THIS YEAR MARKS THE 60TH ANNIVERSARY OF THE THIN-FILM TRANSISTOR (TFT). The TFT is so closely associated with the modern active-matrix display that it may surprise some readers to learn that more than 20 years after the fabrication of the first TFT, no high-resolution LCDs used them. Let's begin the story there. In 1985, Toshiba introduced the T1100—the world's first mass-market laptop computer (Fig. 1), which significantly was compatible with the IBM desktop PC. At a price of US $1,899 ($4,514 in 2019 dollars), Toshiba was delighted to sell 10,000 the first year. The Toshiba T1100, introduced in 1985, was the first laptop PC manufactured in high volume. It used a passive-matrix, twisted-nematic LCD, whose limitations can be seen in the photo. Source: Toshiba. Cramming all of the electronic functionality of an IBM PC into a laptop package was challenging, but there was one component without which the T1100 could not have been made: a monochrome twisted-nematic (TN) LCD capable of displaying 80 × 25 alphanumeric characters and CGA (up to 640 × 200) graphics on a screen measuring 9 × 4 inches. The display had a just-adequate contrast ratio of 4:1 and a vertical viewing angle of +40°/–15°. The display could be tilted to bring the readable zone into the user's field of view. The image quality of the display was poor by modern standards, but it made the T1100 a functional product, and it allowed the computer to run for eight hours on a battery charge. In early 1987, Toshiba upgraded the display to a supertwisted nematic (STN) unit in the T1100 PLUS. The display made the T1100 possible, and this laptop PC was much more portable than the Compaq Portable and a few similar “luggables” that integrated the full-size components of an IBM PC with a cathode-ray tube (CRT) monitor in a suitcase-like container. The T1100 solved the portability problem, but the image quality begged to be improved. Color cathode-ray televisions of the time gave people an idea of what they would like to experience in a computer monitor or laptop screen. But the IBM 5153 Personal Computer Color Display that was introduced in 1983 was nothing like today's full-color monitors and notebook screens. The monitor supported CGA: 16 colors in its 160 × 100 low-resolution mode and black and white (or black and one other color) in its 640 × 200 mode. The IBM CGA adaptor that drove the 5153 also had a composite video output so it could be used with TV sets, but TV picture and computer data tubes were different animals. Picture tubes had higher luminance but much lower resolution, and composite video did not completely separate the chrominance and luminance parts of the signal, which resulted in poorer video quality. However, the IBM 5153 driven with a CGA video adaptor had much better contrast and viewing angles than TN LCDs. The high-resolution TN displays—such as the one used in the T1100—used passive addressing, in which the switching voltage on each pixel is produced by the time-dependent difference between voltages on the row and column electrodes. Because every row has to be scanned during the period of one frame, each pixel experiences its designated voltage for a time that is inversely proportional to the number of rows. As the duty cycle of the voltage decreases, the liquid crystal material at each pixel has less time to change its orientation, and its maximum distortional response is less (i.e., the greater the number of rows, the less the maximum contrast ratio). In addition, the viewing angle (not good to start with) deteriorated further with greater resolution. With some of these displays, the image would turn negative when viewed from even moderate angles. Clever designers ameliorated some of these limitations. With dual scanning, the top half of the display was scanned from the top while the bottom half was scanned from the bottom, allowing the number of lines to be doubled for any target contrast ratio. But the difference between the root mean square (rms) voltage for ON and OFF pixels still was quite low. “For example, in a dual-scan, notebook computer video graphics array (VGA) display, which has 240 multiplexed lines, the rms voltage at a selected ON pixel receives only about 6.7% more voltage than a nonselected OFF pixel.”1 The apparently unbreakable trade-off between contrast and resolution became known as “The Iron Law.” A partial solution was the development of the supertwisted LCD display, in which a chiral additive caused the LC material to assume an unactivated twist of greater than the 90 degrees of TN displays. An angle of 240 or 270 degrees often was designed, which produced a much steeper electrodistortional response (producing greater contrast) for the limited voltage available. Although STN was a great improvement over TN, which is why Toshiba adopted it for its upgraded T1100 PLUS, the technology had its own limitations. Subsequent STN developments constitute a fascinating story of engineering insight and ingenuity. For those readers who care to pursue it, the article by Scheffer and Nehring1 is an excellent place to start. To reach a higher resolution with good video quality, one would need to break The Iron Law. That could be achieved by placing a switch at each pixel location, replacing the passive matrix with an active matrix. The switch would apply the required voltage to a capacitor, which would maintain the pixel at the appropriate voltage for the entire frame time, regardless (almost) of the number of rows. Bernard J. Lechner of RCA Labs originated the idea of an active-matrix display in 1968 and demonstrated the concept later that year with an 18 × 2 dynamic scattering (DS) LCD. The proof of concept used discrete metal-oxide-semiconductor field-effect transistors (MOSFETs). (A team at RCA Labs headed by George A. Heilmeier, including Joseph A. Castellano, invented the DS LCD in 1966.) However, a high-resolution LCD with thousands (later millions) of pixels could not be made with discrete transistors. A technology was needed that