A single ink stain opened the door for 2019 IEEE Medal of Honor recipient Kurt? Peterson opened the door to a lifetime of research into microdevices.
In 1975, Kurt Petersen was the inventor of the microdevice. Kurt Petersen was a bright young researcher with a Ph.D. in electrical engineering from the Massachusetts Institute of Technology working at the IBM Almaden Research Center in California. He was part of the center's optics research group. However, he was bored from time to time, and one day he was strolling through the huge complex and then noticed a large black stain on the linoleum tiles of an ordinary corridor. It was this stain that changed his life and his entire profession.
In an effort to find the source of the stain (and he had nothing better to do), Peterson went into the nearest lab. Eventually he realized that the stain had formed from spilled ink. This is a lab that develops nozzles for inkjet printers, a process that involves punching holes in silicon.
Punching holes in silicon? Peterson had never heard of it, but he remembered an ad he had seen earlier about silicon-based microaccelerators. Suddenly, a larger picture came to him: people were actually making miniature mechanical parts, with various components just a few micrometers long, all made of silicon. Today, we call such devices microelectromechanical systems (MEMS). Peterson wanted to make MEMS, too.
So he started a whole new career path - specializing in MEMS technology, including the device now used to scan all mail in the United States for anthrax, and founded MEMS Enterprises. It was for his contributions in this area that Peterson received the IEEE Medal of Honor in 2019.
Shortly after discovering the ink stain, Peterson began reading everything he could find on the use of silicon to make micromechanical devices, including various journals and magazines such as IEEE Transactions on Electron Devices, Applied Applied Physics Letters, and the Journal of the Electrochemical Society. At the time, there was no specific name for the device and only a few MEMS products on the market. He found that "a lot of people around the world have made different mechanical devices out of silicon, but there is no community. Most of the people working on these kinds of devices don't know each other."
And then Peterson set out to build his first device. Looking at those inkjet printer nozzles under the microscope, he said, "If there was a defect, I could see it at a glance. There were tiny, separate and very fine silica columns under the microscope. And I thought, these tiny mechanical structures might be able to move around. They might be able to steer light, and I could make a dimmer." His development process was similar to how MEMS are fabricated today, starting with a layer of silicon dioxide on top of an epitaxial silicon sacrificial layer, which is then etched away. In the end, only the silicon dioxide cantilever remains, with a thin metal layer on top.
He spent three months building several microregulators, each about 100 microns long and 0.5 microns thick. He took the regulators to a lab equipped with an IBM scanning electron microscope, where a technician helped him set up the wires, and then he powered up the devices and watched them operate.
"She was all mesmerized." Peterson recalls, "She said she had never seen a device that ran under a microscope."
Peterson then spent the next five years building as many different kinds of micromechanical devices as he could from silicon, including gas pedals and electronic switches. He left the optics research group for a specially customized lab that could accommodate only him and an intern.
Based on a deep dive into the literature and the work he had done, Peterson wrote an internal report on the emerging technologies. "A lot of the mechanics may be valuable to IBM." Things like read/write heads for optical and mechanical disk drives and more complex nozzles for inkjet printers, for example, weren't of interest to IBM, he said.
Peterson was disappointed, but he also realized that such devices were not part of IBM's key business. So he revised his report to remove IBM's proprietary information, and submitted it to the IEEE Transactions, filling 50 pages of the entire Transactions. The article, titled "Silicon as a Mechanical Material," was the cover article in May 1982, establishing MEMS as a separate branch of technology.
The paper was comprehensive in its coverage of the mechanical properties of materials for integrated circuits, as well as the various ways in which such materials can be etched into appropriate shapes and structures. "The article extrapolates on things that may come in the future, such as deep reactive ion etching (DRIE), a technology that has revolutionized the field." He said, "Even today, I have many people telling me that it was that article that got them interested in MEMS."
"We all read that article when we were in graduate school." Greg Kovacs, who is now chief technology officer at Stanford's Institute for International Studies? Greg Kovacs said of the institute, which is located in Menlo Park, Calif. "He has played a huge role in the MEMS field. The work he accomplished was more important than pioneering the field, and he pushed it forward. To me, he is a superhero."
