I find it quite annoying that the original papers are not mentioned anywhere in the article. How difficult can it be to include a link/reference/doi to the source? I understand that for many people these scientific articles are too much, and unfortunately most of the times the scientific sources are still behind pay walls, but it is very important if one wants to verify the claims made in the news article.
Agreed. Fortunately there's a media contact details at the end of the article. I'm sure they'd like that feedback as well. Did you email them in addition to posting this comment on HN? :) I just did, expressing similar sentiments.
The advance is not that they "reached 90%", that's absolutely trivial, and nothing to do with a "Solar Power Material". They seem to be claiming that their advance is a black material they can paint onto thermal pipes that is durable at 1000K for an extended period of time in Earth's atmosphere, made out of blends of nanoparticles.
"High temperature black paint created that lasts 5-10 years instead of 1 year, reducing maintenance needs for concentrating solar thermal" would be a more honest title.
A direct comparison with off-the-shelf high-temp black coatings would have been useful. Here's a list of some known values for black coatings for solar collectors. (http://www.solarmirror.com/fom/fom-serve/cache/43.html)
Black chromium plating (absorption 0.87, emission 0.09) is often used for this purpose. It can handle the temperature, tolerates thermal expansion, and, like most hard chrome plating, is a hard plated surface that can be cleaned aggressively, as with pressure washing. There are other black coatings with up to 0.99 absorption, but they aren't as tough.
So what this new work has done is improve performance from 0.87 to 0.90.
(Why are so many "nanotechnology" articles like this? The title and lede sound like it's some earth-shaking development, and then it turns out it's at best a minor improvement.)
it's actually rather surprising, and important, that they were able to get 90 percent solar absorptivity combined with 30 percent infrared emissivity. this is difficult because absorptivity is the same thing as emissivity, but of course you can have different emissivities at different wavelengths; that's why things are different colors. but the vast majority of opaque solid materials have rather pastel colors, which is to say that their absorptivity (and thus emissivity) varies slowly with wavelength. this is undesirable for concentrating solar power, because it makes it hard to transfer the heat energy from the sunlight to the coolant rather than reradiating it as infrared. all the materials you named with albedos of under .1 are also great absorbers of infrared, which means they're also great emitters of infrared, so they wouldn't work very well for this application. (you also happened to pick three that won't withstand kilokelvin temperatures.)
there are a number of 'spectrally selective coatings' already in existence with this unusual combination of emissivities, but they are all very expensive to produce.
the longer lifetime is useful too, but not so useful they bothered to mention it in the abstract of their paper.
the small particles mentioned are made of a silicon-germanium blend, although they claim you can use different semiconductors for this. they picked that blend because its bandgap is about a volt, so it absorbs photons of more than about an electron volt.
i don't know! i didn't know about structural coloration, https://en.wikipedia.org/wiki/Structural_coloration. the crude absorption is not structural (it corresponds to a bandgap in the semiconductor), but clearly they think the structure is important. i didn't read the paper closely enough to know why.
Solar power is a bit of a misnomer here because most people reading that will assume it implies eventual conversion into electricity or motive power.
Electricity is like steak, heat energy is more like hamburger. It's useful but not nearly as useful as electricity, so you're going to need another conversion step (steam turbines are best at this right now) to get to a more usable form of power and that conversion step will have losses (radiation losses, mechanical losses, electrical losses).
So 'overall' efficiency is the key, not the efficiency of a single step in the process (they should at a minimum then list their current efficiency next to the previously achieved maximum for that step and how cost effective this new method is).
> The new material can also withstand temperatures greater than 700 degrees Celsius and survive many years outdoors in spite of exposure to air and humidity.
For simplicity, lets assume that it's 723°C (1333°F), and there is a nice summer afternoon and the environment temperature is 23°C (73°F). In Kelvin, we have 1000K and 300K.
By the second law of thermodynamics, we have that Q_in/T_in < Q_out/T_out,
so Q_in/1000K < Q_out/300K
then Q_out > 3/10 Q_in.
At least the 30% of the heat that enters has to be released to the environment in some type of cooler. So the efficiency of the heat to electricity conversion is at most 70%. If we multiply that by the 90% efficiency of the light to heat conversion, we get at most 63%. (65% in a cool winter morning :) ).
