That was a great story! We have the blue ones because one guy was willing to put it all on the line to get it done.
Amazing so much hangs on just one of us sometimes.
We also have it because the research scientist acquired real electromechanical skill. Most of the time those skills are not there and my mind is on fire thinking about what could be done, and done faster with that kind of know how more broadly distributed.
Not having that PhD sucked mostly due to peers not valuing other skills.
I know a chemistry professor who values these things. I met him while setting some polymer equipment up. (Limiting details here to keep from outing people who may well read here. (Hello, from you know who in Oregon!))
Basically, this prof has a parts and equipment depot. Anytime there is an opportunity to score inexpensive, relevant gear, they do it.
Students often build the gear they need. This may not be science grade, but it is almost always enough to validate a research path, or some other plan, including procurement or access to science grade equipment later on.
In my discussions, those students live the program and know the value they are getting.
Essentially, it is the same high value our Blue LED making friend has seen; namely, more direct agency and control with far fewer, maybe even zero dependencies navigate.
They can explore even higher risk areas of research and then upon seeing potential outcomes worth publishing, can put their stuff to work how they need, when they need.
A quick look back through history shows us a whole lot of the hard won scientific understanding we value and depend on, engineer with, came to us via people who could make things as well as think and observe. Add computation to that list as well.
Academia could use a whole lot more of this as could public research and even private research programs.
And as can be seen, not only incandescent but also LED lighting was only made possible when it was by truly Edisonian efforts.
>Anytime there is an opportunity to score inexpensive, relevant gear, they do it.
Up into the 1980's things were done a bit differently than they are now in industrial research when it comes to equipment.
For energy, well-funded places like Exxon and Shell would store and accumulate used equipment in surplus warehouses when they recommissioned laboratories or replaced individual gear with the latest & greatest. There it would age for 5 to 10 years on average before being tagged for discard.
The material was traditionally being held as a resource as in previous decades, when it was expected that principal investigators would look first in the vast storehouse for useful items before requisitioning & purchasing new equipment for their labs. But nobody was doing that any more, energy had skyrocketed in price and oil companies had plenty of money so they had only been buying new equipment for years.
These were big warehouses, but eventually they would fill up and stay full, and they needed to make room for more on a regular basis so things were auctioned off.
I ended up with a very small (carefully selected) fraction of what was passing through those warehouses, and it was still a nice multiple of the tonnage that any one PhD had at their disposal during an average career. A lot of them don't want to touch the equipment anyway, they make the interns do it. So it's not often the most scientifically advanced one in the lab leveraging their hands-on experience, and conversely seldom the most capable hands-on operator having their abilities leveraged most scientifically.
I collaborated with some of their people who would visit my lab at the time, plus non-research customers and there was nothing to be ashamed of using their second-hand equipment which still had inventory stickers from the original owners. I was constantly validating equal or superior performance to their own in-house work.
They would never think of using their own surplus equipment or even going down to the warehouse to see how much overwhelming tonnage there actually was.
>>The material was traditionally being held as a resource as in previous decades, when it was expected that principal investigators would look first in the vast storehouse for useful items before requisitioning & purchasing new equipment for their labs.
I thought this was a really well-produced video! It's difficult to communicate science to the public in an accessible way at the right level, and I think Derek does a commendable job.
I really liked the LED explanation at the 4:00 mark. Can anyone who is familiar with semiconductor physics opine on how well this explanation models the reality?
Bachelor's in engineering physics (condensed matter experimental)/EE specializing in semiconductors here. The explanation starting at 4:00 is very accurate.
When he talks about the electrons "feeling" the neighboring atoms, he's talking specifically about a result that follows from the materials being crystalline, that is, having regular ordered structure. The regular structure gives rise to a periodic potential. You plug that periodic potential into the Schrodinger equation and apply continuity conditions and translational symmetry to the wavefunction. Computing the solutions to the Schrodinger equation with those conditions reveals that there are allowed and disallowed energy levels, and also reveals the relationship between energy and momentum in the crystal lattice. You can step through this by reading the wikipedia page on the Kronig-Penney Model. This depends on the periodicity, which obviously can change depending on direction in a crystal.
