A Chemical Company in Tokushima

The tension between the powerful consensus and the faint signals from Tokushima and Nagoya defined the moment. But consensus takes time to form, and signals take time to send. In 1979, the moment had not yet arrived. There was no agreement to defy and no contradiction to offer. There was only a chemical company on Shikoku, the smallest of Japan’s four main islands, and a young engineer reporting for his first day of work.

Shuji Nakamura walked through the gates of Nichia Chemical Industries in Anan City, Tokushima Prefecture, carrying a master’s degree in electronic engineering from the local university. He was twenty-five years old. Nakamura had grown up on Shikoku, the son of a farmer in a rural prefecture known for its Naruto whirlpools and its indigo dye. Tokushima was far from Tokyo, farther from the centers of Japan’s electronics industry. Nakamura had interviewed with Mitsubishi and Sony, the companies that mattered. They had not hired him.

Nichia hired him. The company made chemicals, not research scientists.

Nobuo Ogawa had founded Nichia in 1956, naming it after the Japanese word for “sun” and the Chinese word for “center.” The name reflected his ambition. Ogawa was a pharmacist by training, a practical man who had built his company on the production of calcium phosphate, a raw material for fertilizer. Nichia’s business was commodity chemicals: materials sold by the ton, with margins measured in fractions of a yen. The company produced phosphors for fluorescent lamps, compounds that glowed when excited by ultraviolet light. The business was stable, reliable, and limited.

Ogawa had built his factory in Tokushima because he was from Tokushima. The decision was personal, not strategic. He could have located closer to Osaka, closer to suppliers and customers and transportation networks. He chose instead to remain in the rural prefecture where his family lived. The decision defined the company. Nichia was provincial by design, a creature of its place, managed by men who had grown up together and sent their children to the same schools. The boardroom sat in Anan City, population forty thousand, accessible by a two-hour train ride from Osaka followed by a bus connection.

The location attracted a specific kind of employee. Engineers who wanted to work for Mitsubishi or Sony did not apply to Nichia. Engineers who wanted to live in Tokyo or Osaka did not apply to Nichia. The company drew from the local university, the University of Tokushima, a solid regional school without the prestige of Tokyo University or Kyoto University. Nakamura was typical of the hires. He was local, he was available, and he had a degree in a relevant field.

Nakamura’s first assignment was production line maintenance, not research. He was given a section of the phosphor manufacturing process and told to keep it running. The work was practical, immediate, and repetitive. Machinery, chemistry, failure points—these became his daily education. Temperature controls drifted, batch compositions varied, and the results taught him what worked and what did not.

The first blue gallium nitride LEDs were developed in 1971–1973, but they were feeble. The researchers who made them moved on to other materials. The field had concluded that gallium nitride could not be made useful. Effective doping eluded every attempt. Quality crystals refused to grow. The material produced light, but barely. The companies that had tried it—RCA in America, where J.I. Pankove and co-workers put in considerable effort but did not make a marketable GaN LED in the 1960s—abandoned it for zinc selenide, a material that seemed more promising.

Nakamura knew none of this. He was maintaining production lines in Tokushima. But he was bored, and he was ambitious, and he was not suited to routine work. Within a year, he was asking for permission to do research.

Ogawa said yes.

The permission was remarkable, though neither man recognized it at the time. In a large electronics company, a production engineer’s request to move into fundamental research would have died in committee. The engineer would have been told to wait, to apply through channels, to demonstrate qualifications he did not have. He would have been reminded that research positions were for researchers, not for maintenance staff. He would have been told no.

Ogawa gave a different answer. Nichia had no committees, no research hierarchy, and no established researchers to defend their territory. The president believed that engineers should solve problems, and if Nakamura’s problem was boredom, then research was a solution. The company was small enough that the president could make decisions without consultation.

Nakamura was given a small laboratory and a budget. He was not given a research plan or a mandate to produce commercial products. He was simply given permission to work.

The freedom was not total. Nichia was a business, not a university. Nakamura’s research was expected to produce something of value eventually. But the timeline was loose, the oversight was minimal, and the expectations were undefined. Ogawa had created a space for patient capital without using the phrase. He had given an engineer money and time, and he had asked for nothing immediate in return.

Nakamura’s first research projects were modest. He worked on phosphors, extending the company’s existing expertise. He developed new formulations, improved manufacturing processes, filed small patents. The work was useful but not revolutionary. A chemical company could expect exactly this kind of incremental improvement from a small research program.

But Nakamura wanted more. He had applied to Mitsubishi and Sony because he wanted to work on cutting-edge electronics. He had ended up at a phosphor company in rural Shikoku, but his ambitions had not changed. He read the technical literature. He followed the field. He knew that the major electronics companies were racing to develop blue LEDs, and he knew that they were failing.

