Microsystems are systems with an overall size in the order of milimeters, with features in the micrometre range. They can have a large variety of applications ranging from sensors, actuators, chemical reactors, structural components to opticals and information processing.
Microsystems technology (MST) has emerged from a combination of IC-processing and conventional miniaturised fabrication technologies. Microsystems have functions in many domains such as electrical, mechanical, chemical, thermal, and/or optical with applications in many areas. Prominent examples of MST products in mass markets are ink jet printer heads, pressure sensors and accelerometers; the total volume in niche markets however is at least of the same size. All over the world a wave of innovations using MST can be seen, in research environments as well as in market development.
MST is characterised not only by the small size of the systems, but also by the requirement of a multidisciplinary design approach, the integration of many functions in a single system and the potential of mass fabrication. Altogether, MST promises cost-effective design alternatives and/or new types of functionality.
In many technical systems there is a strong trend for miniaturisation towards micrometer and even sub-micrometer range. This trend results from two independent facts.
The first one is that small components and systems perform differently. Small systems can perform actions large systems cannot, simply because of their small size. This can be due to limited space or access (example: minimal invasive surgery). Or it is due to related aspects such as low weight, (aerospace applications) and/or energy efficiency (example: mobile applications). In some cases miniature systems can use physical effects at small scales, such as more efficient chemical reactions or negligible influence of gravity.
The second fact is due to the manufacture of Microsystems, which provides unique possibilities. Material cost is negligible and materials can be used that would otherwise result in prohibitive costs. Technology derived from IC-fabrication processes allows the production of miniature components in large volumes for low prices (examples: pressure sensors for automobile applications, ink jet printers). The fact that large volumes can be realised against low costs has opened entirely new fields of application. Distributed systems which hundreds or thousands of microsystems can be realised.
The general nature of MST makes it a useful addition to the existing range of technologies available and it has applications in almost every field imaginable. Although MST is a fairly new technology that is still under development, already successful applications can be claimed. MST product for mass markets include pressure sensors, crash sensors, inertia sensors, ink jet printer heads and read-write heads. A huge number of microsystems is being commercialised for niche markets, ranging from pressure sensors on catheter tips over infrared sensors for cosmology to complete mass spectrometers. However, the number of success stories is still very small if compared to the claimed potential of MST.
In the end of the '80ies of the last century there was a great enthusiasm on MST, based on the vast numbers of potential application of MST. Work on these applications has been restricted mainly to the university groups. Here we find work everything from micro actuators, robots, to more complex microsystems such as micro components for fluid handling. Industrial research and development has not been nearly as involved. The types of MST products now on the market are rather simple systems. We find sensors for pressure, acceleration and angular rate, in term of market volume for the automobile industry. Numerous types of these and other sensors (notably for fluid flow) are in niche markets. Few other MST products are on the market.
We may state that the euphoric expectations of the 80ies have by far not been met.
The reasons for the limited commercial success of MST are many fold. The complex task of designing a microsystem, the long developing time, and the fact that MST at the same time offers new systems possibilities while using new fabrication technologies in combination with unusual materials, presents a great barrier for companies to step into the technology.
The idea is that the fabrication technology enables us to integrate many physical and chemical functions into a compact system. Due to the small details in the miniaturised systems, one cannot assemble a microsystem from components off the shelf, as is often done for macrosystems. First of all, the components are too small to be assembled for reasonable costs. Further, currently there are no components that fit together. In designing a microsystem, all the components have to be designed during the design process of the whole system. There is no design phase of the system separate from the design phases of components. This includes the package and the interface of the microsystem.
This is not only very demanding on the designers, requiring intimate knowledge of the whole system, it also means that one has to design a microsystems for each particular application, starting the design process all over again. An example may illustrate the situation. In CD players a lens is positioned over a groove with bits. The mass of the lens is by far too large to use existing microactuators to move it with the required acceleration. The whole system concept must be changed to build a CD-player using MST. No company is willing to start this effort.
The fabrication of the microsystem is an integral process. This process is being provided by (silicon) micromachining and to a lesser extent by miniaturisation of traditional machining (cutting, drilling ect.). Micromachining is very complex and not developed to a standard technology. The process space is still largely unexplored. Therefore, designing a microsystem typically also involves the design of the corresponding fabrication process. The lack of off-the-shelf modules, technologies and processes make fabrication expensive, prohibiting wide application of micro systems in mid-size and small series quantities. This is quite different from e.g. IC-design and processing. For IC-processes there are strict and clear design rules that guide the circuit designer, without having to worry how the circuit is actually made.
As the design of a microsystem is an integral process, it also includes all system aspects from mechanics to electronics and application specific knowledge. Microsystems designers therefore must have a large number of skills and a rather broad experience. For the development of new microsystems with applicability in domains as biotechnology, medicine, or minimal invasive surgery the multidisciplinary approach of the research group is indispensable. Knowledge in Microsystems design, fabrication technology must meet knowledge of the particular application of the microsystem.
This of course is the interesting aspect of MST: we can design systems with functionality not met by other means. Nonetheless, the scaling of systems properties to the micro scale is a considerable challenge for the engineer (we note that there exists no engineering curriculum at universities in The Netherlands that prepares engineers for this task - some curricular are emerging in Germany and in the US).