Just a heads up to those who haven't published in peer review journals: hyping the potential commercial applications of your research, no matter how mundane the discovery, is required.
Worse yet, the published work concerns simulations. Simulations are cool, and have great value, but they don't constitute discovery. Confirmation of a theoretical prediction by experimentalists constitutes a discovery.
Even if the effect is confirmed, the road to putting it into commercial use is long and most likely a dead end. I know first hand that "nanoscience" isn't so much science as it is an art. Tiny imperfections result in large changes to desired effects. This is due to the increased contribution of interfaces and surfaces relative to the bulk. The very same property that leads to novel physics is the one that defeats the potential for practical applications.
I long for the day when people can do good science just for the sake of good science, and not have to spin every single paper as being the start of some new revolution that never seems to come.
> I long for the day when people can do good science just for the sake of good science, and not have to spin every single paper as being the start of some new revolution that never seems to come.
I absolutely agree - but I think that's only going to happen when people's livlihoods are not dependent on the immediate value of their research.
J. J. Thompson (who discovered the electron) said this about 100 years ago:
If you pay a man a salary for doing research, he and you will want to have something to point to at the end of the year to show that the money has not been wasted.
In promising work of the highest class, however, results do no come in this regular fashion, in fact years may pass without any tangible result being obtained, and the position of the paid worker would be very embarrassing and he would naturally take to work on a lower, or at any rate a different plane where he could be sure of getting year by year tangible results which would justify his salary.
The position is this: You want one kind of research, but, if you pay a man to do it, it will drive him to research of a different kind. The only thing to do is to pay him for doing something else and give him enough leisure to do research for the love of it.
Our research group prided ourselves on cutting edge nanotechnology that was also commercially viable.
Just focusing on my project alone, and not the other great ones, if I explained it in the most concise, layperson fashion it was as follows:
You like sugar cookies? Well, guess what, you mix chocolate chips in and BAM 10x better! All you have to do is mix them in when you're mixing the batter the way you already do, no added steps or infrastructure.
My research project was discovering and demonstrating that LED efficiency can be increased by mixing polymer-grafted zirconia nanoparticles in with the LED encapsulant and the encapsulant and phosphor for phosphor concerted white LEDs (just look at the back of your phone, that flash, of yellow, is likely a YAG:Ce phosphor on a blue or royal blue LED, the mixing of the yellow and blue gives you that white).
Despite being humble about mentioning our success, this result is actually pretty important. The US Department of Energy's Solid State Lighting goals include getting even small increases in efficiency through better light extraction (more light out instead of reflecting back in), better thermal conductivity (more heat out than in, heat from the semiconductor and phosphor decrease both of their efficiency), and better lifetime.
We worked with the same companies that manufacture the LEDs, phosphors, and the silicone encapsulant materials so that we were using the same thing companies were using in their manufacturing process for personal and commercial applications. I chose to do a phosphor conformal coating the same way they do in industry (other researchers try crazy versions of phosphor placement to try to get more efficiency but it makes tooling significantly more comprehensive, complex, and you don't get much if any increase in efficiency. You actually get bigger LED structures, so they're more of fun theoretical models of "what you can do").
More briefly: my research set-up was what they do with the materials they use in industry for manufacturing the LEDs we use.
I showed that increase in efficiency by just mixing the nanoparticles into the encapsulant. We did not notice a change in a 1000 hour reliability testing. Other solutions like epoxy encapsulants or Titania nanoparticles have issues like yellowing or other degradation.
Other key things:
It was not insanely expensive to do this, while I never quantified the cost per LED I can assure you it was low (we allocated $10,000 for materials and by the end of the project I was told I needed to burn through the last $6,000 somehow so that funders wouldn't give us less next time. I'm frugal, spending where we need but not frivolously, but our materials weren't super expensive).
The entire production of nanoparticles was scalable (blew my mind, often nanoparticles synthesis and polymer grafting like this can have a drop off in yield as the process scales). Whether you we're making 2.22g yielding 0.7g nanoparticles or 22.2g yielding 7g, (I think that was the conversion, it's been a while) you got the conversion of precursor to nanoparticles to grafted nanoparticles.
