Yes, early days and all that, but...
An effect that requires temps below -73C seems not so useful in practice.
US boffins have documented a transition from metal to semiconductor that can be controlled by exposure to light. The researchers, from Washington University in St Louis, created a thin film of gold nanorods coated with zinc oxide. The result: with no illumination, the gold/zinc oxide mix conducted electricity as a metal, but …
An effect that requires temps below -73C seems not so useful in practice.
Most materials science begins in incredibly impractical environments. Microgravity (space) situations, particle accelerators, temperature extremes, etc... For example, the science underlying modern automotive safety glass was developed in experiments onboard the Space Shuttle.
Once you have observed an effect, regardless of its practical environment, and learned how to manipulate it there, you can begin to experiment with it in practical situations. Starting the research in a practical environment isn't efficient as there are too many unknown variables. Sometimes an effect can't be reproduced in practical environments, sometimes it can; but that's the nature of science and the reason for experiments.
Does that really matter? It sounded to me that at any temperatures normally found on the surface of the earth, the material remains a metal until illuminated, at which point it becomes a semiconductor. Turn the light off and it is back to a metal. It also becomes a semiconductor when it gets really, really cold, but that is an effect that is not likely to be used in most practical applications.
Correct. The light/dark induced metal/semiconductor transition occurs at room temperature. The metal/semiconductor transition will only occur in the absence of light at the cold temperatures.
Actually, I think the effect is the other way. At all temperatures above -73C the material exhibited semiconductor effects irrespective of illumination. At temperatures BELOW -73C it behaves as a metal, unless illuminated. When illuminated it behaved as a semi-conductor.
The graphs show the region from -73C to be remarkably similar, where as the difference is the area below 200K (actually, it looks move likely to be below 180K or so).
Also, the scale is different in the two graphs which further confuses things. The linked article does not clear things up.
No. You've all missed the point of the graphs. The left-hand scale is resistance.
The metal section has resistance on a scale of 10^7
The semiconductor has units of 10^5. That's a factor of 100, and since it's resistance, you'll see a massive spike in current. If they were on the same scale, you could not see the fine detail.
That's why this whole thing is important
" For example, the science underlying modern automotive safety glass was developed in experiments onboard the Space Shuttle."
Excuse my ignorance, but what's different about "modern automotive safety glass" compared to the thin, laminated, toughened glass that's been in use for many decades, and was in widespread use before the space shuttle was built?
since Register readers can stand having the detail
Hurrah for a media outlet that doesn't pander to the lowest common denominator of a red top reader.
(For the non brits, a "red top" is a tabloid newspaper written to appeal to the ignorant masses. Someone once told me that journalists for red-tops have to assume their readers have the mental skills of a seven-year old)
>...readers have the mental skills of a seven-year old bacon sandwich
There, I fixed it for you...
That's not something you want to think about at lunch time.
You mean a red top like El Reg has! Oh Christ! For ~10 years I've been partaking of information for the proles? Oh god! How will I get the stench off? I'll have to destroy all my computers. I guess the data center has to go to. Arrrrgh!!!!!!
That would have been as a safety device for pyrotechnic systems. Illumination would have been needed to give a low resistance path to fire the bolt.
Otherwise I'm not really seeing the sizzle here. There are things called "Light Dependent Resistors" that change their resistance by something like a factor of 10 000 or 1000 000 on illumination using CdS. They've existed since at least the 1920's.
Sorry folks but I'm obviously missing something as this take a very roundabout route to duplicate something you can already get off the shelf. Perhaps it demonstrates something clever but I'm not sure what it is.
>...Perhaps it demonstrates something clever but I'm not sure what it is.
The best presentations in the field of research usuall end with statements like this.
Of course, the marketing department will be along shortly and make it more 'sellable', but until then we could just continue to classify this as 'interesting...'
The thing is, how quick does the switch happen? If it's somewhere in the multi-gigahertz or higher range, you have a 10nm electronic switch that's rather speedy, no?
Interesting question. We could call it a "light switch".
When an unusual response to stimulus like this occurs, it isn't always obvious how to exploit it, but 'twas doubtless the same is early semiconductor work, early superconductor work, and certainly early bucky ball/nanotube work. At school I was taught about graphite and diamond and that's it. The other forms did not exist as far as we knew. Now look, were starting to find applications we never imagined before.
Discover a previously undocumented interaction, optimise it, play with it, make it repeatable, then leave it for the engineers to apply it in the world.
I will play the ignoramus and ask what usages can be forseen for which which we do not already have a solution. We already have light sensitive switches etc
I belong to the unwashed masses please enlighten me... ( I lack imagination too)
I too wondered how this may be more useful than a conventional photodiode / phototransistor. One thought is that semiconductors all suffer a voltage drop across the junction (0.6v for Silicon). This may not seem much, but for high power applications this can be a significant problem. When in its metal state I assume there is no such drop, so could be useful in that respect.
Annother idea is to protect solar arrays. Solar panels react rather badly to having current shoved up them the wrong way (such as when part of an array is in shadow and taking current from the illuminated panels). If the unilluminated panels can be made to "turn metal" when in shadow, and hence act like a shunt, they can be protected.
When the laser was first demonstrated it was for quite a while known as "The solution in search of a problem" and was regarded as "pretty but of little commercial use", how prescient they were as the laser is now nothing more than a laboratory curiosity showing once again what a waste of money fundamental research always is.
Look seriously you have to do this nebulous research and probably in most cases it gets nowhere but the failures and dead-ends are worth doing for the occasions where it succeeds spectacularly and unexpectedly as it is impossible to predict in advance what will be useless, slightly useful or game changing.
Or to bring it round to IT, there was a time when no one but a few crazy researchers saw a use for ultra high density storage or low latency error checking memory.
Having cut my teeth on chip design assisting with a silicon chip that functioned perfectly until you put a lid on it and shut the light out I dont think this is new - just in a different material. Its interesting science but I cant see a use for it that is not already available and probably cheaper.