Why did the blue LED take thirty years?

An LED makes light when electrons drop across an energy gap inside a crystal. Each drop gives off one particle of light, and the size of the gap sets its colour. Red and green LEDs were being made by the late 1960s. Blue needs a much wider gap, and the material that could do it, gallium nitride, defeated labs for three decades. Isamu Akasaki and Hiroshi Amano showed a bright blue LED in 1992, Shuji Nakamura found his own way at Nichia, and the three shared the 2014 Nobel Prize in Physics. Blue mattered because a blue LED with a yellow coating makes white light, and that's the LED lamp in your ceiling.

For most of history, light came from heat

A candle flame glows because of soot. University of Bristol chemists describe the bright yellow part of the flame as hot, carbon-rich soot particles heated until they glow. Only 0.4 percent of the candle's energy comes out as visible light. The UK's National Physical Laboratory had people measure their own candles at home and got between 15 and 86 watts of total power per candle. The share that becomes light, NPL notes, is tiny.

A light bulb does the same thing with electricity. The current heats a thin metal filament to about 3,000 °C, according to the same NPL report. The US Department of Energy says old incandescent bulbs released 90 percent of their energy as heat and only 10 percent as light.

The usual way to score a lamp is lumens per watt. A lumen measures light the way the eye responds to it, so the score is how much useful light you get for each watt you pay for. The Nobel committee's background notes put Edison's light bulb at about 16 lumens per watt and the fluorescent tube, which makes mercury vapour glow, at about 70.

An LED skips the heat

The Nobel committee's popular explainer describes it as a stack of semiconductor layers. One layer has spare electrons. The other has a shortage of them, which physicists call holes. A voltage pushes both into a thin active layer in the middle, and when an electron meets a hole it falls into it and gives off a photon, a particle of light. The whole light-making part is "no larger than a grain of sand."

The fall has a fixed height. Electrons in a semiconductor can only sit at certain energies, with a gap between the high ones and the low ones. Drop across a bigger gap and the photon carries more energy, which means a shorter wavelength. HyperPhysics, Georgia State University's physics reference, works one example. Red light at 683 nanometres is a photon of 1.82 electron volts.

So the colour of an LED is set by its material. The committee's scientific background gives two that matter here. Gallium arsenide has a gap of 1.4 electron volts, which gives infrared light you can't see. Gallium nitride has 3.4, which gives ultraviolet. Blue sits close to the top of that range.

Slide the gap below and watch the colour change, then drop an electron.

Try the three buttons. Gallium arsenide's gap gives infrared, which you can't see. Gallium nitride on its own gives ultraviolet. Blue is in between, and for decades nobody could build it.

Red and green came early

The first LEDs were built in several labs in the 1950s and 1960s, and they glowed from infrared up to green, according to the Nobel committee. By the late 1960s manufacturers in several countries were making red and green LEDs from gallium phosphide. Nick Holonyak Jr. and his colleagues at General Electric worked on a mix of gallium, arsenic and phosphorus that pushed the light into the visible range, and reported red laser light from it in 1962. Those early LEDs ended up as the red numbers in digital watches and calculators and as little on/off lights.

Everyone knew blue was the missing piece. White light for a room needs blue at the short end of the rainbow, so without blue there could be no white LED lamp.

Why blue took three decades

"The blue LED had remained a challenge for three decades," the Nobel press release says, despite considerable efforts in labs and in industry.

The obvious materials failed. Early attempts with zinc selenide and silicon carbide didn't give efficient light. Gallium nitride looked right on paper, but two practical problems made it look hopeless. Nobody could grow a gallium nitride crystal of high enough quality, because there was no fitting surface to grow it on. And it was close to impossible to make the p-type layer, the one with holes, which an LED can't work without.

Akasaki and Amano at Nagoya University cracked the crystal first. In 1986 they put a layer of aluminium nitride on sapphire and grew good gallium nitride on top of it. At the end of the 1980s they noticed their material glowed more brightly after it had been examined in an electron microscope, and that beam turned out to be making the p-type layer work. In 1992 they showed a diode that gave off bright blue light.

Nakamura, at Nichia, a small chemical company on the island of Shikoku, started on his blue LED in 1988. He grew a thin layer of gallium nitride at a low temperature first and the rest hotter, and got high-quality crystals by 1990. He also worked out why the electron beam helped. Hydrogen in the crystal was blocking the p-type layer, and simply heating the material drove it out. Nichia's company history marks 1993 as the year of its blue LED.

Both groups then improved their LEDs through the 1990s by mixing indium or aluminium into the gallium nitride. The blue LEDs in white lamps today are indium gallium nitride, the DOE's 2022 report notes.

Blue plus yellow makes white

Here's why a single colour won a Nobel Prize. In 1996, Nichia made what it calls the world's first white LED by combining a blue LED with a yellow phosphor. The phosphor soaks up some of the blue light and glows at longer wavelengths, and the Nobel background notes that the mix looks white. The other route is to put red, green and blue LEDs side by side and let the eye blend them.

The prize citation says it plainly. It went to Akasaki, Amano and Nakamura "for the invention of efficient blue light-emitting diodes which has enabled bright and energy-saving white light sources."

How good LED lighting got

The numbers kept climbing after the prize. A 2022 Department of Energy report says white LED packages, the light-making parts, went from about 35 lumens per watt to well over 200 in fifteen years. The best ones in production in 2020 reached 240 for a cool white, though most products sold were lower. The DOE's guide to efficient lighting says LED bulbs use at least 75 percent less energy than incandescent bulbs and last 25 times longer.

LEDs aren't heat-free, though. They still turn part of their power into heat, which is why the DOE report treats heat sinking as part of designing an LED lamp. But compared with a filament at 3,000 °C, they waste very little.

From a glowing crystal to your next question

Colour is a rabbit hole of its own. What bees see covers the ultraviolet just past where blue LEDs sit, and How Wi-Fi works follows radio waves instead of light. Learnfold turns a question like "why do some semiconductors glow and others don't?" into a planned course, and every follow-up becomes a linked page, so a curiosity rabbit hole gets a map. Building a course from your own question is part of Learnfold's paid subscription. It's an AI tutor and can make mistakes, so check its explanations against sources like the ones below.