Proceedings Article | 16 August 2004
KEYWORDS: Packaging, Molecular nanotechnology, Semiconductors, Nanotechnology, Standards development, Microopto electromechanical systems, Sensors, Head, Nanolithography, Electronics
Forecasters and analysts predict the market size for microsystems and microtechnologies to be in the order of $68 billion by the year 2005 (NEXUS Market Study 2002). In essence, the market potential is likely to double in size from its $38 billion status in 2002. According to InStat/MDR the market for MOEMS (Micro Optical Electro Mechanical Systems) in optical communication will be over $1.8 billion in 2006 and WTC states that the market for non telecom MOEMS will be even larger. Underpinning this staggering growth will be an infrastructure of design houses, foundries, package/assembly providers and equipment suppliers to cater for the demand in design, prototyping, and (mass-) production. This infrastructure is needed to provide an efficient route to commercialisation. Foundries, which provide the infrastructure to prototype, fabricate and mass-produce the designs emanating from the design houses and other companies. The reason for the customers to rely on foundries can be diverse: ranging from pure economical reasons (investments, cost-price) to technical (availability of required technology). The desire to have a second source of supply can also be a reason for outsourcing. Foundries aim to achieve economies of scale by combining several customer orders into volume production. Volumes are necessary, not only to achieve the required competitive cost prices, but also to attain the necessary technical competence level. Some products that serve very large markets can reach such high production volumes that they are able to sustain dedicated factories. In such cases, captive supply is possible, although outsourcing is still an option, as can be seen in the magnetic head markets, where captive and non-captive suppliers operate alongside each other. The most striking examples are: inkjet heads (>435 million heads per year) and magnetic heads (>1.5 billion heads per year). Also pressure sensor and accelerometer producers can afford their own facilities to produce the numbers they want (several millions per year). The crossover point where building a dedicated facility becomes a realistic option, can differ very much depending on technology complexity, numbers and market value. Also history plays a role, companies with past experience in the production of a product and the necessary facilities and equipment will tend to achieve captive production. Companies not having a microtechnology history will tend to outsource, offering business opportunities for foundries. The number of foundries shows a steady growth over the years. The total availability of foundries, however, and their flexibility will, undoubtedly, rely on market potential and its size. Unlike design houses, foundries need to realise a substantial return on the "large" investments they make in terms of capital and infrastructure. These returns will be maximised through mass-produced products aimed at "killer" applications (accelerometers are only one example). The existence of professional suppliers of MOEMS packaging and assembly is an essential element in the supply chain and critical for the manufacturing and commercialisation of MOEMS products. In addition, the incorporation of packaging and assembly techniques at the front-end of the engineering cycle will pay back in terms of financial savings and shorter timescales to market. Packaging and assembly for MOEMS are, in general, more costly than their equivalents for standard integrated circuits. This is, primarily, due to the diversity of the interconnections (which are multi-functional and may incorporate: electrical, optical, fluidic etc). In addition, the high levels of accuracy and the potential sensitivity of the devices to mechanical and external influences play a major role in the cost aspects of the final MNT product. This article will give an overview of the package/assembly providers and foundry business models and analyse their contribution to the MOEMS supply chain illustrated with some typical examples. As we believe that commercial services are the main basis for the breakthrough of MOEMS technology, we only cover commercial package/assembly and foundry services and not the ones offered by universities and research labs.
