Why It Was Almost Impossible to Make the Blue LED
Why It Was Almost Impossible to Make the Blue LED
The blue LED was supposed to be impossible—until a young engineer proposed a moonshot idea.
Really annoying that the company shat on him for years, and continued to do so after he multiplied the value of the company. Toxic behavior.
It's an extreme example that perfectly illustrates how profit is extracted from employees by the employers. He didn't have any leverage to get a larger share of the profit from his labor, as is the case with most employees. You could call it toxic behavior, and it is, but it's the expected behavior, the behavior incentivised by the system.
It also shows how capitalism hinder innovation. It doesn't create it. The potentially innovative path took money without any guarantee of creating profit. It's bad business to be innovative. Capitalism prioritizing profit never chooses the best path, even if it gets a good ending eventually despite itself.
Makes me presume power harassment.
On the flip side, he was using up millions and millions of company dollars on his singleminded pursuit with no obvious results to show for it. Had things gone even a little differently, things would've gone very differently indeed. Hard to imagine most companies tolerating an employee flat ignoring instruction to change to another task when their old task was proving fruitless.
Hindsight is clear enough here, but in context it was pretty nuts what the guy was doing.
Makes you wonder how many great inventors of revolutionary tech were shoved off their path by dumb luck.
Probably far fewer than never had the opportunity to realize they could be great in the first place.
If greatness is one in a billion we have 8 (boy would the richest like us to believe that!). If it’s one in 100 million (I’m bad at math. I think it’s like) 80. Or if it’s one in a million, that’s 350 in the US alone. I’m inclined to lean toward the later, after all, if there aren’t a lot of greats waiting to be called up, how the fuck did we beat the odds by such a large margin??
Didn't fire him though. I don't think my boss would let me sit in an office doing my own project and binning notes from him for 20 years.
I bet he was ok with : Nobel prize bitch!
That's Japan, baybee. They love their toxic work culture. Thankfully, it is slowly changing with the younger generations, however.
Shuji Nakamura was a researcher at Nichia who was determined to create the first blue LED, which had eluded scientists for decades. Through innovative crystal growth techniques and materials discoveries, he succeeded in developing bright blue and white LEDs in the early 1990s. This breakthrough enabled LEDs to be used for full-spectrum lighting. Nichia's fortunes grew enormously as a result, though Nakamura was not properly compensated for his invention. Today, LEDs powered by Nakamura's blue LED technology are ubiquitous and have brought enormous energy savings worldwide.
Something interesting I found was that Nakamura persisted in his research for blue LEDs against the wishes of his company management, who saw it as a waste of resources. His stubbornness and belief in his work paid off by solving a problem that had stumped the electronics industry for 30 years.
He really got screwed. They didn't want him even working on blue LEDs and then when he was right and actually made one they gave him nothing and made hundreds of millions of dollars. Then sued him when he left to work for another company for "leaking company secrets" which was really all his work. He counter sued and the courts awarded him like 189 million, then the company counter sued back and he got 8 million which just covered his legal fees.
Wow... for brief, fleeting moment i thought justice prevailed in the end. Silly me.
Um... if someone pays you to do a thing, then they own it. Imagine if you paid me a hundred thousand dollars to build a house and then it's my house to live in. Doesn't make any sense at all.
I'm not defending the company, but the law is pretty clear on this. If you want to own your own work, then start your own company.
full-spectrum lighting
this doesn't sound true.
iirc full spectrum means "every wavelength" (like sunlight) and not just "3 wavelengths that add up to white".
That's true (here's an interesting video on the subject of colored light: https://www.youtube.com/watch?v=uYbdx4I7STg), but as mentioned at https://youtu.be/AF8d72mA41M?si=i8wjHjaKRaQbf23b&t=1516, once you have a blue LED, you can use a phosphor to convert the light to a range of longer wavelengths.
Excellent counter example to anyone claiming that we need patent and copyright to innovate.
This man made nothing on his invention and was not motivated by money but fame.
There are endless of examples of how those who do things for money hold back the creativity that leads to innovation. This is one of them. It almost didn't happen because his pursuit was not seen as profitable.
Sure, but the company fronted the millions of dollars required to develop the technology. The investment needs to come from somewhere.
That doesn't have to be a private company, though. We need public funding that retains the patent rights, if not just to make the invention free from licensing costs to manufacture.
The insane thing about our current system is that we do have public funding, but private companies wind up with the patent anyway
The company didn't invent it. A person did. The company almost stopped it from being invented. They didn't spend millions inventing this. A person spent tens of thousands of hours inventing it.
That the funding is only available from a company is a result of the patent system. It does not spur development, it perverts it. Any ideas to the contrary are propaganda.
