The Origin of the LED: From the Discovery of Electroluminescence to a Global Lighting Revolution

In 1907, British wireless researcher Henry Joseph Round was conducting experiments passing voltage through silicon carbide (SiC) crystals when he noticed something strange. Light was leaking from a point on the crystal. He reported the observation in a letter of just 150 words to Electrical World, then moved on to other research.^[1] He offered no explanation of the principle, no account of why it happened. Neither Round himself nor anyone else at the time could have known that this single letter would mark the beginning of a technology that today accounts for more than half of the global lighting market.

A Discovery No One Noticed: The History of Electroluminescence

Round’s discovery is the earliest recorded instance of what we now call electroluminescence — the phenomenon in which a semiconductor or certain crystalline material, when an electric current is passed through it, converts electrical energy directly into light energy. Unlike an incandescent bulb, which first generates heat and then light from that heat, the energy is converted directly into photons.^[2]

Yet Round’s finding attracted little attention from the academic world of the time. The preoccupations of early twentieth-century physics lay elsewhere, and the theory of semiconductors had not yet been established. It was a young Russian inventor who first took electroluminescence seriously.

Oleg Losev worked in the 1920s at a radio research institute in Nizhny Novgorod in the Soviet Union, where he systematically studied the emission of light from crystal detectors used in radio receivers.^[3] In 1927 he published a detailed research report describing the electroluminescent properties of silicon carbide and zinc oxide crystals, and the following year filed a patent application for a device applying the phenomenon under the name “light relay.”^[3] Losev was among the first to attempt a quantum mechanical interpretation of why the phenomenon occurred, rather than merely describing it.

Yet Losev’s research remained poorly known both within the Soviet Union and in the West. The reasons were several: the language barrier of the USSR, the upheaval of the Second World War, and the tragedy of Losev’s own death. He died of starvation during the Siege of Leningrad in 1942. He was 39 years old.^[3] It was not until the 1950s, when semiconductor physics began to develop in earnest, that the academic world rediscovered his work.

Invention Without Infrastructure: The Semiconductor Revolution of the 1940s and 1950s

The reason Round’s and Losev’s discoveries remained confined to experimental observation was not a problem with the physical devices themselves. The theoretical foundation needed to understand them simply did not exist. In 1947, when John Bardeen, Walter Brattain, and William Shockley invented the transistor at Bell Labs, semiconductor physics was opened up in earnest. The three were awarded the Nobel Prize in Physics in 1956.^[4]

The arrival of the transistor was also a prerequisite for LED development. To understand and control electroluminescence, one had to understand the operating principle of the p-n junction — the interface between a p-type semiconductor (where holes are the majority carrier) and an n-type semiconductor (where electrons are the majority carrier).^[5] An LED is essentially a p-n junction diode. When current flows through the diode in the forward direction, electrons and holes combine at the junction and release energy; that energy emerges as light. The wavelength of the light emitted is determined by the bandgap energy of the semiconductor material used.^[5]

From the mid-1950s, various researchers began conducting experiments to produce electroluminescence in III-V compound semiconductors such as gallium arsenide (GaAs) and gallium phosphide (GaP). However, the light emission at this stage was mostly in the near-infrared region and not efficient enough to produce visible light.

The First Visible-Light LED: Nick Holonyak’s Achievement

In October 1962, American electrical engineer Nick Holonyak Jr. created the world’s first visible-light LED at General Electric’s Syracuse laboratory.^[6] The material he used was gallium arsenide phosphide (GaAsP), whose bandgap corresponds to energies in the red visible-light range.

Holonyak had been John Bardeen’s first graduate student — that is, he had learned semiconductor physics directly from the Nobel laureate who co-invented the transistor.^[7] Holonyak himself was convinced this invention would lead to practical lighting. From the moment he announced his red LED, he publicly argued that “this will eventually replace Edison’s light bulb” — a statement that struck contemporaries as wildly unrealistic.^[7]

The LED of 1962 was a long way from a practical lighting device. The first commercial LED was, in the same year, an infrared LED (SNX-100) developed by James Biard and Gary Pittman of Texas Instruments, priced at $130 per unit — equivalent to roughly $1,300 today.^[8] Holonyak’s visible red LED was also extremely expensive in its early days, and its brightness was far too low for indoor lighting. Nevertheless, the technology soon found its footing as indicator lights for calculators, digital watches, and electronic equipment.