The Wikipedia article is not very clear ( http://en.wikipedia.org/wiki/Energy_conversion_efficiency ), but I guess that in a real power plant, we'd get at most a 40% or 60% heat to electricity conversion efficiency instead of the theoretical 70%.
It's not a noticeable improvement in efficiency for solar heating. Solar heating was already ~70% efficient at converting incident sunlight to heat for water heating [0]. So it's important to note that this efficiency includes losses for circulating the water within the storage system.
> As of 2008, the world record for solar to electric efficiency was set at 31.25% by SES dishes at the National Solar Thermal Test Facility (NSTTF).
The 90% * 60% number is probably very optimistic.
The Wikipedia article is also interesting, because it discuss that when the temperature is too high the heat receiver is an important heat emitter. I didn't thought about that.
15%-20% is for photovoltaic solar cells (http://en.wikipedia.org/wiki/Photovoltaics): where sunlight is converted directly into an electical current. The article here talks about converting solar irradiation into pure heat.
Yeah, but gus_massa calculated the upper bound of the sunlight->electric current conversion using the material we were talking about, which seems like an improvement to me. However I don't know what the ballpark efficiency of solar heat plants is, so it could turn out to be not that big of a deal if current modern plants achieve similar levels of efficiency.
Also, efficiency of solar power isn't even extremely relevant, except perhaps for very space-limited applications like solar vehicles. Since the input to solar power is effectively free, a better measure is dollars per watt.
Cost and sustainability swamp density for sure. Now if these things could be made without conflict materials or "artisanal miners" risking being buried in unsanctioned strip mining.
The heat does not have to be converted to electricity, 60% of a home's energy usage is in the form of heat. 42% for space heating and 18% for water heating.
I am deriving 33% of the energy used for heating our startup headquarters with a solar thermal system we rigged up out of cardboard and black aluminum foil. We are bootstrapping and the electric heat the building has is much more expensive per BTU than oil/gas.
Black aluminum foil will reach 156F in direct sunlight in December in Southeast Connecticut. I have read that black aluminum foil will convert 88% of light to heat. We have 50 square feet of these panels that spend every clear or partly cloudy day between 130-156F, warming the building through convection. On a clear day with an outside temperature of 27F our rig heats two floors to 61F. We are going to add another 50 sq. ft. of panels to attempt to reach 70F+
And the heat mass of the building keeps it warm for four hours after the sun sets. We have some large floor to ceiling windows and we installed the panels so they stand behind the bottom three feet of some of the south-southeast facing windows.
The total cost of our system was $94 and a couple hours of time.
If we owned the building I would install solar hot air panels (basically an insulated rectangular box with a plexiglass top and black aluminum foil tacked into the inside with an intake and exhaust) and run a couple air ducts into the building.
This building already had a solar hot water heating system, with two 4' x 8' water heating panels on the roof. This gives us much of our hot water.
We are awash in solar energy, and heating air with the sun is the cheapest and most efficient way of capturing it. Solar hot air is THE low hanging fruit of solar power.
The only downside with our system is that there is no energy storage. But that could be engineered easily. A concentrated solar trough connected to a buried large cement block or other heat sink could easily store a day or two worth of heat for night time and cloudy days.
55% of the days here are clear or partly cloudy and great for solar heating.
Update:
If you were wondering what we are doing in Southeast CT, we are participating in the first of what we are calling The Winter Startup Challenge, our own laid back, low burn rate version of Y Combinator.
My team rented a waterfront three bedroom house in a beach community for the offseason, October-May. Being the off season we got a great deal.
We have to leave May 31st. We have 8 months to develop the product and reach ramen profitability or get funding. The product challenges have been overcome, the prototype is being built, we have committed buyers waiting for the first production run. Five months left.
Can you share some pictures of your setup, and how it works? From your description, I think the aluminium foil is propped inside your windows, and absorbs energy from the sun. This then warms the room via convection.
What I don't get is why thin aluminium foil is good for this. Don't you need a bigger heat absorber? Or is it surface area that's important?
The aluminum foil is cheap, and I found some anodized foil that is really dark for great absorption. Anodized black foil is much better than painting foil black.
I don't believe the total BTUs would be any higher if one replaced the thin foil with a 1" sheet of aluminum. An aluminum sheet with fins on the backside for heat dissipation might help, but I still think the total heat would be the same because the amount of energy hitting the surface is the same.
