From: veritasium
Light-Emitting Diodes (LEDs) are components where the color of the light comes from their internal electronics, not from plastic covers [00:00:09]. The casing simply helps distinguish different LEDs [00:00:12].
Early Visible LEDs
In 1962, General Electric engineer Nick Holonyak created the first visible LED, which glowed a faint red [00:00:15]. A few years later, engineers at Monsanto developed a green LED [00:00:23]. For decades, only these two colors were available, limiting LEDs to applications like indicators, calculators, and watches [00:00:28].
The ability to create blue light was crucial, as mixing red, green, and blue light would allow for the production of white light and every other color, unlocking LEDs for all types of lighting globally [00:00:38]. This includes light bulbs, phones, computers, TVs, and billboards [00:00:48]. However, blue light proved exceptionally difficult to produce [00:00:54].
Throughout the 1960s, major electronics companies like IBM, GE, and Bell Labs competed to create the blue LED, recognizing its potential worth in billions [00:01:00]. Despite extensive research efforts, success remained elusive [00:01:12]. After 30 years, hope for using LEDs for general lighting faded [00:01:18], with a Monsanto director stating they would only be used in appliances and car dashboards [00:01:26].
How LEDs Work
LED stands for Light Emitting Diode [00:03:49]. Unlike incandescent bulbs, which generate light as a byproduct of heat from a tungsten filament, LEDs primarily create light, making them much more efficient [00:03:35]. A diode is a device with two electrodes that allows current to flow in only one direction [00:03:59].
Energy Bands and Semiconductors
In a solid, the outermost electrons of atoms form energy bands due to the influence of multiple nuclei [00:04:25].
- Valence Band: The highest energy band containing electrons [00:04:48].
- Conduction Band: The next higher energy band [00:04:53].
- Band Gap: The energy difference between the valence and conduction bands [00:05:19].
In conductors, the valence band is partially filled, allowing electrons to move easily [00:05:00]. In insulators, the valence band is full, and the band gap is large, preventing electron movement [00:05:16]. Semiconductors are similar to insulators but have a much smaller band gap, allowing some electrons to jump into the conduction band at room temperature [00:05:41]. The empty spots left behind in the valence band are called “holes” and act as positive charge carriers [00:05:57].
Doping and P-N Junctions
To enhance functionality, impurity atoms are added to the semiconductor lattice in a process known as doping [00:06:22].
- N-type semiconductors: Created by adding atoms (e.g., phosphorus to silicon) that contribute extra valence electrons, leading to mobile negative charge carriers (electrons) [00:06:29].
- P-type semiconductors: Created by adding atoms (e.g., boron to silicon) that have fewer valence electrons, creating mobile positive charge carriers (holes) [00:07:16].
When p-type and n-type semiconductors are joined, a p-n junction forms [00:07:48]. Electrons from the n-type diffuse to the p-type, filling holes and creating a “depletion region” where mobile charge carriers are absent [00:07:52]. When a battery is connected in forward bias, the depletion region shrinks, allowing electrons to flow from n to p [00:08:41]. When an electron falls from the conduction band into a hole in the valence band, the band gap energy is emitted as a photon of light [00:08:50].
The size of the band gap determines the color of the emitted light [00:09:06]. For example, pure silicon has a band gap of 1.1 electron volts, emitting non-visible infrared light used in remote controls [00:09:10]. Red and green LEDs were developed first because they require less energy and smaller band gaps than blue light, which needs a larger band gap [00:09:25].
Shūji Nakamura and the Blue LED
By the 1980s, despite hundreds of millions of dollars spent, no electronics company had successfully created a blue LED [00:09:40]. However, researchers understood that high-quality crystal with minimal defects was a critical requirement, as defects dissipated energy as heat instead of visible light [00:09:50].
Shūji Nakamura, a researcher at the small Japanese chemical company Nichia, took on the challenge [00:01:52]. Despite the semiconductor division being on its last legs and internal pressure to quit, Nakamura proposed a radical plan to Nichia’s founder, Nobuo Ogawa, to create the elusive blue LED [00:02:04]. Ogawa gambled, investing 500 million yen ($3 million) into Nakamura’s “moonshot project” [00:03:06].
Technical Breakthroughs
Nakamura focused on gallium nitride (GaN), a material largely abandoned by other researchers in favor of zinc selenide [00:13:00]. GaN presented three major problems:
- Crystal Quality: It was hard to grow high-quality GaN crystals due to a significant lattice mismatch (16%) with its sapphire substrate, leading to over 10 billion defects per square centimeter [00:13:06].
- P-type GaN: Scientists had only successfully created n-type gallium nitride [00:13:22].
- Light Output Power: Blue LEDs needed a total light output power of at least 1,000 microwatts to be commercially viable, two orders of magnitude greater than any prototype had achieved [00:13:32].