His explanation, and the result that "the" band gap is a single number, isn't dishonest because when we grow semiconductor devices, we grow them such that the crystal is oriented such that current flows in the desired direction, so that simple result holds true.
Even his portrayal of the bands leaning down as potential/voltage is applied mirrors how potential change is shown in diagrams of semiconductor devices, see Streetman and Banerjee - Solid State Electronic Devices.
I liked it, though it bugs me a little when people equate infrared and heat. Infrared is light. Light can heat things, and hot things can glow, but "infrared is heat" isn't exactly right.
I know basically nothing about physics, so sorry if this is a dumb question.
The existence of infrared LEDs seems to indicate to me that infrared light can exist without heat.
The existence of infrared thermometers seems to imply that hot stuff radiates infrared light, at least usually.
So my question is, is there any case where heat does not cause infrared radiation?
What are those cases? Some special materials? Special colours(perhaps outside the visible spectrum)?
I agree with you, but I think that for the general public you have to relate to what they experience, and thus intuitively know, and that is that heat “seems” to be different from light.
I would agree. Having said that I still think it was a bit wishy washy. The whole treatment of band gap energies I think is quite complicated beyond the simple diagram shown.
The explanation is really well done, it captures the essence of the Pauli exclusion principle without delving too deeply into the weeds. In my opinion the best part of the video is the explanation of the "hole" quasiparticle at 6:10 (I learned this as a pseudo-particle but will defer to Wikipedia [1]).
While a great introduction to semiconductor behavior this does gloss over a very important detail namely direct vs indirect semicondoctors as some others have mentioned. In the video the detail that's glossed over relates to the nature of crystals, namely that they're highly ordered repeating structures but that they don't look the same when viewed from every direction. This means that there isn't a single band-gap but multiple ones depending on the direction of the crystal you're contemplating.
At this point you may reasonably ask why the direction matters and now we unfortunately get deep into the weeds with quantum mechanics again. When a single photon is absorbed in the semiconductor system both momentum and energy must be conserved. The momentum of the photon for something like the Silicon bandgap is quite small (something like the equivalent of an electron traveling at 1500m/s) while the momentum of room-temperature conduction electrons is substantially faster [2] so as a very slight simplification transitions due to the absorption of photons are not accompanied by a change in momentum and so we only care about the band structure (and the accompanying free carriers) associated with a particular crystal direction.
In particular in Silicon you have what's called an indirect bandgap, namely the minimum energy conduction band electrons have a different momentum from the valence band holes ([3]) and as a consequence while you can _absorb_ a photon in order to make a detector you cannot make it efficiently _emit_ a photon as an LED should (something the video got wrong).
None of this matters for the heart of the video, which focuses blue LEDs in the GaN materials system which is definitely a direct bandgap material, however if someone does manage to create a manufacturable light emitter in pure Silicon expect an absolute revolution with regards to optical computing and photonics. (Not for lack of trying, this has been the holy grail for at least 20 years, possibly longer)
That explanation closely follows the outline of the equally good and very accessible explanation offered in the Halliday/Resnick/Walker's Fundamental's physics (11 edition, chap. 41)
I agree. Setting how LEDs work aside, I never really got how semiconductors worked, despite reading about it and talking with experts for years, until this video.
I mean, I could explain how they worked in the same ways that they were explained to me, but I couldn't connect those explanations to a true physical understanding.
But thanks to this, I finally actually understand.
Thanks for the billions in revenue, but remember how I told you to quit trying to use GaN to solve the blue LED problems? Well here's $147 for your patent and clean out your desk, because you're fired.
I feel like I remember he complained that Japan would likely fall behind by their best and brightest going outside Japan knowing that inside they would not be compensated.