It is surface area and temperature of that surface that counts and we don't need fans because of convection.
If you take an infrared thermometer and take a reading off a baseboard heater you will see 140-180F temps radiating off the casing and the metal fins inside that are put there to increase surface area.
These flat panel aluminum foil based collectors work the same way.
When light/heat energy enters a room through the window some of it turns to heat and some is reflected back out the window. That is why you can see objects from the other side.
Seeing those objects is how we lose our heat. These collectors let us capture and convert a much higher percentage of the energy coming into the window already.
If you look at one of these collectors through the window from the outside it appears invisible almost, they are hard to see, nothingness from which precious free energy is harnessed and consumed and not allowed to escape.
>60% of a home's energy usage is in the form of heat. 42% for space heating and 18% for water heating.
and 90% of all statistics on the Internet are made up. Unless you meant to say, "climate control" instead of space heating, this is just not true. You can't just take the "average" of what people use energy for in a country like the US, and expect it to be close to true for everyone.
For those of us living in the South, Southwest US (it was 65 and sunny here on Christmas day), energy usage is much higher for air-conditioning than for heating, and you can't just take a passive solar system and expect it to cool your house down.
Yes, passive solar heating can be great for some people, but not for everyone. Geography matters.
I just wanted to say thank you for that link; I had always dreamed of using the oppressive heat to create an A/C system and now I know it's a real thing [plus, the adage about "no new ideas" is again confirmed].
If you want to use heat in your home you're going to have to transport it there somehow. The only forms of energy that we can transport without huge losses are fossil fuels and electricity. The latter is obviously the more convenient one, though natural gas is also (relatively) safe and easy to transport.
Heat is extremely hard to transport over longer distances without huge loss.
Your focused on moving heat from "Power Plant" to "Home". That's not where/what the OP is talking about.
He's talking about installing solar devices to collect heat at the place of business/home and use it there. On-site creation (if you consider using sunlight creation) and consumption.
There is nothing stopping the creation of electricity/gas/etc and transporting that over long distances... while ALSO using solar devices at the business/home.
Why use electricity to heat water, when you can use the sun? Why run an electric heater when you can use the sun to provide some/all of the heat needed?
It's a great idea and he makes a lot of good points. If the engineering can be simplified and brought to the masses
You really should have a good look at the material and the devices described in the article. I don't think they lend themselves to decentralized use at all.
Solar concentrators are typically massive and are relatively dangerous to operate. Especially if you do two axis concentration (I should know, nearly set my old office on fire with one...).
The title explicitly mentions conversion into heat, so it's not fair to call it a misnomer. And turning heat into electricity is how most power plants work, not just concentrated solar power ones.
True, but I can picture many uses for the heat as is - such as off-grid heating water for homes/hotels/factories, which would normally burn natural gas or use electricity from the grid.
There are plenty of uses for a temperature difference that don't require arbitrary conversions to and from electricity. Furthermore, if an application desires electricity instead of a heat source, the Stirling cycle generators are tough to beat in terms of cost, reliability and end-to-end efficiency, even with exotic solar cells and lenses.
While what you say is totally true, huge part of the energy budget of humanity is heating stuff. If we remove heating for cooking, heating homes and water ... we will reduce drastically our footprint.
Also my inner steampunk geek will love the sight of huge sterling engines moving around ...
No ... you can never convert electricity to heat at > 100% efficiency.
What heatpumps do is transfer heat from one place to another while using electricity. And they have some nice limitations on temperature operating ranges.
in context, this is pedantry. using a heat pump, you can add four joules of heat to your house by dissipating one joule of electrical energy. that's four times as efficient as an electric space heater (bar fire) or lightbulb, which are very nearly '100 percent efficient.' in this sense, im3w1l is correct, even though the phrasing could confuse people into thinking that we are talking about perpetual motion machines.
Yeah ok, sorry for being a little imprecise. Anyway, the main point still stands, you can heat a house more with one Joule of electricity than one Joule of heat.
Converting light into heat is by far not as difficult as turning it into electricity. So don't mix this up with the 20-24% efficiency that are achievable in photovoltaics.
What's as useless as the pointed article. I tought the university PR would care to add such things as absorption rate of infrared, and why should I care about this instead of looking at traditional black nickel.
