Issue link: https://iconnect007.uberflip.com/i/1085192
FEBRUARY 2019 I DESIGN007 MAGAZINE 21 called the "Dream Team." We combined a vet- eran group of experienced talent with a rook- ie group of individuals who did not know the technology barriers of the past. The team be- lieved that we were going to launch a multi- billion-dollar business based on novel prod- ucts realized by a portfolio of innovations— designs materials, processes, manufacturing equipment. Goldman: When you were designing some of those early flexible or rigid-flex boards, there must have been a lot of failures and having to do it over again until you caught on to it such as the direction of the copper grain at flex points and things like that. Am I right? Gamota: You're right. It was the days of Mac- Gyver because people were fabricating a bot- tom-gate transistor using any materials you had available. We had people using a Juicy Fruit gum wrapper, paperclip, and semicon- ductor ink to make a transistor. They looked at everything in an open transistor and IC design world where you simply required a conductive material for the gate, dielectric material for the gate insulator, another conductive material for the source and drain, and a material that had semiconductor-like properties. When we would visit printing services com- panies—because ultimately, we were going to use printing equipment such as a screen printer, gravure, or flexography—we would go to their shelf of inks that they were using for graphic arts printing, and we would look at the ink chemical compositions. We learned that in graphic arts printing (e.g., a magazine) several of the inks used have electrical prop- erties. Some inks are great conductors while others are great insulators. Except for not hav- ing a semiconductor ink, the printers had all the necessary materials to print transistors as well as the process know-how to align and print very fine features necessary for electron- ics. They knew how to ensure that transistor structure layers were pinhole-free and how to achieve sub-50 micron registration from print- ed layer to printed layer. It seemed like everything was going to align perfectly, and the only thing that was a po- tential concern for us was the semiconduc- tor material. The advancements in semicon- ductor material electrical mobility were a lit- tle bit tempered and the materials weren't as high performing once you took them out of a well-controlled environment—a chemistry lab. Once you moved the material into a manufac- turing or NPI environment, you lost a couple orders of magnitude in mobility performance. You still have these issues today when transi- tioning technologies from lab to fab. Also, we lack the ability to consistently print 1,000 tran- sistors and ensure that all 1,000 provide the same electrical performance. That's where the deficiency of printed tran- sistors lies—robust high mobility semiconduc- tor inks. Printed transistors would have a very difficult time replacing silicon based transis- tors. Once we realized that printing 1,000 tran- sistors that all worked within a certain opera- tion window would be very difficult and we observed the cost coming down quickly for sil- icon-based ICs, an all-printed transistor-based circuit was no longer an option. This was the start of flexible hybrid electronics (FHE). Today you see flexible hybrid electronics (printed electronic devices combined with sili- con-based devices assembled on a flexible sub- strate)—enabling the design and manufacture