My search-fu is failing though. I did find this interview
a good leader need to be humble enough to spot when they are wrong and correct course. This guy sounds like he had control issues, and kept a grudge. very childish behavior.
There's little of that to go around in Japan. They strongly subscribe to power structures (see Senpai-Kohai) - seniors are unimpeachable and highly respected.
After the blue LED came onto the market in the 1990s, it took less than a decade for them to become shelf items of a couple dollars each at electronics retailers.
It was also around that time that web-based communities of computer technicians really took off. Web forums, etc.
The coincidence led (yup!) to a love-at-first-sight relationship. Funny as it may seem now, being a mere 20 years later (or: "holy crap, it's been 20 years!, how did that happen??"), there were a few years there in the 2000-2005 region during which the de-riguer of computer nerdery was to go blue LED crazy.
It felt elite, cutting edge, rare, oh-so-techy. And it's funny now, to look back at the windowed PC cases full of LEDs and garishly lit by cold cathode tubes, with our mouses and speakers and other electronic gadgets painstakingly swapped over to blue.
And within a further couple of years - from about 2005 onward, if not sooner - the commercial market had taken over the trend and made it boring, passe. We hackers and overclockers weren't interested any more. Indeed, these ultra bright things began to get annoying. Within about 5 years the modding scene's blue LED craze began, peaked, commercialised, became a liability ... at which point we began to hastily obscure our blinding modifications with another, very different, product whose very identity hinges upon the colour blue : Blu-Tack! [0]
So where did it all end up, this short-lived cultural crossover between blue LEDs and hackers? Well, basically, the commercial market morphed into the "RGB" movement of lit-up computer hardware. RGB fans, RGB cases, RGB panels on graphics cards, etc. But I still think the blue LED is pretty cool.
Yes - there was Time that every consumer device insisted on using extremely bright blue LEDs for “power on” status. Even mundane things like DVRs or TVs. It became so annoying that various forms of tape were use to cover them.
I still remember the blue power LED on my Toshiba laptop fading on and off in a pattern that was not quite the patented Apple "breathing" pattern, but close enough to suggest it.
Which is why I was always baffled by the decorative lights in my old office.
It was a wall with small scattered lights in different colors, so they used recessed LEDs. Fine. But instead of color LEDs with a neutral diffuser, it had red+green+blue triple LEDs to make white light, with a red/green/blue plastic in front to recolor it!
I understand how this could be cheaper to assemble or maintain, but I'll never not balk at a system that has components undoing each other's work. Feels almost disrepectful to the technology.
The blue is nicer if you do that.
Technology Connections has made 3 videos over the last 5 years mostly centering around how he hates the blue led lights used in holiday lights. I think I've only seen the 2 yr old one, but now that I have I can't unsee it. The blue is just too blue. If you see a set that is all blue instead of multi-color it's unbearable. It's just too blue. White light in blue plastic is where it's at.
Something odd about that ultra-blue color - when I'm outside at night, I can see any lit up sign in relatively good focus with my glasses on. Green signs, red signs, etc.
But anything blue always looks blurry unless I'm very close to it.
But the commenter said it was red, green, and blue LEDs together, with a blue diffuser over them. Depending on the diffuser, that could produce a more pleasant result (by allowing some monochromatic red and green through), but it presumably wouldn't solve the underlying problem that monochromatic blue light can be unpleasant.
Are you sure they were RGB triplets and not white LEDs? Either way, is fun to imagine continuing this - grouping three of these filtered LED lights to make a new white light source, which you can then again filter with translucent plastic, etc!
Imagine, if you built a giant grid of such things, and could modulate the tranlucency 30 times a second or so, you could show some sort of.... moving picture show.
Though it is incredibly frustrating how ungrateful Nichia Corp was to Mr. Nakamura, the underdog who pushed through every obstacle to ultimately give them the vehicle for more than 65% of their revenue!