The blue LED was the missing piece. Red LEDs had been available since the 1960s. Green LEDs followed. But blue remained elusive. Without blue, there could be no white light from LEDs, no full-color displays, no solid-state lighting revolution. Every major electronics company wanted it. None could make it work.

The consensus was clear. Zinc selenide was the material of the future. Gallium nitride had been tried and found wanting. The best minds in the field had attempted it and failed. The papers had been published, the conclusions had been drawn, and the research programs had moved on.

Nakamura read the papers. He reached a different conclusion.

The judgment was characteristic. Nakamura was not a member of the field. He had not attended the conferences, participated in the discussions, absorbed the consensus. He was reading the literature from outside, without the social pressure to agree. He saw that gallium nitride had produced blue light, however feeble. He saw that zinc selenide had produced nothing useful despite decades of effort. He wondered why everyone had given up on the material that worked, however poorly, to pursue the material that did not work at all.

In 1988, Nakamura made a request. He asked Ogawa for permission to pursue blue LED research using gallium nitride.

Ogawa was seventy-six years old. He had built Nichia from nothing, had made it profitable, had provided stable employment in a prefecture that needed it. He was not a gambler. But he was also not a manager who trusted consensus over his own judgment. A career spent building a company from scratch had taught him to make decisions that others might not have made. He had located his factory in Tokushima when logic suggested Osaka. He had hired local engineers when he could have recruited from Tokyo. He had given a bored maintenance worker a research budget because the man asked.

Now his engineer wanted to pursue a material that the entire electronics industry had abandoned. The request made no business sense. Nichia was a chemical company, not an electronics company. It had no expertise in semiconductors, no connections to the device market, no infrastructure for producing LEDs. If Nakamura succeeded—and the consensus said he would not—Nichia would have to build capabilities it did not possess. If he failed, the company would have spent money on nothing.

Ogawa approved the request. He also approved a different request: Nakamura wanted to spend a year at the University of Florida, studying metalorganic chemical vapor deposition, the technique he would need to grow gallium nitride crystals.

The decisions were extraordinary. A large electronics company would have required a formal proposal, a budget projection, a timeline with milestones, a committee review. Ogawa gave his approval in conversation. There was no proposal, no committee, no formal process. The president of a small chemical company in rural Japan decided to fund fundamental research into a material that the field had declared impossible, and he did it because one of his engineers asked.

The year in Florida transformed Nakamura’s technical capability. He learned the mechanics of crystal growth, the parameters that mattered, the equipment he would need. He returned to Tokushima in 1989 with knowledge that Nichia did not possess and could not have acquired any other way. The company had sent an engineer abroad to study, had paid his salary and his expenses, and had asked for nothing immediate in return.

But Ogawa was aging. In 1989, he ceded the presidency to his son-in-law, Eiji Ogawa. Nobuo Ogawa remained as chairman, but the company’s daily management passed to a new generation.

The transition brought a change in culture. Eiji Ogawa was more cautious than his predecessor. He had not built the company; he had inherited it. He felt the responsibility of preservation more keenly than the impulse of risk. He looked at Nakamura’s gallium nitride project and saw money disappearing into a material that the entire industry had rejected.

Eiji Ogawa ordered Nakamura to suspend work on gallium nitride. The project was consuming too much time and money. It distracted the engineer from work that might produce revenue. The decision was not unreasonable. From a business perspective, Nakamura’s research was a speculative bet with no clear timeline and no guaranteed return. Nichia was a profitable company, but it was not a rich one. The money spent on gallium nitride was money not spent on improving the phosphor business, on expanding production, on reliable returns.

Nakamura argued. He was not a man who accepted orders easily, particularly orders that stopped his work. Years of preparation had gone into this research. He had won the right to pursue it, had studied the technique, had designed the equipment in his mind. Now a new president wanted to take it away.

Nobuo Ogawa intervened. The founder, now chairman, supported his engineer. He had approved the research, and he had not changed his mind. The decision stood. Nakamura could continue.

The intervention was brief, but it defined the company’s direction for the next decade. The founder’s authority overrode the president’s caution. The engineer’s ambition overrode the business logic. Nichia would pursue gallium nitride despite the consensus, despite the cost, despite the new president’s doubts.

The decision revealed the structure that made Nichia different. In a large company, a president’s decision would have been final. The organization would have enforced it. But Nichia was a family firm, a provincial enterprise, a place where personal relationships mattered more than organizational charts. Nobuo Ogawa had hired Nakamura, had approved his research, had sent him to Florida. He was not going to let his son-in-law cancel the project on a calculation of cost.

The structure was accidental. Nichia had not been designed to foster impossible research. It had evolved into an organization that could approve speculative bets because it lacked the bureaucracy to prevent them. The distance from Tokyo, the provincial culture, the founder’s personal authority—these were not features of a research strategy. They were features of a company shaped by its place and its history. They happened to be the features that allowed gallium nitride research to proceed.