1. High-Quality Crystal: The Two-Flow Reactor
Nakamura first mastered Metal Organic Chemical Vapor Deposition (MOCVD), a technology for mass-producing clean crystals [00:10:17]. After assembling a new MOCVD system himself, he discovered that adding a second nozzle to the reactor, which released a downward stream of inert gas, pinned the reactant gases to the substrate, forming a uniform, high-quality crystal [00:16:33]. He called this his “two-flow reactor” [00:17:19]. This innovation allowed him to use gallium nitride itself as a buffer layer on the sapphire substrate, yielding even cleaner GaN crystals than previous methods [00:17:29].
2. P-type Gallium Nitride: Annealing
Akasaki and Amano at Nagoya University had previously achieved p-type gallium nitride by exposing magnesium-doped samples to an electron beam, though the reason for its effectiveness was unknown, and the process was too slow for commercial production [00:19:26]. Nakamura suspected energy was the key and successfully created a completely p-type sample by heating magnesium-doped gallium nitride to 400 degrees Celsius in a process called annealing [00:20:06]. This method was scalable and revealed that hydrogen atoms from ammonia (used in MOCVD) were bonding with magnesium, plugging the holes needed for p-type behavior. Annealing released the hydrogen, freeing the holes [00:20:30].
3. Efficiency and True Blue: Active and Blocking Layers
After creating a prototype blue LED that was still inefficient (42 microwatts) [00:21:20], Nakamura addressed the efficiency problem. He incorporated an active layer of indium gallium nitride (InGaN) at the p-n junction, which slightly shrinks the band gap, encouraging more electrons to fall into holes and bringing the light to a true blue color [00:22:02]. When electrons overflowed from this well, he created an opposite structure: an aluminum gallium nitride (AlGaN) layer with a larger band gap that blocked electrons from escaping the well [00:23:19].
By 1992, Nakamura had completed the complex structure of the blue LED [00:23:48]. He created a bright blue LED with a light output power of 1,500 microwatts and emitted a perfect blue at 450 nanometers, over 100 times brighter than previous prototypes [00:24:21].
Impact of LED Technology on Modern Lighting
Nichia announced the world’s first true blue LED in 1994, stunning the electronics industry [00:24:46]. Orders flooded in, and by the end of 1994, Nichia was manufacturing 1 million blue LEDs per month [00:25:03]. Within three years, the company’s revenue nearly doubled [00:25:11].
In 1996, Nichia made the jump from blue to white by placing a yellow phosphor over the blue LED. This chemical absorbs blue photons and re-radiates them across the visible spectrum, creating white light [00:25:15]. This breakthrough unlocked the final frontier for general LED lighting [00:25:33]. By 2001, blue LED products accounted for over 60% of Nichia’s revenue, which approached $700 million annually [00:25:40].
Advantages of LED Lighting
Compared to incandescent or fluorescent bulbs, LED bulbs offer numerous advantages:
- Efficiency: Far more efficient [00:28:12].
- Longevity: Last many times longer [00:28:14].
- Safety: Safer to handle [00:28:16].
- Customization: Completely customizable, with high-end bulbs offering 50,000 different shades of white [00:28:17].
- Cost-effectiveness: Their price has dropped, and their efficiency allows users to recoup the cost within months and save money for years [00:28:28].
This led to a lighting revolution; in 2010, only 1% of residential lighting sales were LED, but by 2022, it was over half [00:28:44]. Experts predict nearly all lighting sales will be LED within the next 10 years [00:28:56]. The energy savings from a full switch to LEDs could save an estimated 1.4 billion tons of CO2, equivalent to taking almost half the cars in the world off the road [00:29:03].
LEDs are now ubiquitous, found in house lights, streetlights, phones, computers, TVs, traffic lights, and displays [00:27:33]. The blue light from screens can, however, disrupt circadian rhythms if viewed before bed [00:27:50].
Legacy and Future Innovations in LED Technology
Despite quadrupling Nichia’s fortunes, Nakamura received only a modest salary increase and a 8 million, which primarily covered his legal fees [00:26:47]. This payout was minimal for an invention that now underpins an $80 billion industry [00:27:26].
In 2014, Shūji Nakamura, alongside Dr. Isamu Akasaki and Dr. Hiroshi Amano, was awarded the Nobel Prize in Physics for creating the blue LED [00:31:11].
Nakamura’s current research focuses on the next generation of LEDs, including:
- Micro LEDs: As small as five microns, these could be used for near-eye displays in AR and VR [00:29:21].
- UV LEDs: These could sterilize surfaces in hospitals or kitchens, killing pathogens like COVID-19 in seconds [00:30:04]. UV LEDs use aluminum gallium nitride, which has a much bigger band gap [00:30:27]. Currently, UV LEDs have an efficiency of less than 10% and high costs, but expected improvements in efficiency to over 50% could make them comparable to mercury lamps [00:30:37].
Beyond LEDs, Nakamura has also ventured into nuclear fusion research [00:30:58]. His career highlights his determination, critical thinking, and problem-solving skills, which allowed him to succeed where thousands of others failed [00:32:15].