Once the IEEE Transactions paper was published, Peterson was invited to speak at conferences around the world, and researchers came to Almaden in droves to get a glimpse of the author. "People doing all kinds of crazy research get in touch with me in one way or another, like the researchers on microfluidic cryogenic coolers." He said. He seems to have become a MEMS technology gatekeeper overnight.
The field had been growing steadily through the 1980s. When Peterson's paper was published, there were about 30 or 40 people around the world working on the technology. By 1990, he estimated that there were about 600 people working on the technology. Pressure sensors for disposable blood pressure monitors and new fuel control carburetors appeared on the market. MEMS-based accelerometers also began to be used in the aerospace industry. The first micro-mechanical inkjet printer printheads enter mass production. There were a lot of startups that were eager to evolve with the technology. Peterson said the field was officially named at the 1987 NSF workshop.
Not surprisingly, Peterson was contacted by several companies. He ended up accepting the invitation, and in 1982, with Jim Knutty, he was able to work on the field. Jim Knutti to co-found Transensory Devices to develop and manufacture MEMS devices.
He recalls feeling "nervous" about giving up the stability of a corporate research job. He had two young sons, so financial security was important. About $1 million of the startup money ended up coming from out-of-state oil investors, not Silicon Valley ones. "There were some startups in Silicon Valley at the time, but nothing like what we have today. Fundraising was a tough thing to do back then." He said.
Their team later moved to a 280-square-meter lab in Fremont, California, and built some of their own equipment, including wafer-bonding equipment to encapsulate and protect silicon wafers. They contracted with large companies to produce samples for them, including the kind of dimmers Peterson had made at IBM. At the same time, they began developing their own MEMS devices.
"We were demonstrating a lot of devices," Peterson said, "but none of them went into production." At one point, a tire-pressure sensor for the trucking industry was nearly successful, but the executive they were working with died. Peterson believes that it was because both he and Kneutty lacked manufacturing experience that their research was not able to be commercialized.
Contract manufacturing kept Transensory running smoothly, but Peterson still wanted to bring his MEMS device to market. He thought it was time to start a second company.
In 1985, Peterson teamed up with Januszky Brzeszek, the founder of Transensory, to create a new company. In 1985, Peterson teamed up with Janusz Bryzek and Joseph Mallon to develop a MEMS device. Joseph Mallon to create NovaSensor, with $5 million in startup capital from oilfield services giant Schlumberger. Brizek had previously co-founded two companies developing MEMS pressure sensors. "Jankowski and his partners have production and manufacturing experience" that Transensory lacked, Peterson said.
NovaSensor was founded and began manufacturing three types of pressure sensors: one for the aerospace industry, another for the oil industry, and a high-temperature pressure sensor that was not targeted to a specific market. The last proved to be the most successful, with the pressure sensor even being used in the tires of the space shuttle. "We found a way to isolate the resistor from the substrate using a MEMS process. We bonded single-crystal silicon wafers to a silicon oxide wafer with a pressure sensor diaphragm, then etched away most of the upper wafer, leaving only the resistor." Peterson said. He believes this sensor was the first silicon-on-insulator device, and this device has been in common use ever since.
In 1991, Lucas Industries acquired NovaSensor, making Peterson a "MEMS Millionaire," and NovaSensor's product line is now sold by Amphenol.
Peterson's stake continued to grow over the next few years. During that time, he focused on fusion bonding, a process that involves etching two different patterns of wafers and then joining them together. This process allows for the creation of very complex devices, such as gyroscopes. His business card has always had a picture of one of the first devices made using this process.
By the time Peterson left NovaSensor in 1995, MEMS pressure sensors were used in a wide variety of systems, including diving equipment and HVAC control systems, and MEMS accelerometers were just beginning to be used in collision-awareness systems in automotive airbags.
Petersen left NovaSensor without any arrangements.
Petersen left NovaSensor without any arrangements. A researcher at Livermore National Laboratory, Allen Northrup, has been working on the project for several years. Allen Northrup, a researcher at Lawrence Livermore National Laboratory, suggested to him that a MEMS device could greatly speed up the polymerase chain reaction (PCR), a relatively new method of replicating DNA sequences.
