Concept of the reduction projection-type exposure system

In 1966, the Agency of Industrial Science and Technology (present-day National Institute of Advanced Industrial Science and Technology) started research into ultra-high-performance computers. These computers required ultra-high-performance transistors, and as the performance of these transistors progressed, the circuits configured with them needed to become more miniaturized. The Institute therefore developed an electron beam lithography system that used electron beams to draw minute circuit patterns on the wafer, and completed the system in 1967.

This system used an electron beam to draw a range of 2mm square, and when this range was drawn, the wafer was moved, and the process repeated. This process became known as "step and repeat" and the machines that performed it were called "steppers." Later reduction projection-type exposure systems became collectively referred to as steppers.

Nikon was then entrusted by the Institute to produce the stepper mechanism of the electron beam lithography system. The person who headed this assignment was Shoichiro Yoshida^*1 (currently a senior counselor of Nikon). Later, Yoshida had the idea that if the electron beam-related parts of this system were replaced by a mercury lamp and a projection lens, it would become a reduction projection-type exposure system. It would take time for an electron beam lithography system to draw a circuit pattern, resulting in low throughput. Yoshida surmised that the reduction projection-type exposure system could expose circuit patterns all at the same time and thereby achieve high throughput.

• *1Shoichiro Yoshida
  Joined Nikon (formerly Nippon Kogaku K.K.) in 1956. After working on the development of astronomical telescopes and ruling engines, he developed IC steppers and scanners. He became president in 1997, Chairman & CEO in 2001, and from 2011 a senior counselor for Nikon. In 1991, Yoshida received the Medal with Blue Ribbon, and in 2012 was awarded the Order of the Rising Sun, Gold and Silver Star.

Exposure system concept is turned down

In 1978, GCA Corporation became the first in the world to commercialize a reduction projection-type exposure system. Around 1972, five or six years earlier, Yoshida and Zenji Wakimoto^*2 made a presentation of the reduction projection-type exposure system concept that they had in mind to the president of the David Mann Company with the aim of forming a technology tie-up. The president turned the concept down, commenting that "this sort of high-precision equipment is not suited to mass production." Yoshida's concept also made no further progress since then. Later, however, the David Mann Company was acquired by GCA Corporation and became the core for the development of the reduction projection-type exposure system.

• *2Zenji Wakimoto (1924-1996)
  Joined Nikon (formerly Nippon Kogaku K.K.) in 1948, worked in the Optical Design Division and Camera Design Division, headed the Optical Division, and was appointed as a director of Nikon in 1977. In 1983, Wakimoto received the Medal with Purple Ribbon for his development of ultrahigh-resolution lenses.

Request from the VLSI Technology Research Association to produce a prototype

In March of 1976, under guidance of the Ministry of International Trade and Industry (currently the Ministry of Economy, Trade and Industry), the VLSI Technology Research Association was launched with the participation of the Agency of Industrial Science and Technology and five domestic computer manufacturers. The VLSI Technology Research Association aimed to develop an electron beam lithography system of even higher performance, with submicrometer^*3 resolution. Since the limit of optical resolution at the time was considered to be about 1 micrometer, the VLSI Technology Research Association felt a need for electron beam technology that could achieve higher (submicron) resolution.

Of course, it is natural to surmise that an electron beam lithography system would achieve lower throughput than a reduction projection-type exposure system. Researchers were concerned about this drawback, so they also looked into the possibilities of the reduction projection-type exposure system. They therefore assigned Nikon with the task of developing a reduction projection-type exposure system capable of producing circuit patterns of one tenth the size, and Canon was assigned the task of developing an exposure system capable of 1 to 1 projection.

• *3submicrometer:
  smaller than 1 micrometer (1/1000 millimeter)

Why Nikon was selected

Prototype No. 1

Nikon may have been chosen to develop a reduction projection-type exposure system, which is more difficult to produce than a 1 to 1 projection system, because it had gained experience with precision machining technology through its Ultra Micro-NIKKOR, a high-resolution projection lens, and its ruling engines; and with measurement control technology that utilizes photoelectric sensors. Nikon was also evaluated highly for its contribution to research into ultra-high-performance computers (mentioned above), which led to the formation of the VLSI Technology Research Association.

Two years later, in July of 1978, Prototype No. 1 was delivered to the VLSI Technology Research Association. Yoshida also headed the development of this system, bringing to fruition his idea of 10 years earlier.

Prototype No. 1 was operated at the VLSI Technology Research Association, leaving much data that was later used by semiconductor manufacturers and manufacturers of semiconductor fabrication equipment. Then in January of 1980, Nikon completed Prototype No. 2. It kept one system internally for evaluation purposes, and in April delivered another system to a semiconductor manufacturer.

Launch of the NSR-1010G

NSR-1010G reduction projection-type exposure system

Nikon's Prototype No. 1 and No. 2 were highly praised. The number of units to be manufactured, a time for their release, prices and other details were determined, and the NSR-1010G was publicly announced in February of 1980. Achieving 1-micrometer resolution and highly-precise alignment, the NSR-1010G became a promising system for the manufacture of VLSI circuits and attracted considerable attention. The first NSR-1010G was delivered in January of 1981.

However, the NSR-1010G's throughput was still low and many semiconductor manufacturers requested improvements. Nikon then changed the projection magnification from 1/10 to 1/5, and developed a projection lens with a wide exposure field. In December of 1981, Nikon released the NSR-1505G and NSR-2005G featuring this lens and improved throughput. The NSR-1505G, in particular, was used a lot by semiconductor manufacturers on their VLSI circuit production lines, paving the way for Japan's semiconductor industry to expand globally.

Nikon's second largest business after cameras