The initial shock of a major earthquake lasts just a short time, tens of seconds, although it can seem forever for those trapped by its shaking.

A tsunami wave train lasts minutes to hours before ocean waters re-stabilize after their initial disturbance.

But the persistence of radiological disasters is measured in decades.  Decades in which the principles of entropy apply to the dispersion of the highly concentrated radioactive materials found in damaged nuclear power plants, for example, until the materials make their way from containment into the environment at large, a “gradual decline into disorder” with potentially catastrophic consequences.

It is just such a catastrophe that is now Japan’s tragedy to confront as a result of the March 11, 2011 Great East Japan Earthquake (東日本大震災) and the subsequent (and continued) meltdown of three nuclear reactors at the Fukushima Dai Ichi Nuclear Power Station.

MEF 2015 OverallThe MEMS Engineer Forum 2015 (April 20 and 21 2015, Tokyo, Japan) featured a superb presentation by Dr. Yang Ishigaki, CTO of Yaguchi Electric Corp., and founder of, on Participatory Environment Monitoring Developed with Social Media: Action Research for Radiation Monitoring in Japan.

In an emergency situation, where information being released by local and national governments about immediate environmental risks (fine-particulate matter air pollution; air or water-borne pathogens; radiation) is potentially unreliable, or, worse yet, suspected of being false, and where the sensors to measure events may not be readily at hand, what is the appropriate response by at-risk citizens?

For Dr. Ishigaki and his team, the answer to the specific problems posed by the release of radioactive material from the Fukushima Dai Ichi Nuclear Power Station was to develop, using Kickstarter – crowd-sourced – participatory development methods, a smartphone-based pocket Geiger counter.

From “Our mission is to make radiation information available to all people. Children, mothers, senior citizens – everyone has a right to know the true radiation level in their residing cities. Learning the radiation level should be [as] easy as listening to the weather forecast or reading a thermometer. To accomplish this mission, we are now developing a simplified survey meter (dosimeter), and working to supply it steadily at a low price.”Pocket Gieger Image

By using a cheap PIN photodiode, rather than the expensive and less durable Geiger-Muller tubes or scintillator crystals found in conventional radiation monitoring devices, Radiation-Watch was able to greatly reduce the cost and size of its sensor, making the Pocket Geiger truly pocket-sized, and truly affordable; and bring the Pocket Geiger to market in the space of three months, with radiation level data being shared extensively, and publicly, in Japan and in other parts of the world.

Dr. Ishigaki’s lesson at MEF 2015 is a doing-well-by-doing-good story; giving residents the (cheap) tools to evaluate, from an environmental quality standpoint, their locales, and then creating for those same residents the online community in which the quality-of-life data can be shared as easily as a Facebook post.  It’s a powerful lesson, and is a working model that extends far beyond the unfortunate need to monitor radiation levels in Japan over the next several decades.

Dr. Weileun Fang, National Tsing Hua University, Taiwan, had another of the important lessons shared at MEF 2015: Dr. Fang’s was the lesson of frugality, or a kind of frugality, based on building MEMS structures and devices directly into fully-processed TSMC CMOS wafers: 180nm process node – 200mm wafers, the kind of wafer that TSMC builds in its sleep, on the cheap.

“CMOS MEMS: Technologies for the Next Big Things – IoT,” Dr. Fang’s paper, barks up the same tree as the recent publication by 3D InCites’ own Bill Martin and his co-author, writing about how “IoT Requires the Evolution of the ‘New’ 200mm Fab.”

Matsumoto-joUsing wafers baked in TSMC’s standard CMOS process lines, Dr. Fang and his group, with follow-on MEMS fabrication process steps, have been able to create integrated CMOS MEMS motion sensor hubs, integrated CMOS MEMS environmental sensor hubs, integrated CMOS MEMS resonator hubs, and integrated CMOS MEMS acoustic / optical hubs.

It’s not exactly monolithic 3D in the classical sense, at least as monolithic 3D has been presented in 3D InCites; it’s actually more powerful than that, so long as the MEMS processing does not adversely affect the CMOS.

It’s the best kind of piggybacking, and offers the potential for the best kind of frugality (IP reuse) in the sense that the “chasing-the-latest-technology” business model is not going to be the answer for the IoT, at least from the semiconductor / MEMS sensor standpoint. (See more about this in “IoT Anxieties in McKinsey-GSA Study,” by Junko Ishida, EETimes.)

We hope to hear more from Professor Fang in a future 3D+ post.  We need more lessons like his from MEF 2015.

The final lesson from the MEMS Engineer Forum 2015 is a lesson on the rewards of perseverance, and of vision, and it goes something like this.

In the early days of semiconductor fabrication many of the big companies, most notably IBM, made their own processing tools.  The thinking was that the special capabilities home-grown tools gave a chip manufacturer were as much a competitive advantage as was the design or functionality of a given product.

It wasn’t just the features of an IBM processor itself that gave IBM big iron its lead in data centers and data processing; it was also the way the processing chip was fabricated, something which was maybe inseparable from the chip’s performance.

(Oh, and another advantage: nobody ever got fired buying IBM.)

Were wafer fabrication tools one of IBM’s core competencies? Ultimately, IBM decided no, and began to retire its IBM-only processing gear for commercial off-the-shelf units.

Much to the semiconductor capital equipment industry’s gain.

In MEMS, famous for “one chip – one process” thinking, the original (and still best) deep silicon etching process was invented at Robert Bosch, in Germany, and initially used solely for Bosch commercial MEMS products.

The Bosch Process was surely a great competitive advantage for MEMS fabrication at Bosch, and yet, the story goes, as Bosch was not at its core a MEMS capital equipment company, maybe it would be better, in Bosch’s thinking, if Bosch were to license the technology to interested parties in the capital equipment business, and thereby, in the end, be availed of much better equipment.

So a deal was struck.  Enter Mr. Susumu Kaminaga of Sumitomo Precision Products, Japan, who became involved with MEMS in 1988 and played a major role running Surface Technology Systems (STS), UK, a subsidiary of SPP, since its acquisition by SPP in 1995.Kaminaga-san

Starting in 1994, STS pioneered development and commercialization of Deep Reactive Ion Etching (DRIE) technology based on the Bosch Process, and, under Kaminaga-san’s management, shipped the first commercial Bosch Process tools in 1995.

Many, many DRIE tools later, from STS / SPTS (now part of Orbotech), and others, Bosch Process DRIE technology has enabled the MEMS world to expand rapidly in the last decades, as a result of this key enabling technology being available to the wide commercial market.

And how did that work out for Bosch?

Well, Bosch is still the leader in MEMS.  According to IHS, “Thanks to design wins in Apple products, Bosch maintained its leadership in microelectromechanical systems (MEMS) in 2014, with sales growing nearly 17% to reach $1.17 billion … In a fragmented market, Bosch increased its market share to 12% in 2014, up from 11% in 2013.”



The lesson is one of perseverance and one of vision.

As presented at the MEMS Engineer Forum 2015.

From Pittsburgh, PA, thanks for reading ~PFW





Paul Werbaneth

Paul Werbaneth is a long-time Contributing Editor at 3D InCites. Since entering the semiconductor industry…

View Paul's posts

Become a Member

Media Kit