Any random black paint absorbs more than 90% of the light, and lots of black coatings survive 700°C.
thats why the story itself (which is a couple weeks old) and most discussion isn't catching the "real story".
You can find materials and substances that'll survive years at 700C although its non-trivial (not home depot, but no challenge for real engineering). And you can find materials and substances with ridiculous awesome IR profiles that are unfortunately incredible fragile or at least not "engineering useful" tough. But both characteristics in the same material is where the real story is. Its pretty cool.
There is another aspect of industrial scale use, not discussed in the weeks since the story came out, where for example, nickel oxide powder has some serious issues for lungs, and no one seems to know the toxicology of this new stuff. I donno where this stuff fits on the continuum of health, if its better or worse that your example of nickel oxides. It would suck if everything downwind turned into a superfund site, then again maybe its harmless as a charcoal bbq.
I guess a bad computing analogy would be its not very hard to find a low power computing device. Think of TI and microchip pic series of picowatt or nanoamp things. I've fooled around with the 10F220 series for no reason, "just because". And its not hard to find a computing device that does stuff really fast, any ole COTS desktop CPU or high end video card has interesting specs. But finding a processor that simultaneously uses nanowatts of power while making gigaflops of calculations, in 2014 anyway, would be pretty impressive. And getting a good ratio of flops/watt while being engineering-worthless isn't that impressive, but getting a good ratio of flops/watt while being engineering useful and deployable is really cool.
What was the previous best conversion rate? This doesn't get us all the way to electricity, so its not really interesting info without a point of reference.
Does this mean that we are closer to replacing fossil fuels? It seems as though for energy we could do it now if we wanted, yet we still obsess about fracking and to a lesser extent middle eastern oil? The cost difference is certainly not the reason we don't dive in and make these changes.
Those in charge still seem to be of a 'secure the oil and the heroin' and you rule the world.
Replacing fossil fuels is not as simple as having high-efficiency solar panels (which these may not be when converting sunlight to electricity instead of heat).
Fossil fuels also have the advantage of having extremely high energy density. They can be transported cheaply and efficiently and they store well. Electricity suffers large losses when being transported long distances directly, and batteries are significantly heavier per kWh. Batteries also degrade over time, are expensive to replace, and often require substances that can be as (or more) destructive to mine than oil is to burn.
It's about 80% efficient and would probably get more so. So you can solve the storage and intermittency problems of non fossil fuel storage pretty simply.
Pumped hydro has nowhere near the capacity necessary to consider it a complete solution to long-term energy storage[1].
Tesla has shown that batteries can be viable in some cases (e.g., passenger vehicles), but that is but a tiny fraction of the cases where the versatility, mobility, and energy density of fossil fuels are used. Virtually all other long-haul transport (passenger flight and cargo hauling for example) uses them exclusively for starters.
Nothing about transitioning away from fossil fuels is "simple", and it's likely to involve a patchwork of partial solutions[2].
Tesla has not proved that at all. You are completely ignoring the fact that they have no cheap solution to the batteries going bad. That's why there is no pricing available on battery replacements and why they have to have the depreciation matching program with gas cars.
Pumped storage is cool for daily cycles, but the real source of electricity storage is going to be bigass batteries that can respond to demand as quickly as a natural gas turbine: http://www.ambri.com/technology/
Hardly all the problems are solved. The holiday road trip I just took, from Missouri to Virginia and back, would be effectively impossible on the current Supercharger network, and would take much longer on the 2015 planned Supercharger network.
When there's a dense network of battery hot swap stations, then perhaps.
We could replace fossil fuels even today everywhere (with enough capital investment) ... except air travel - we are 3 or 4 major breakthroughs away from that.
With an energy source, you can manufacture fossil fuels from atmospheric carbon, giving you carbon-neutral fossil fuels.
The goal is to substantially reduce the CO2 being pumped into the atmosphere, not to just talk about how People Like Us are better than Those Fossil Fuel People, right?
The problem with renewables lies in unpredictability of wind power and photovoltaics, and in transmission infrastructure designed for predictable (and controllable) power sources.
We don't have good storage for electricity at the scale that is needed, not within few orders of magnitude, so at all times energy produced must be equal to energy used (the available storage is insignificant globaly, it is very expansive and is limited by geography - basicaly we can pump water upstream and let it flow back through turbines - this is only possible when you have a lot of water and some place much lower, where it can safely go, and it's still not cheap).