Bill McMillan, a friend of Peterson's wife who works in the biotech field, said that the MEMS device could greatly speed up PCR, a relatively new method of copying DNA sequences. Bill McMillan, a friend of Peterson's wife who works in biotechnology, recognized the promise of PCR. Peterson then began working on a plan to reduce the size and cost of PCR machinery, with the goal of creating handheld devices that doctors could use in their offices.
He and McMillan had lunch at the Magnolia Cafe*** in Palo Alto. "I gave him a general overview of my idea and he started sketching the business plan on a paper placemat." Peterson said. He still has that placemat today.
Peterson's 1982 paper hinted at the possibilities of deep reactive ion etching, a technique that can carve deeper holes and grooves into silicon than traditional chip production processes. He began applying deep reactive ion etching to microfluidic chips, delivering tiny amounts of liquid into precise channels.
"We had an idea that we could use MEMS technology and microfluidics to rapidly heat and cool samples to make a small but responsive PCR device that doctors could use to make diagnoses in the office." Peterson said.
To commercialize the technology, Peterson co-founded Cepheid in 1996 and received a grant from Lawrence? Livermore National Laboratory to license the underlying technology. By 1997, the company had raised $3.2 million in funding from the U.S. Department of Defense, which wanted the company to develop a bioweapons detector.The first device Cepheid developed was called the Smart Cycler, which used a MEMS structure to enable rapid heating and cooling of a few microliters of liquid while using a fluorescent sensor to monitor the progress of the reaction. It's not a handheld device, but that's not a problem. More importantly, it automates the PCR process.
Cepheid's second product, the GeneXpert, is designed to further simplify PCR by automating the extraction of DNA from biological samples and then adding the reagents needed for testing.
The company went public in 2000, just as the tech bubble was bursting. Before the market shrank, "we were one of the last companies to have a successful IPO." Peterson said.
Through the public stock offering, the company secured enough capital for the team to put the Smart Cycler into production, and by the end of the summer of 2001, the company had shipped 80 units. After the first prototype was produced in December 2001, development of GeneXpert continued to progress incrementally.
Then came the anthrax terrorist attacks in the United States. In late September and October 2001, letters carrying anthrax spores were mailed to the U.S. news media and members of the U.S. Senate, ultimately infecting more than 20 people and killing five.
By then, Cepheid had already established that its technology could quickly detect anthrax bacteria and became an overnight sensation. "We worked with Sanjay? Dr. Sanjay Gupta to conduct a live PCR trial via the Good Morning America show and CNN." Peterson recalled.
Worried about another letter-carrying bioattack in the future, the U.S. Postal Service invited all the companies that had mastered biosensor technology to demonstrate their products, and Cepheid's device passed the test in December 2001. "It was running perfectly. It worked perfectly," Peterson said.
After a few months of additional testing, the company partnered with Northrop Grumman to develop the PCR biosensor, which easily interfaces with mail sorters. The product was introduced to the market in 2003, and today, all mail in the United States is still screened for anthrax through Cepheid machines, Peterson said. The company's systems are now used for medical diagnostics related to strep, norovirus, influenza, chlamydia and more. The company sells more than 20 tests that are approved by the U.S. Food and Drug Administration and applicable to Cepheid machines.
By 2003, Peterson was ready for a new chapter in his career. This time, he wanted to develop silicon resonators, devices that produce a constant frequency that can be used for precise timing. "I built some of the first MEMS resonators while at IBM, but they weren't ideal. They were not comparable to quartz crystal oscillators." He said.
Tom? Tom Kenny, Markus Lutz, and Jennifer K. Kennedy have been working on MEMS resonators for some time. Markus Lutz, and Aaron Partridge. Aaron Partridge, three researchers, came up with a better solution. "They used monocrystalline silicon to make the resonator, which is the most perfect material in the world." Peterson said, "Polycrystalline materials under stress produce tiny shifts at grain boundaries. Over time, even if only one or two atoms are displaced, this can lead to changes in mechanical properties." Whereas monocrystalline silicon does not change over time, its resonant frequency changes with temperature, so the difficulty lies in how to address its temperature dependence.
