coating cools solar cells to boost efficiency

http://phys.org/news/2015-09-transparent-coating-cools-solar-cells.html

In this case, the purpose of the design feature that reflects the radiation the solar cell doesn't use is to boost its efficiency by reducing how hot it gets in the sun. But I can think of another reason to do that:

I assume no solar panel can ever be exactly 100% energy efficient because that would break the laws of thermodynamics.
But, I have long thought that the 'ideally' solar panel should only absorb the solar radiation that it manages to convert into electricity while reflecting near-enough ~100% of what it fails to convert into electricity upwards where most of it would be radiated to outer space. That way, whatever solar energy the solar energy it fails to convert into useful energy would tend to be reflected upwards and thus contribute towards the albedo of the Earth and thus help counteract global warming.
I personally think one of our long term objectives should be to design solar panels optimized to do that.

I think even at the cost of the solar panel only absorbing, say, half of the solar energy and reflects the other half, I would still imagine it to be easily possible to make it cost effective and, so to minimize global warming, even at that cost, it would be worth so designing it to minimize the proportion of solar radiation it absorbs that it wastes as useless thermal radiation. I would guess it would be better to have a solar panel that converts ~50% of the solar energy into electricity and reflects ~50% of the remaining solar energy than have a solar panel that converts ~90% of the solar energy into electricity but absorbs and thus converts the remaining ~10% of the solar energy into waste heat that contributes to global warming. Well, that is what I think. I don't know if anyone here would disagree.

Originally posted by humy
Well, that is what I think. I don't know if anyone here would disagree.

It all depends on cost. It may be cheaper and more effective just to paint the rest of your roof white.

The land area covered by solar panels is probably never going to be significant when it comes to global warming, and the savings in fossil fuels that the solar panels gain by generating electricity is probably a far far greater effect.

Originally posted by twhitehead
It all depends on cost. It may be cheaper and more effective just to paint the rest of your roof white.

The land area covered by solar panels is probably never going to be significant when it comes to global warming, and the savings in fossil fuels that the solar panels gain by generating electricity is probably a far far greater effect.

I accept I could be wrong but I envisage a time in the far future where any amount of energy generated would be put to good use (what is not used for other purposes could always be used to run billions of research labs run by AI and robots so no human labor costs and all these labs could be designed built by billions of AIs and robots in billions of factories -all powered by stupendous amounts of renewable energy. After millions of years of construction, the labs would end up as one continuous lab covering most of the surface of the Earth in the form of a continuous layer several kilometers thick, with it being floating like a permanently anchored ship over most of the worlds oceans. And all our cities, towns, roads, parks, conservation areas etc were we live will be built on top of it so we will all be living on top of a vast global research lab in disguise for ever doing stupendous amounts of research to advance science. Pure fantasy? ) no matter how massive that amount is and where every rooftop, road and footpath will be completely covered in solar panels (the surface of the solar panels on all the roads and paths being covered in a wear-resistant transparent surface, such as hexadecimal diamond film ) . I think if that were to actually happen, I think the land area covered by solar panels and the proportion of solar radiation they reflect is going to have a significant effect on the global heat budget. Remember, the proportion of solar energy that they absorb is going to near enough all end up as waste heat (excluding the tiny proportion that would be converted to radio waves and other radiation that is shone directly out of the atmosphere ) that heats the atmosphere, even that proportion that is first converted to electricity. That is because, no matter how efficiently that electrical energy is used, it generally all ends up as waste heat.

It is possible we may eventually make fusion energy viable, in which case solar may no longer be the best option.
Also, if you are going solar on a massive scale then it is better to go into space. Stick the AI research centre on the moon.
Although it is entirely possible that powerful AI will not require all that much power.

Originally posted by twhitehead
It is possible we may eventually make fusion energy viable, in which case solar may no longer be the best option.
Also, if you are going solar on a massive scale then it is better to go into space. Stick the AI research centre on the moon.
Although it is entirely possible that powerful AI will not require all that much power.

Again, I admit I could be wrong, but I am assuming here that the philosophy that would be adopted is, the more AIs and labs, the faster the research can be done (because you can have more of them simultaneously doing a greater number of more specialized lines of research and in an coordinated way ) and therefore the better. So AIs and labs would be put over nearly the whole surface of the Earth AND the moon AND Mars AND Mercury etc and anywhere else they could be put with reasonable practicality. I have even speculated that vast floating labs may be manufactured on the surface of the oceans of the gas giants using the heavier chemical elements in the oceans and atmospheres there providing a practical way is found to do just that despite the vary harsh conditions there. The types of research done there would be severely limited by the stupendously high pressures. I think that, for example, you can just about rule out doing particle physics there. Perhaps the labs there can just highly specialize on extremely high-pressure physics/chemistry there?

And, no mater how energy efficient the AIs and the labs, one of the main limiting factors that would limit the number and size of them, apart from the availability of building material, would be the availability of energy because, with all else equal, the more energy generated, the more of them you can run simultaneously. And, while I don't have the slightest doubt now that fission power can be made viable, I am pretty sure (because I once did a crude mathematical estimate of this albeit with some simplistic assumptions ) that, for example, the practical maximum amount of solar energy that can be generated on Earth each day would dwarf the practical maximum amount of fission energy that can be generated on Earth each day.

But, that said, I think nuclear fission could become the main source of power where there is a significant lack of availability of solar energy such as in the outer solar system.

Originally posted by humy
And, while I don't have the slightest doubt now that fission power can be made viable, I am pretty sure (because I once did a crude mathematical estimate of this albeit with some simplistic assumptions ) that, for example, the practical maximum amount of solar energy that can be generated on Earth each day would dwarf the practical maximum amount of fission energy that can be generated on Earth each day.

I think geothermal energy should not be overlooked it is a vast largely untapped resource.

Originally posted by twhitehead
I think geothermal energy should not be overlooked it is a vast largely untapped resource.

It is a vast resource but I wonder of the consequences of tapping that energy, effects on earthquakes and such, if you extract heat you would make the rocks cooler and stick in place which might build up crustal pressure maybe causing volcano activity or some such. I would expect we could still get quite a bit of energy out before that happens though.

Originally posted by sonhouse
It is a vast resource but I wonder of the consequences of tapping that energy, effects on earthquakes and such, if you extract heat you would make the rocks cooler and stick in place which might build up crustal pressure maybe causing volcano activity or some such. I would expect we could still get quite a bit of energy out before that happens though.

It may even stop volcanoes or earthquakes. However, as you say, it would probably take an awful lot for us to have any significant effect.