a range that is sufficient for measuring glucose, urea, cholesterol, drugs of abuse, and many large and small molecules. Weetall emphasizes that biosensors have difficulty measuring analytes in the 10-9M range, the range required to measure hormones and other serum components. Thus sensitivity is an issue, as indicated earlier in this chapter, that still requires improvement.
Weetall cautions, with tongue in cheek, that many scientists developing biosensors do not consider the manufacturing aspects of the device they want to work with. But this aspect is essential and should be brought into consideration early in the development process. Finally, this author predicts the use of biosensors in the next 15 years in the following areas: single-molecule detection, nano-size sensors, multianalyte arrays, home imaging systems for wellness screening (using noninvasive biosensors), and interface between the human central nervous system and artificial intelligence (using implantable sensors).
Was this article helpful?