would permit the mass production of small, thin transistors behind the pixel array. Such transistors—TFTs—already had been invented (Fig. 2). In 1957, John Wallmark of RCA filed a patent for a thin-film MOSFET in which germanium monoxide was used as the gate dielectric. Paul K. Weimer (Fig. 3), also of RCA, implemented the concept and made the world's first TFTs in 1962, which were fabricated with thin films of cadmium sulfide. In 1966, T. Peter Brody and H.E. Kunig (Westinghouse Research Labs) fabricated MOS TFTs using indium arsenide. The thin-film transistor (TFT) embodied the performance of a bipolar transistor in a flat structure that made the economical fabrication of transistor arrays possible. Source: Joseph A. Castellano Paul Weimer made the world's first TFT in 1962. He also invented the first color video camera. Source: National Academy of Sciences/National Academy of Engineering In 1973, Brody (Fig. 4) and colleagues developed a cadmium selenide (CdSe) TFT. In 1974, Brody and Fang-Chen Luo demonstrated the world's first flat active-matrix LCD (AMLCD), which used CdSe, and Brody coined the term active matrix in 1975. But CdSe devices never were produced in volume because of complications in controlling material properties, device reliability over large areas, and cadmium toxicity. Nonetheless, Brody continued to argue the case for CdSe, in print and in person, over the remainder of his long life. T. Peter Brody, along with Fang-Chen Luo, demonstrated the world's first flat active-matrix LCD in 1974. In 1975, Brody coined the term active matrix. Source: Sarah Brody Webb (CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=52490762). A breakthrough came with the development of the hydrogenated amorphous silicon (a-Si) TFT with silicon nitride gate dielectric by P.G. Le Comber and colleagues (University of Dundee) in 1979. The material soon was recognized as being more appropriate for large-area AMLCDs than pre-existing materials, which led to its commercial development for AMLCD panels in Japan. In May 1983, Shinji Morozumi of Seiko-Epson announced a transmissive, full-color TFT-LCD at the Society for Information Display conference. The display measured 2.13 inches square and had 57,600 pixels. It was the first practical TFT-LCD television display. Castellano recalls that many people in the audience were amazed by Morozumi's presentation because the display appeared to have few, if any, defects. In August 1984, a slightly smaller version of the display appeared in Epson's ET-10 color pocket TV, now regarded as the world's first commercial AMLCD product (Fig. 5). The pocket TV category quickly proliferated. Epson's ET-10 color pocket TV, introduced in August 1984, is regarded as the world's first commercial AMLCD product. The pocket TV category quickly proliferated. Source: Epson Only three years after Toshiba introduced the T1100 with a monochrome TN display in 1985, Sharp produced a 10.4-inch TFT-LCD panel for notebook PCs, and in 1990 opened the first-generation (Gen 1) TFT-LCD manufacturing facility. In 1988, Sharp demonstrated a 14-inch full-color LCD, which convinced the electronics industry that LCDs eventually would replace CRT as the standard television display technology. In 1992, a joint venture between Toshiba and IBM Japan introduced a 12.1-inch color Super Video Graphics Array (SVGA) panel for IBM's first commercial color laptop, the PS/2 CL57 SX (Fig. 6). It was an impressive display, but the computer was not well received, probably because of its $5,595 price tag ($11,060 in 2020 dollars). In 1992, Toshiba and IBM Japan introduced this 12.1-inch color SVGA panel for IBM's first commercial color laptop, the PS/2 CL57 SX. Source: Laptop Pics. On its 60th anniversary, the TFT a-Si remains the primary material for TFT arrays because of its low cost and ease of fabrication, but a-Si has a low carrier mobility and is unstable when used with current-driven emitters, such as OLEDs. Careful laser annealing of a-Si produces low-temperature polysilicon (LTPS), which has much higher carrier mobility that permits higher refresh rates and lower power consumption. The TFTs also can be much smaller, which makes them suitable for high-resolution displays, and they are stable when used with AMOLEDs. But the LTPS fabrication process is not (yet) practical for large substrates and large displays. Considerable research indicated that transparent metal oxides were a promising material for TFTs. They originally were subject to various modes of instability, but years of development finally produced stable materials. Currently, amorphous indium gallium zinc oxide (IGZO) is the preferred oxide and is used widely for large AMOLED and LCD displays. The current interest in flexible displays has fueled an interest in organic TFTs. OTFTs initially had carrier mobilities considerably worse than that of a-Si TFTs, but after years of development, their mobilities are now superior. The physics, chemistry, and fabrication of TFTs have rarely attracted attention outside the fraternity of their practitioners, but their effects are highly visible. Users are moved (and motivated to buy) by the performance of display-centric products, which depends upon the characteristics of their TFTs. It's been 60 years, and we're not done yet. Ken Werner is the principal of Nutmeg Consultants and specializes in display manufacturing, technology, and applications. He is the 2017 recipient of the Society for Information Display's Lewis and Beatrice Winner Award. He can be reached at kwerner@nutmegconsultants.com. Joseph A. Castellano founded Stanford Resources in 1976, where he worked until retirement. He joined RCA Laboratories in 1965, where he performed pioneering research in photochemistry and liquid crystal materials. Castellano received a PhD in chemistry from New York University's Tandon School of Engineering (formerly the Polytechnic Institute of Brooklyn). In 2000, he received a SID Special Recognition Award. He can be reached at drjcast@aol.com.