MEMS Manufacturability


This Site:

MEMS Technology
MEMS Products
MEMS Adaptive Optics


Up MEMS Design Tools MEMS Manufacturability MEMS Reliability MEMS Image Gallery

The demonstration of the first micromachined motors in the mid-1980's generated huge interest throughout the engineering community.  There was widespread speculation about the revolutionary products that would be enabled by this incredible new technology.  In the two decades since, there has been a vast array of impressive MEMS research and demonstration prototypes. The transition of this work to high value commercial products has been much slower.  While there have been some significant successes, many of the exciting research demonstrations have failed to make it out of the lab and into products.  The pervasiveness of the technology is much less than what had been anticipated at it's inception.  Once of the key barriers to more widespread commercial success has been challenges associated with manufacturing.  MEMS technologies have suffered from a variety of manufacturing issues.  One important issue is the problem of stiction.  Stiction is the fundamental property that small devices want to stick to each other.  Another significant challenge is in the area of MEMS packaging.  Many devices that work properly at the wafer level can not be effectively packaged for use in real applications. 

There has been significant strides in overcoming stiction issues, and achieving manufacturable MEMS devices.  The stiction problem has been addressed at both an engineering and a physics level. Engineering wise, stiction can be minimized by careful consideration during design.  Devices can be designed to minimize stiction by creating devices that a stiff,  and have a minimum area of potential contact.  Stiff devices have a larger restoring force to battle against stiction, and designing devices with dimples to limit contact area reduces the stiction force. On the Physics front treating the surfaces with Self Assembles Monloyars (SAMs) can lower the surface energy of the devices and reduce the stiction forces. Combinations of the Engineering and Physics approaches has led devices with manufacturability approaching that of Integrated Circuits.

The chart below shows the product yield during the development of a complex MEMS-based mirror chip.  It can be seen that sustained yields in the high 90% were obtained. This chart represents over 25,000 devices manufactured.

For this product, the stiction problem was overcome through the use of sophisticated design techniques, and development of a proprietary back-end-of-line release capability.  This combination of design and processing results in stiction free devices suitable for use in real products. 

This data shows that with proper design and process, it is possible to move beyond prototyping and to achieve the levels of manufacturability required for real applications.


Contact Information:

Electronic mail:

Copyright 2003-2008
privacy policy