Green, Yellow, and the Brightness Revolution

The decade following the introduction of the red LED was a period of expanding the color palette and improving brightness. In 1972, M. George Craford, a former student of Holonyak’s, developed the first yellow LED while working at Monsanto. He also succeeded in increasing the efficiency of the red LED by more than tenfold.^[9] In the same year, Herbert Maruska and Jacques Pankove at RCA Laboratories produced a violet LED using a thin film of magnesium-doped gallium nitride (GaN).^[9] This research on magnesium-doped GaN would later prove an important stepping stone on the path to the blue LED.

In 1976, T. P. Pearsall designed a high-brightness infrared LED for optical fiber communications. This development broadened the applications of LED technology from lighting into the field of telecommunications.^[10]

In the 1980s, high-efficiency red and infrared LEDs using aluminum gallium arsenide (AlGaAs) materials were commercialized, followed by aluminum gallium indium phosphide (AlGaInP) materials that enabled efficient emission across the range from red to yellow-green.^[11] But the blue LED remained out of reach. Even with red and green available, without blue it was impossible to combine the three primary colors of light to produce white. The absence of blue was the decisive barrier preventing LEDs from advancing into general-purpose lighting.

Thirty Years of Obstruction: What Stood in the Way of the Blue LED

Why was the blue LED so difficult? Blue light carries more energy than red. Therefore, producing a blue LED required a semiconductor material with a larger bandgap. From the 1970s, researchers focused on two candidate materials: zinc selenide (ZnSe) and gallium nitride (GaN).

GaN had theoretically ideal bandgap properties. But growing GaN crystals was extraordinarily difficult. It was hard to find a substrate material whose crystal structure and lattice constant matched those of GaN, and p-type GaN in particular was a seemingly insurmountable problem. Without p-type GaN, a p-n junction cannot be formed; without a p-n junction, an LED cannot function.^[12]

Many research teams abandoned GaN and shifted toward ZnSe. The prevailing view in the academic world tilted toward “it is effectively impossible to produce a blue LED with GaN.” By the late 1980s, mainstream research funding had dried up in this area. It was at precisely this point that Isamu Akasaki of Nagoya University and his student Hiroshi Amano began to find a way through.

Nagoya and Nichia: Making the Impossible Possible

In 1986, Akasaki and Amano succeeded in producing a high-quality GaN thin film by first growing an aluminum nitride (AlN) buffer layer on a sapphire substrate and then depositing the GaN crystal on top of it.^[13] This was a decisive technological breakthrough. Where previous GaN had been riddled with defects, it was now possible to obtain a far more uniform and pure crystal.

The next obstacle was p-type GaN. In 1989, while observing a magnesium-doped GaN specimen under an electron microscope, Amano accidentally discovered that irradiating the material with the electron beam changed its properties. The electron beam was removing hydrogen atoms around the magnesium dopant and activating the p-type characteristics.^[13] Without this discovery, p-type GaN — and therefore the LED — would not have followed.

At the same time, a researcher at Nichia Chemical Industries, a small chemical company in Tokushima, was independently tackling the same problem. His name was Shuji Nakamura. The major corporations had by this point either abandoned GaN research or already pivoted toward ZnSe. Nakamura’s choice of GaN was, paradoxically, motivated by the absence of competition.^[14]

In 1988, Nakamura persuaded the company’s president, Nobuo Ogawa, to fund a year of research at the University of Florida (Gainesville) along with more than three million dollars in research funding. Upon his return in 1989, he modified a metal-organic chemical vapor deposition (MOCVD) apparatus suited to GaN crystal growth and developed an original technique he called “two-flow MOCVD.”^[14] This modification of the equipment was the key to producing high-quality GaN thin films.

In 1992, Nakamura found a way to produce p-type GaN: activating magnesium-doped GaN through thermal annealing rather than electron-beam irradiation. This method, far more practical than electron-beam exposure, opened the door to mass production.^[14] In 1993, Nakamura and Nichia announced the world’s first high-brightness blue LED. The following year, 1994, they announced a cyan LED as well, dramatically expanding the spectrum of colors that LEDs could produce.^[15]

White LEDs and a New Era of Lighting

The arrival of the blue LED immediately made the next goal — white LED — achievable. There are broadly two ways to produce white light with LEDs. The first is to combine red, green, and blue LEDs and mix the three primary colors of light. The second is to coat a blue LED with a phosphor that, when stimulated by the blue light, emits yellow light; the two colors combine to appear white.^[16]

The second approach is by far the most widely used today. The decisive moment came in 1996, when Nichia commercialized white LEDs using a blue LED coated with yttrium aluminum garnet doped with cerium (YAG:Ce) phosphor.^[16] This structure was simple to manufacture, highly efficient, and above all amenable to mass production.