The contrast with larger companies was sharp. At Toshiba, at Matsushita, at Sony, research was managed through committees and milestones. Projects were evaluated against competing priorities. Resources were allocated through processes designed to minimize risk and maximize return. The processes were rational. They produced rational outcomes. They killed gallium nitride research because gallium nitride research was irrational.

Nichia had no such processes. It had a founder who trusted his judgment and an engineer who wanted to work. The combination was enough.

The company’s financial position mattered. Nichia was profitable, though not spectacularly so. The phosphor business provided steady revenue. The company employed several hundred people, a size that allowed for flexibility. A larger company would have had more resources, but it would also have had more constraints. A smaller company would have had more flexibility, but it would have lacked the resources to fund fundamental research. Nichia occupied a middle position: large enough to absorb the cost of failure, small enough to approve the attempt.

The location mattered. Tokushima was far from the centers of the electronics industry, but it was not isolated. The prefecture had a university, technical institutes, a population educated in engineering. Nakamura had graduated from the University of Tokushima, a regional school without prestige but with competent programs. The company could draw on local talent without competing against the major corporations for the best graduates. The engineers who came to Nichia came because they were from Tokushima or because they had not been hired elsewhere. They were available, not elite.

The culture mattered. Nichia was a company where engineers could talk to the president. The hierarchy was flat, the distances short. An engineer with an idea could present it directly to the man who could approve it. There were no middle managers to filter proposals, no committees to deliberate, no processes to navigate. The engineer asked, and the president answered.

The structure created vulnerabilities as well as opportunities. The same lack of process that allowed Nakamura to pursue gallium nitride also meant that the project depended entirely on personal relationships. When Nobuo Ogawa died, when Eiji Ogawa consolidated control, the protection would disappear. The company would eventually confront the consequences of its decision to fund research that no other company would touch.

But that confrontation lay in the future. In 1989, Nakamura had his mandate. The founder’s backing was secure, a budget was available, and a material waited to be pursued. A year of training in crystal growth had given him technical capability, and a clear objective focused his work. A small laboratory in a chemical company far from the centers of the electronics industry had become the unlikely site of a serious attempt.

The company’s revenues told the story of its position. In 1989, Nichia’s sales were modest, measured in tens of billions of yen. The company was not poor, but it was not wealthy. It could afford to fund research, but the funding came with expectations. Nakamura would have to produce results.

The expectations were not formalized. There was no contract, no schedule, no specification of what constituted success. Ogawa had given his approval in trust, expecting that his engineer would work diligently and report honestly. The trust was personal, not institutional. It depended on Nakamura’s character and on Ogawa’s judgment of that character.

Nakamura was not an easy employee. He was stubborn, argumentative, resistant to authority. He had fought to continue his research when the new president tried to cancel it. He would fight again, many times, over many issues. The company that employed him was signing up for conflict.

But Nakamura was also a worker. A decade in production, small research projects, and preparation had equipped him for this moment. He had not been idle. His time had been spent learning, experimenting, developing the skills he would need. The request to pursue gallium nitride was the culmination of years of thought, not a whim.

The gallium nitride community was small and discouraged. Isamu Akasaki and his student Hiroshi Amano at Nagoya University continued to work on the material, producing results that the field largely ignored. Theodore Moustakas at Boston University pursued a parallel path, filing patents and publishing papers that attracted little commercial attention. The major companies had moved on. The field was a backwater.

Nichia entered this backwater because it did not know it was a backwater. The company had no history in semiconductors, no relationships with the research community, no investment in the prevailing consensus. It was a chemical company that made phosphors for fluorescent lamps. When its engineer proposed gallium nitride research, the company saw an engineer proposing research, not a field rejecting a material.

The ignorance was strategic. If Nichia had known what the electronics industry thought of gallium nitride, the company might have hesitated. If Ogawa had consulted experts, he would have been told that the material was hopeless. If Nakamura had attended the conferences, he might have absorbed the consensus. But Nichia did not know, Ogawa did not consult, and Nakamura did not attend. The research began in isolation, and the isolation protected it.

The protection was temporary. Success would bring attention, and attention would bring competition, and competition would bring conflict. The company that funded impossible research would eventually have to decide what that research was worth, and the engineer who performed it would eventually have to decide whether the company’s valuation was fair. Those decisions would destroy the relationship that made the research possible.

But the destruction lay in the future. In 1989, the relationship held. The founder trusted his engineer. The engineer trusted his founder. The company provided money and time and freedom. The engineer provided labor and commitment and hope.

Nakamura returned from Florida and began to build. The equipment he needed did not exist. The metalorganic chemical vapor deposition reactors available commercially were designed for materials other than gallium nitride, for processes other than the one Nakamura envisioned. He would have to construct his own.

The construction would take time. The failures would accumulate. The money would disappear. The company’s patience would be tested. Nakamura had his mandate and funding from Nichia; the pressure now shifted to the concrete, physical struggle of execution.