People have been inventing shit longer than corporations have existed. People have been inventing things without any guarantee on return on investment for most of human history.
Capitalism is bullshit and the capitalization of ideas harms humanity.
My favorite thing about widely-available blue LEDs was the effect on TV scifi.
Watch the Star Trek shows made in the 1980s and 1990s and the tricorders, alien gadgets, and other props were always twinkling with red, yellow, and green LEDs to look futuristic. A generation later and every single hand prop on 2000s Doctor Who, Torchwood, etc. glowed and twinkled blue because the LEDs had just become cheap enough for prop makers, but weren't yet widespread in day-to-day life so the viewers were seeing something strange and unusual.
Now every color of LED imaginable is just common and whatever, but for a good stretch of time glowy blue became the standard "scifi" color just because that particular tech happened to turn up at that particular time.
9's sonic screwdriver
I'm not sure that LEDs were the thing that kicked off the trend. They made it easier to implement, but even in the 80s and 90s, you had things like Tron that might have kicked off the futuristic look with neon lines/tubes.
Making blue LEDs is easy. Just make a red one, then move towards it really fast.
Doppler effect?
https://en.m.wikipedia.org/wiki/Redshift#Blueshift
Essentially yes
Yeah but it only works if the source is moving towards the observer
This was an yet another glorious episode from veritasium.
I hope we get well past UVC LEDs. (i.e., shorter wavelengths) UV LEDs are already available. Unfortunately, this progress will stop before X-ray light. With +1 KeV energy, you pretty much must blast off the electrons from the atoms to emit X-rays, which an x-ray tube already does. Or by peeling off a piece of scotch tape.
Maybe making X-ray emitters cheap enough to put in a flashlight isn't the best idea anyway.
More efficient compact X-ray generators would be pretty huge for science work. We run the diffractometer in my lab at 2 kW and it still takes hours to get a good quality scan
Sorry sir I have no idea what you are talking about
-
Light = energy, shorter wavelengths= higher energy. Blue light has a shorter wavelength than red light. UV has even more energy. X-Rays have a lot more energy. For reference in the visible spectrum were talking about maybe 1-4 eV (this may be wrong, I'm too drunk to look it up rn).
-
If we want to produce light, the aim is to find an energy gap that has the exact energy gap that corresponds to the wavelength we're interested in. Typically this corresponds to an electronic transition, i.e. an electron "jumps" into a higher orbital, on its way down it will emit the energy difference as light.
2.1 X-Rays rn are produced by accelerating electrons onto a metal plate with high voltage. The impact of the electron "rips" out an electron in the close vicinity of the nucleus. Another electron will take the place of that electron, the energy gap associated with that process is large, which is why it produces X-Rays.
-
If we want to produce LEDs that emit in the far UV range we have to find large energy gaps in materials which is difficult. We still have to have a way to get the electron across the energy gap using electricity.
-
X-Ray LEDs are probably not realistic, as the energy of x-rays is so large that we have to rip out electrons from the close vicinity of the nucleus... which is already what we're doing with X-ray tubes.
-
Get past uvc for what purpose?
I imagine that lithography for integrated circuits would be an application, assuming you could make an appropriate photo-resist. The shorter the wavelength, the smaller the possible feature size. Current lithography relies on constructive and destructive interference between wavelengths to create super small features.
It seems that the blue led is picked by many manufacturers now for its coolness factor. There are so many appliances people have in sleeping areas with blue lights glaring and disturbing sleep
I carry a bit of gaffer's tape everywhere for those little obnoxious blue bastards.
It’s such a big problem that there’s literally LED dimming tape on the market. It’s semi-transparent tape that you stick over the blue LEDs, to knock them down to a more reasonable brightness. It’s akin to putting sunglasses on your appliances.
I just layer on masking tape until it's dim enough, usually 2 layers
The last like 3 computer cases have had blue power leds. They often are brighter than night lights. I have gotten to the point where I only connect the power button as the leds are so bright and the number of times I’ve inadvertently hit the reset button.
I made a hobby PCB with indicator LEDs and I have to admit that was the reason for my choice. The surface mount frosted "milky" ones look especially great, where they have the diffusion layer so it's not a harsh glare.
A few details as further info, focusing mostly on the technical aspects:
It's considerably easier to decrease the band gap than to increase it. Decreasing it only requires that you insert some material to provide an intermediate band, while increasing it would likely need alloying it to force some structural change.
The material being in the right band gap is not enough. You need to make sure that it can be p-doped and n-doped, that its crystalline structure is stable even with some temperature variation. Ah, it should be also relatively straightforward to produce industrially.
Then you get the little gem that Nakamura found.
