Trading Firm Mitsui & Co. Establishes MEMS R&D Subsidiary

Global Emerging Technology Institute

Mitsui & Co. Ltd., a Japanese giant trading firm with an annual turnover of about 12 trillion yen (US$100b) recently announced the establishment of a wholly-owned subsidiary that specializes in research and development of MEMS (micro-electro mechanical systems). The new company, Device Nanotech Research Institute (DNRI), is the fourth venture to be established by Mitsui in regard to its nanotechnology research business development efforts. Mitsui & Co. set up the Bio-Nanotech Research Institute (BNRI) and Carbon Nanotech Research Institute (CNRI) in July 2001. BNRI will research nanoporous membranes and develop relevant process technologies, while CNRI will develop fullerene and carbon nanotube (CNT) volume-production technologies. Mitsui also established INRI in October 2002 to manage intellectual property rights of the affiliated nanotechnology research entities. The latest subsidiary DNRI is headquartered in Tokyo and capitalized at 300 million yen (US$2.5m). Mitsui will expand DNRI's capital to 1.1 billion yen ($9.2m) next year and spend 1.9 billion yen ($15.8m) on R&D activity at DNRI in three years. DNRI currently employs 4 researchers and plans to expand its staff to 15 in the second quarter of 2003.

At present, DNRI is working on 12 research topics, including MEMS devices for use in information & communications, medical & healthcare, and environment & energy applications. For example, in the information & communications category, the company will develop optical signal techniques for use in next-generation ultra-fast optical communication systems. The goal is to develop "signal dispersion compensation" techniques in order to achieve transmission speeds of 10 tera bits per second in broadband optical fiber systems by 2010. In order to achieve this goal, DNRI plans to develop a femto-second optical scope for high-sensitivity, real-time monitoring of optical signals, and a photonic crystal waveguide that corrects the dispersion of optical signals. For the medical & healthcare category, the company will develop a microchip for speedy SNP (single nucleotide polymorphism, pronounced "snip") typing. SNPs are minute variations in a person's DNA (deoxyribonucleic acid). By profiling SNPs and correlating these to a person's ability to metabolize a drug, personalized drug therapy might become a reality. At a press conference, a company researcher unveiled a schematic view of the prototype that consists of 2 layers of silicon substrate. The top layer has a sample reservoir and injection mechanism. The bottom layer has micro channels that facilitate the mixing of the sample with the reagent for hybridization and for detection. The proposed design will require 10 masks to fabricate the necessary 3-dimentional structures of silicon.

DNRI will also develop nanofabrication process technology called "nano-imprinting," which will be commonly used to produce optical, micro fluidic, electronic and other devices. Nano-imprinting is essentially a technique to fabricate highly precise molds in order to produce replicas cost-effectively. DNRI thinks the development of the nano-imprinting technique will increase device production throughput, broaden the selection of materials to use, and enable the fabrication of devices with high aspect ratios (the ratio of depth to the width of a structure) or laminated structures. DNRI will collaborate with manufacturing partners to market the nano-imprinting equipment and sell the devices that will be made using the equipment.

The MEMS industry is highly fragmented, with a large number of relatively small niche businesses. Therefore, it is important that companies in the future develop business models that can show the potential to go beyond growing these small markets and/or consolidating them into larger ones. DNRI executives believe they will need to be flexible in regard to working with different business models depending on which markets will be targeted. Their basic strategy, however, is to make most of the intellectual property rights they plan to develop. Therefore, collecting licensing royalties is expected to be one major revenue source. The company also plans to accrue revenue from helping to start and take stakes in joint venture businesses that put the intellectual property to use and commercialize it. At this moment, company executives do not have a definitive answer as to when DNRI will be profitable. The current plan is to start announcing device prototypes by the fourth quarter of 2003. DNRI then plans to disclose more detailed business plans at that time.