All companies have to do is mix it in using the same materials they have been, and we get increased LED efficiency.
Even with it so easy and proven, the road to commercialization is still slow.
One of my peers continued on to a company to help push this technology into application but even such a proven, simple thing is challenging for large industries to pick up.
Unfortunately, our research group was rare in this regard. So many others focus on ivory tower "cool" discoveries, but leave a major gap to commercialization. You can spin your research into a startup company, partner with the corporations that will use it, or try to pitch it, but even a proven enhancement will require luck and a long road.
I used to work in an LED startup for 1 year as my first real job in Canada, then it went under. Also in a fancy phosphor niche, with some "smart software" sugar on top.
$7m capital - eaten like a candy.
Any conventional high tech startup in the West is crazy expensive outright from the beginning, unlike internet and software companies.
Close to a million went outright to high profile C-levels even before they did anything. The biggest sum I've seen paid to anybody to just to hang around.
A million, to "a real deal" science team of photonics PhDs and engineers with 20+ years experience.
Few millions for all kinds of super exotic engineering services, like doing thermal simulation of an LED package, making a custom process metrology tool for patterned LED chips, custom process control tools.
When I asked one of engineers about the price, they told me that they used 10 years old, few times resold tools to save money, and that by standards of the industry this is something that people don't even blink about.
The rest went into initial batch, and then the biggest financial backer backed away, and that was it, despite the first batch being more or less sold.
I still can't fathom the mind of people who can wrap up a $7m enterprise on a whim while being 5 minutes away from the finish line.
And it still feels surreal to me how I was in the middle of it for a year while basically being just an office errands boy.
I'm no tax expert but I'm under the impression it's never advantageous to deliberately enter a business arrangement so that you can incur a loss for the tax offset. But, it can be advantageous to strategically realize losses that were previously paper-only.
Great content. I've been doing freelance machine learning work for a while, and it's hit or miss. Some clients are great, others are a complete pain.
Most recent gig I had, the client cut hours and then rates... all due to the fact that the CEO didn't manage the project properly from the outset, and they were hemorrhaging cash. That fell squarely into the "not my problem" category, so I quit.
It's a tough business, and one I've not completely figured out yet.
I have been a freelancer for 10 years now, and I noticed that machine learning gigs are a bit harder to find because they is usually proprietary info and they want to keep the "models" in house.
Gigs seem to fall into a couple categories: one group of clients has heard all the buzz and wants to get in on the action, without really knowing much about it.
The more serious types are indeed protective of their data, intellectual property, and processes. I have even seen paranoia around leveraging open source frameworks (Tensorflow, Pytorch) out of fear of their sponsor corporations coming for the client... not sure I see the logic on that one, but whatever.
Whenever I don't have much success in a venture, I try to look inward to see what I'm doing wrong. I suspect that there is a fair amount I have to learn about getting good gigs and being successful in the freelance game. Unfortunately, it's one of those "have to learn the hard way" type of things.
Very cool stuff. To what extent would you recommend people study the underlying physics of QM, vs. the more domain specific content of quantum computing?
I got my PhD in physics back in 2012, after doing my dissertation in spin dependent transport phenomena in magnetic materials.
got a job at Intel as a back end process engineer, which was pretty cool, but I was then laid off in 2015.
These days I'm all about machine learning. Teaching myself by creating content, and hopefully educating others. It's been really cool seeing the overlap in some concepts (systems seeking minimum energy vs. gradient descent) between ML and physics, and I'm hoping my background will pay dividends as I get deeper into the field.
Worse yet, the published work concerns simulations. Simulations are cool, and have great value, but they don't constitute discovery. Confirmation of a theoretical prediction by experimentalists constitutes a discovery.
Even if the effect is confirmed, the road to putting it into commercial use is long and most likely a dead end. I know first hand that "nanoscience" isn't so much science as it is an art. Tiny imperfections result in large changes to desired effects. This is due to the increased contribution of interfaces and surfaces relative to the bulk. The very same property that leads to novel physics is the one that defeats the potential for practical applications.
I long for the day when people can do good science just for the sake of good science, and not have to spin every single paper as being the start of some new revolution that never seems to come.