People like the Nichia CEO at the time, a nephew who inherited the business nepotism-style (ditching the successful methodologies of his uncle) are just goddamn fools. Any success is in spite of their unimaginative, bean counting petty mindedness fighting tooth and nail every decimeter of the way.
I think it was narcissism. Nakamura ignored direct orders to desist for years, and wound up proving that those orders were massive strategic blunders. So insubordination combined with making his superior look a fool. (On the other hand it's pretty clear we're only getting Nakamura's side of the story.)
I'm totally with you, but what's the other guys story going to be?
"This weird researcher wouldn't follow orders, and I didn't dig deeper to understand his level of commitment or anything. If I had, I would've at least seen his level of dedication and possibly rallied to support him."
That's the best possible version, and highly unlikely since, as you noted, narcissism.
Anytime I see someone intelligently committed to a cause the way Nakamura was, I respect it and will support them however I can. Even if it doesn't pan out, it's still a good story.
It's not narcissism. I have dealt with this situation myself. It's recognizing that the boss is in over his head technically and is making the wrong decisions. This situation happens when the researcher cannot explain clearly why he knows a particular strategy is the correct one or why he knows it will work. Yes, it's confidence of a particular kind, one that you get when you understand every detail, and it requires trust.
The first ten minutes is the best explanation of conductors, insulators, and semiconductors I’ve seen. The rest is a gripping human story with science sprinkled in. The maker / experimentalist spirit strongly resonates with me. There’s a line at the end for climate change and fusion folks.
This video is to adult me what Back to the Future was to kid me: it has it all.
Kudos for mentioning Back to the Future. When I was 8/10 yo, I could recite all the lines from the first movie up to 15 min or something. I really, really loved it.
Readit News
Amazing so much hangs on just one of us sometimes.
We also have it because the research scientist acquired real electromechanical skill. Most of the time those skills are not there and my mind is on fire thinking about what could be done, and done faster with that kind of know how more broadly distributed.
Not having that PhD sucked mostly due to peers not valuing other skills.
I know a chemistry professor who values these things. I met him while setting some polymer equipment up. (Limiting details here to keep from outing people who may well read here. (Hello, from you know who in Oregon!))
Basically, this prof has a parts and equipment depot. Anytime there is an opportunity to score inexpensive, relevant gear, they do it.
Students often build the gear they need. This may not be science grade, but it is almost always enough to validate a research path, or some other plan, including procurement or access to science grade equipment later on.
In my discussions, those students live the program and know the value they are getting.
Essentially, it is the same high value our Blue LED making friend has seen; namely, more direct agency and control with far fewer, maybe even zero dependencies navigate.
They can explore even higher risk areas of research and then upon seeing potential outcomes worth publishing, can put their stuff to work how they need, when they need.
A quick look back through history shows us a whole lot of the hard won scientific understanding we value and depend on, engineer with, came to us via people who could make things as well as think and observe. Add computation to that list as well.
Academia could use a whole lot more of this as could public research and even private research programs.
Again, great story. Love it.
This is what made the white LED possible.
And as can be seen, not only incandescent but also LED lighting was only made possible when it was by truly Edisonian efforts.
>Anytime there is an opportunity to score inexpensive, relevant gear, they do it.
Up into the 1980's things were done a bit differently than they are now in industrial research when it comes to equipment.
For energy, well-funded places like Exxon and Shell would store and accumulate used equipment in surplus warehouses when they recommissioned laboratories or replaced individual gear with the latest & greatest. There it would age for 5 to 10 years on average before being tagged for discard.
The material was traditionally being held as a resource as in previous decades, when it was expected that principal investigators would look first in the vast storehouse for useful items before requisitioning & purchasing new equipment for their labs. But nobody was doing that any more, energy had skyrocketed in price and oil companies had plenty of money so they had only been buying new equipment for years.
These were big warehouses, but eventually they would fill up and stay full, and they needed to make room for more on a regular basis so things were auctioned off.
Cashthedayofthesaleasiswhereisnowarrantiesofany kind.