Haven't watch the video yet, but I remember how impressed my step dad was with the blue LED when we got our PlayStation 2. I was like, yeah great whatever let's play games, at the time.
I love Veritasium's deep dives into the scientists behind various inventions. We really ought to celebrate more people like Nakamura.
The first blue LED I ever saw was on the dashboard of my mom's VW Golf. I always wanted one like that, but now they're everywhere!
Before there were blue LED, the indicator light for full beam was a blue tinted incandescent bulb. My parents had a Volkswagen Passat from the 1980s (?) where the usually blue indicator light for full beam was a green LED, since blue ones were not invented back then.
Very interesting. I think my mom bought her car in 1996
Here is an Odysee Link for those that don't want to give youtube support.
Just a PSA for those who don’t know… no shade against Odysee… I’ve just encountered folks here who don’t know this:
Veritassium and many others on YouTube make their living by the advertising shown on YT. If you’re a premium member, even more money goes to the creator when you watch their content. It’s this very money that allows independent creators to create more / better content!
I wish he posted his stuff on Nebula as well. His stuff totally fits the vibe of the platform, and would potentially make him even more money.
If they weren't comfortable with not getting YT ad revenue, they wouldn't be uploading their content to alternative sites.
Relying on YT as the gatekeeper to your entire livelihood also has a cost. It's not trivial to calculate but I imagine it's greater than the loss of AdSense money. There's a reason many people who rely on video content creation to survive hedge through the likes of Nebula, Floatplane or, indeed, Odyssey.
Capitalism is just sad
Here is an alternative Piped link(s):
https://piped.video/AF8d72mA41M
Piped is a privacy-respecting open-source alternative frontend to YouTube.
I'm open-source; check me out at GitHub.
Here's an Odyssey link too:
https://odysee.com/@veritasium:f/why-it-was-almost-impossible-to-make-the:9
Here's the lbry link to that Odysee link:
Samsung (etc) caught with their pants down
It makes you wonder how much longer it would take to commercialize blue LEDs without his contributions.
I didn't watch this, but i bet technology connections video about blue Christmas lights is more entertaining!
Which one? The man can't stop making videos about blue Christmas lights!
All of them! I like Veritasium though, not every video is a banger but this one definitely is
I freaking hate blue LEDs.
I actively avoid buying anything with a blue LED because they are so obnoxious. So bright. Why do I want to read by the light of my HDD? Does this video explain why they have to be like that?
Maybe if you have a separate wing of the mansion to do computing stuff it is not annoying. But if like a lot of people you have electronics in your living space, these lights are extremely disruptive.
It seems that can't really be dimmed.. I had to give up on a couple of blue backlit alarm clocks because there is no way that the time can be visible without illuminating the whole area around them.
For whatever reason, red is the best one. I would prefer another color aesthetically. For whatever reason, red is the only color that does what it has to do and nothing more.
This is actually a biological phenomenon that most humans experience! Our eyes are more attuned to greens and blues rather than reds, so green and blue light appear brighter as the cones in our eyes are more sensitive to those colors. Similarly, our cones are less sensitive to red so it appears darker.
There's also a physics component to this as well since red light has about half the energy (twice the wavelength) as blue light. However, since there's a difference in energy, the engineer must take that into account when designing multicolor LED applications so as to keep a level light intensity when changing or blending colors.
Here's an eli5 question with some more info: https://www.reddit.com/r/explainlikeimfive/comments/ghx9g6/eli5_why_does_red_light_seem_darker/?rdt=58820
I'm the guy that takes an awl and literally shatters them, or I will just place tape over top if they are recessed too far.
Front facing LEDs are a menace.
Awesome video. I had no idea about this story.
This video is recommended by Tournesol:
[33🌻] Veritasium: Why It Was Almost Impossible to Make the Blue LED#Tournesol is an open-source web tool by a non profit organization, evaluating the overall quality of the information in videos from community made comparisons, to fight against misinformation and dangerous content.
im guessing toshiba tried hard
Seen it, its well made and the end is almost like a good cliffhanger for the next Vid
If it was so hard to make the first one that hundreds of researchers couldn’t do it for years… then how is it so cheap and easy to do today?
Scale.
I don't get the dislikes, it is a great question. But as someone above, I think the cost was to dicover a way to make blue LED at all and make it scalable, then it's just like any other product.
Thanks - and yeah I absolutely pre-suppose that much. I was just hoping for a more specific answer. There could be just as interesting of a story about how it went from prototype —> ubiquity. I personally don’t just wave that process off and say “meh economies of scale.” Production scaling is often quite innovative too. And many lab innovations die on the vine because they can’t cross this chasm. So the fact that this one could, after being such a holy grail for so long, is something I would love to understand more.