Peterson, Kenny, Lutz, Partridge, and Joe? Brown (Joe Brown, Peterson's colleague at IBM, with whom he ****ed at both Transensory and NovaSensor) were once again dining at the Magnolia Cafe ****, once again drafting a business plan on a paper placemat. Robert? Bosch AG owns some of the core intellectual property, so in addition to attracting investors, Peterson had to convince Bosch executives in Germany to license the technology.
"In Stuttgart, I had a big meeting with their board." He said, "I told them, 'This is what I do. I started the company, and our company's equipment is responsible for the anthrax screening of all the letters in the United States.' Their board of directors not only agreed to license the technology, they made a significant investment in our company."
The new company, SiTime, was founded in December 2004 with the goal of changing the multibillion-dollar material used in the timing industry from quartz to silicon. The company's first resonators were delivered in 2007. Today, the company's MEMS oscillators are used in a wide range of timing systems for mobile devices and other electronic instruments.
In 2008, when SiTime was doing well, McMillan, one of Peterson's partners at Cepheid, approached him with another entrepreneurial idea: to develop an implantable continuous glucose monitor. "People had been working on it for 30 years, but no one had been successful," Peterson says. Peterson says. Once the sensor is implanted inside the body, "the body uses collagen to isolate it, ultimately preventing the blood sugar from touching the sensor," he explains. He explained.
So McMillan, along with Duke researcher Natalie Wisniski, has been working on a project to develop a blood glucose sensor. Natalie Wisniewski and came up with a solution: using structured hydrogels to avoid foreign body reactions and fluorescent readouts to measure blood glucose concentrations. Peterson used his previous knowledge of optics to help develop the product and spent a year at the start-up company Profusa. That company now has about 30 employees and $100 million in funding.
Peterson says running the company will be his last full-time job. "I just didn't want to keep dealing with the day-to-day business of the company. I started angel investing, which is much more fun."
He also couldn't resist the temptation to build another team. Two Berkeley students had developed technology related to MEMS resonators, but had struggled to commercialize the technology. Peterson and K.G. Ganapathi joined the students' company, which was renamed Verreon, and Peterson became the company's chief technology officer, helping to coordinate the company's sales to Qualcomm in 2010.
This is Peterson's third stint as CTO or in a similar role. Of all his startups, he has only served as CEO at SiTime. "While at NovaSensor, the other two guys wanted to be chairman." The company's marketing consultant, Roger? Roger Grace said, "Kurt didn't care, and he took on the role of chief technology officer. He's not an egotist."
"In the MEMS world, people praise Kurt for being very kind, considerate, and helpful." Grace said, "There are a lot of smart people out there, but he's unique in that he's humble. You feel at ease with him."
Ganapati agrees, "It's rare to find someone as successful and well-liked as Kurt."
For now, Peterson is back in the big business of angel investing, targeting MEMS companies, medical devices, and the biotech sector. He says he's invested in about 70 companies, nearly half of which have been successful, with a 350 percent return on investment, a remarkable record given that a recent study showed that, in general, long-term angel investors with broader investment horizons have a 250 percent return on investment.
"It's as if he has a mysterious power to detect promising products. It takes 3 or 15 years for a product to be successful, but he has a keen sense of that." Ganapathi said.
In 2012, Peterson joined the Silicon Valley Angel Investment Gang, an invitation-only organization of about 200 investors who meet regularly to learn and share information. Now, he heads the organization's hardware division. He also sits on the boards of two companies and serves as an advisor to dozens of others. He meets with several people a day who come to him for advice and makes phone calls to companies in Canada and the eastern seaboard of the United States.
Peterson is 71 years old, but he has no intention of retiring. "Entrepreneurs are dynamic, energized and ambitious, and it's a pleasure to deal with them." He said.
Semiconductor Industry Watch
The #1 Vertical Media for Semiconductors
Huawei｜Telespac｜Jiangbei New District｜Samsung｜IC｜AI｜Bosch｜ARM
Reply to Contribute to the newsletter to find out "How to become a member of the Semiconductor Industry Watch". "
Reply to Search to find other articles of interest!