I ended up with a very small (carefully selected) fraction of what was passing through those warehouses, and it was still a nice multiple of the tonnage that any one PhD had at their disposal during an average career. A lot of them don't want to touch the equipment anyway, they make the interns do it. So it's not often the most scientifically advanced one in the lab leveraging their hands-on experience, and conversely seldom the most capable hands-on operator having their abilities leveraged most scientifically.
I collaborated with some of their people who would visit my lab at the time, plus non-research customers and there was nothing to be ashamed of using their second-hand equipment which still had inventory stickers from the original owners. I was constantly validating equal or superior performance to their own in-house work.
They would never think of using their own surplus equipment or even going down to the warehouse to see how much overwhelming tonnage there actually was.
It just wasn't done.
>>The material was traditionally being held as a resource as in previous decades, when it was expected that principal investigators would look first in the vast storehouse for useful items before requisitioning & purchasing new equipment for their labs.
I really liked the LED explanation at the 4:00 mark. Can anyone who is familiar with semiconductor physics opine on how well this explanation models the reality?
When he talks about the electrons "feeling" the neighboring atoms, he's talking specifically about a result that follows from the materials being crystalline, that is, having regular ordered structure. The regular structure gives rise to a periodic potential. You plug that periodic potential into the Schrodinger equation and apply continuity conditions and translational symmetry to the wavefunction. Computing the solutions to the Schrodinger equation with those conditions reveals that there are allowed and disallowed energy levels, and also reveals the relationship between energy and momentum in the crystal lattice. You can step through this by reading the wikipedia page on the Kronig-Penney Model. This depends on the periodicity, which obviously can change depending on direction in a crystal.
His explanation, and the result that "the" band gap is a single number, isn't dishonest because when we grow semiconductor devices, we grow them such that the crystal is oriented such that current flows in the desired direction, so that simple result holds true.
Even his portrayal of the bands leaning down as potential/voltage is applied mirrors how potential change is shown in diagrams of semiconductor devices, see Streetman and Banerjee - Solid State Electronic Devices.
The existence of infrared LEDs seems to indicate to me that infrared light can exist without heat.
The existence of infrared thermometers seems to imply that hot stuff radiates infrared light, at least usually.
So my question is, is there any case where heat does not cause infrared radiation? What are those cases? Some special materials? Special colours(perhaps outside the visible spectrum)?
While a great introduction to semiconductor behavior this does gloss over a very important detail namely direct vs indirect semicondoctors as some others have mentioned. In the video the detail that's glossed over relates to the nature of crystals, namely that they're highly ordered repeating structures but that they don't look the same when viewed from every direction. This means that there isn't a single band-gap but multiple ones depending on the direction of the crystal you're contemplating.
At this point you may reasonably ask why the direction matters and now we unfortunately get deep into the weeds with quantum mechanics again. When a single photon is absorbed in the semiconductor system both momentum and energy must be conserved. The momentum of the photon for something like the Silicon bandgap is quite small (something like the equivalent of an electron traveling at 1500m/s) while the momentum of room-temperature conduction electrons is substantially faster [2] so as a very slight simplification transitions due to the absorption of photons are not accompanied by a change in momentum and so we only care about the band structure (and the accompanying free carriers) associated with a particular crystal direction.
In particular in Silicon you have what's called an indirect bandgap, namely the minimum energy conduction band electrons have a different momentum from the valence band holes ([3]) and as a consequence while you can _absorb_ a photon in order to make a detector you cannot make it efficiently _emit_ a photon as an LED should (something the video got wrong).
None of this matters for the heart of the video, which focuses blue LEDs in the GaN materials system which is definitely a direct bandgap material, however if someone does manage to create a manufacturable light emitter in pure Silicon expect an absolute revolution with regards to optical computing and photonics. (Not for lack of trying, this has been the holy grail for at least 20 years, possibly longer)
[1] https://en.wikipedia.org/wiki/Quasiparticle [2] https://www.chu.berkeley.edu/wp-content/uploads/2020/01/Chen... [3] https://www.iue.tuwien.ac.at/phd/wessner/node31.html
A cool chance to show the importance of determination!