From the late 1990s through the 2000s, the efficiency of white LEDs rose rapidly. The luminous efficacy of early white LEDs was around 10 lm/W (lumens per watt), but by 2005 it had surpassed 100 lm/W, and by the 2010s laboratory-level results exceeding 300 lm/W had been reported. By comparison, a standard incandescent bulb achieves around 15 lm/W and a fluorescent lamp 50–100 lm/W.^[17] LEDs became not merely a new option but a technology that surpassed every previous lighting technology in efficiency.

The 2014 Nobel Prize in Physics: Belated Recognition, and Controversy

In 2014, the Royal Swedish Academy of Sciences awarded the Nobel Prize in Physics to Isamu Akasaki, Hiroshi Amano, and Shuji Nakamura. The citation was “for the invention of efficient blue light-emitting diodes which has enabled bright and energy-saving white light sources.”^[18] The Nobel Committee described the invention as “a 21st-century invention comparable to the invention of the light bulb in the 20th century.”^[18]

The award, however, was not without controversy. After leaving Nichia, Nakamura filed a lawsuit against the company demanding a share of the patent revenues. In 2004, a Japanese court ruled that Nichia should pay Nakamura approximately 20 billion yen (around $190 million), but on appeal the amount was drastically reduced to 840 million yen through a settlement.^[19] The case provoked wide-ranging debate about the rights of inventors, the relationship between corporations and individual researchers, and Japan’s intellectual property culture.

Nick Holonyak lived to see the achievements of his students Craford, Nakamura, and others, passing away in 2022 at the age of 94. In 2021 he belatedly received the Queen Elizabeth Prize for Engineering, but the Nobel Prize never came to him.^[7]

From Component to Ecosystem: The Spread of LED Applications

After the arrival of the blue LED, LED technology transcended the invention of a single component and gave rise to a vast technological ecosystem.

Displays: The backlights of smartphones, tablets, and laptop screens use LEDs in most cases. Mini LED and micro LED technologies are opening a new display paradigm in which each individual pixel is itself an LED. Micro LED in particular promises higher brightness and longer lifespan than OLED, and has been partially commercialized in large-format displays.^[20]

Signals and transportation: LED traffic signals consume 80–90% less energy than conventional incandescent-bulb systems and have lifespans more than ten times longer.^[17] In most countries today, LED traffic signals have become the standard. Automotive headlights and taillights are also converting to LED at a rapid pace.

Horticulture and agriculture: Plants photosynthesize only in response to specific wavelengths of light. Because LEDs can precisely control red and blue wavelengths, they have become an ideal light source for vertical farms and greenhouse lighting that optimizes plant growth while improving energy efficiency.^[20]

LiFi and optical communications: LEDs are also a key component of visible light communication (VLC), or LiFi. This technology transmits data by modulating LED lighting at very high speeds — imperceptible to the naked eye — and is being researched as a near-field wireless communications medium capable of complementing or replacing Wi-Fi.^[20]

Medicine: LEDs are used across various medical fields, including photodynamic therapy using blue-light wavelengths, phototherapy for neonatal jaundice, and sensors in medical devices.^[20]

The Efficiency Paradox: The Rebound Effect

The spread of LEDs has dramatically reduced energy consumption in the lighting sector. According to the U.S. Department of Energy, the energy savings from LED lighting by 2035 are projected to reach 569 TWh per year — equivalent to the annual output of 92 power plants with a capacity of 1,000 MW each.^[17]

Yet there are analyses suggesting this savings may not be as large as expected. As lighting efficiency improves, people may use more light — the so-called “rebound effect.”^[21] As formerly expensive lighting became cheap, outdoor nighttime illumination expanded explosively, and urban light pollution increased alongside it. Recent research notes that even after the transition to LEDs, the brightness of artificial light at night around the world has actually increased.^[21]

This paradox — that improvements in efficiency do not necessarily translate into reductions in total consumption — is not a limitation of the technology itself, but a problem of how the technology is used and of incentive structures. The tool that is the LED is complete. How it is used remains an open question.

A Short Letter, a Long Journey

From the moment Round observed light emerging from a silicon carbide crystal in 1907 and recorded it in 150 words, 86 years passed before the blue LED became possible. Those 86 years were a mixture of theoretical voids, the Cold War, language barriers, corporate logic, the tenacity of individual researchers, and multiple verdicts of “impossible” from the academic world.

What is striking is that all the key breakthroughs that completed the blue LED came from outsiders. Losev explored alone in a remote Soviet research institute; Nakamura pressed ahead with GaN research — abandoned by major corporations — as a researcher at a small company; and Amano unlocked the secret of p-type GaN through an accidental observation made during an experiment. The technology that makes light, looked at from a slightly different angle, is also something that outsiders kept lit.