I mean, I could explain how they worked in the same ways that they were explained to me, but I couldn't connect those explanations to a true physical understanding.
But thanks to this, I finally actually understand.
Also, the LED story was fascinating.
My search-fu is failing though. I did find this interview
https://www.jsap.or.jp/jsapi/Pdf/Number02/Interview.pdf
Deleted Comment
Dead Comment
It was also around that time that web-based communities of computer technicians really took off. Web forums, etc.
The coincidence led (yup!) to a love-at-first-sight relationship. Funny as it may seem now, being a mere 20 years later (or: "holy crap, it's been 20 years!, how did that happen??"), there were a few years there in the 2000-2005 region during which the de-riguer of computer nerdery was to go blue LED crazy.
It felt elite, cutting edge, rare, oh-so-techy. And it's funny now, to look back at the windowed PC cases full of LEDs and garishly lit by cold cathode tubes, with our mouses and speakers and other electronic gadgets painstakingly swapped over to blue.
And within a further couple of years - from about 2005 onward, if not sooner - the commercial market had taken over the trend and made it boring, passe. We hackers and overclockers weren't interested any more. Indeed, these ultra bright things began to get annoying. Within about 5 years the modding scene's blue LED craze began, peaked, commercialised, became a liability ... at which point we began to hastily obscure our blinding modifications with another, very different, product whose very identity hinges upon the colour blue : Blu-Tack! [0]
So where did it all end up, this short-lived cultural crossover between blue LEDs and hackers? Well, basically, the commercial market morphed into the "RGB" movement of lit-up computer hardware. RGB fans, RGB cases, RGB panels on graphics cards, etc. But I still think the blue LED is pretty cool.
[0] https://en.wikipedia.org/wiki/Blu_Tack
I’m glad those days are over.
It was a wall with small scattered lights in different colors, so they used recessed LEDs. Fine. But instead of color LEDs with a neutral diffuser, it had red+green+blue triple LEDs to make white light, with a red/green/blue plastic in front to recolor it!
I understand how this could be cheaper to assemble or maintain, but I'll never not balk at a system that has components undoing each other's work. Feels almost disrepectful to the technology.
https://youtu.be/PBFPJ3_6ZWs?si=sTeRrqQ5umHsNCgzhttps://youtu.be/cQgcTkXacAc?si=CDj0G9Sh7S-wbLjNhttps://youtu.be/va1rzP2xIx4?si=cAp65hnmwtkrXgDc
But anything blue always looks blurry unless I'm very close to it.
But the commenter said it was red, green, and blue LEDs together, with a blue diffuser over them. Depending on the diffuser, that could produce a more pleasant result (by allowing some monochromatic red and green through), but it presumably wouldn't solve the underlying problem that monochromatic blue light can be unpleasant.
https://images.squarespace-cdn.com/content/v1/595d3fb837c581...
Though it is incredibly frustrating how ungrateful Nichia Corp was to Mr. Nakamura, the underdog who pushed through every obstacle to ultimately give them the vehicle for more than 65% of their revenue!
People like the Nichia CEO at the time, a nephew who inherited the business nepotism-style (ditching the successful methodologies of his uncle) are just goddamn fools. Any success is in spite of their unimaginative, bean counting petty mindedness fighting tooth and nail every decimeter of the way.
"This weird researcher wouldn't follow orders, and I didn't dig deeper to understand his level of commitment or anything. If I had, I would've at least seen his level of dedication and possibly rallied to support him."
That's the best possible version, and highly unlikely since, as you noted, narcissism.
Anytime I see someone intelligently committed to a cause the way Nakamura was, I respect it and will support them however I can. Even if it doesn't pan out, it's still a good story.
This video is to adult me what Back to the Future was to kid me: it has it all.