Additionaly each transmission line has maximum capacity and it will destruct itself if you try to send more.
With old-style energy sources we had baseload produced by water dams, coal, gas and nuclear plants, and peak load produced by plants that can be quickly (in less than 30 minutes) switched to produce more or less energy depending on demand - these are usually gas or coal powerplants. So with average demand for X, and peak deamd of Y you need X of stable powerplants and (Y-X) of controllable powerplants. Network is designed around that Y, using assumptions that power is divided into smaller and smaller lines from source to destination.
Only thanks to that, and to accurate predicting of demand by each level of power distribution hierarchy (I know about details of Polish system, but I guess it's similar everywhere) - the system works.
If you just swap 1000 MW of coal/gas/nuclear baseload plants with on-average 1000 MW of photovoltaics or wind turbines - you will literaly destroy your energy network. Half of the time it will produce too much energy (which results in blackouts and costly repairs), the rest of the time you will have blackouts because of not enough energy produced. You need to prepare infrastructure for that.
You need to adjust the whole network to bigger load, and to change the network topology from "connected stars" to peer-to-peer (or at least orders of magnitude more stars, connected orders of mganitude more intimately), and you need to expand the transmission lines a lot, because the only way to provide baseload with renewables is to average out production from a lot of places with different weather.
Right now electric networks aren't designed for that, and upgrading the whole infrastracture is going to be very expansive.
Germany is trying to do that - props to them, but in the meantime they are kinda problematic to their neighbors, because when they cannot deal with excess energy produced by renewables - they dump in on Polish, Czech,etc networks. Germany needs to pay for that (too much power is exactly as bad as not enough power - Polish and Czech controllable powerplants need to adjust production because of that and it costs both ways, also routing the power through the lines need to change because some lines may exceed capacity with that additional energy, and that cascades through the whole network. Energy distribution is organised in such a way, that when somebody predicted demand or production wrong - they pay for the rebalancing of network caused by that.
From the article (which I suspect you didn't read):
>CSP power plants create the steam needed to turn the turbine by using sunlight to heat molten salt. The molten salt can also be stored in thermal storage tanks overnight where it can continue to generate steam and electricity, 24 hours a day if desired, a significant advantage over photovoltaic systems that stop producing energy with the sunset.
Think how handy something like this would be in Saudi Arabia to run de-sal plants. As long as you have big enough drinking water tanks to handle the occasional weather related or maintenance related you're golden.
For example, if you have tanks big enough to hold a week or so of drinking water, then as long as you produce more than a weeks drinking water per week, you're all good.
This applies to almost any process that requires energy that is either a long term average or you can store in a tank. Electro-refining copper (good luck keeping aluminum cells warm, lol) or pumping irrigation water into fields. I would imagine you can solar distill ethanol whenever you feel like it, more or less. Given big enough tanks, you could generate chlorine for anything from swimming pool size to municipal water system size. A country needs X megatons of ammonium nitrate fertilizer per season and yes all of it is made today from natgas, but you could theoretically make it from solar power, air, and water, as long as the long term average power generated is high enough to make enough fertilizer per year.
Some industries like aluminum refining and some chemical plant work have a batch or shutdown time longer than the average weather related issue, so for basic chemistry scientific reasons they're screwed and will always require coal powered electricity, but most others theoretically only need a couple minutes of UPS power to shut down. Because electrical power is now super cheap, there are industries that have not put the slightest engineering effort on it... think CNC machinery, where there's no scientific reason a machine can't enter "sleep mode" within seconds, and return to work in seconds after sleeping, although in practice no one has ever made or shipped a machine like that, that I know of. As a simple example there are very few gcode programs that don't involve a dwell or toolchange where it could simply sleep in the middle of it, aside from the more difficult task of interrupting a cut. I guess this is an awesome industrial startup idea, not just for CNC machines but numerous other industrial machines. (And edited to add, there are numerous 3rd world areas overseas and California where the coal powered electric grid is totally unreliable, and machines that respond cooperatively to power interruptions would be quite handy even if there's no solar or wind generation involved.)
The biggest challenge to solar powering those industries is going to be non-renewable energy is very cheap compared to long term capital costs, so if its cheaper to run on diesel than to let it sit while continuing to paying the bank loan, or if your competitors can out produce you because they're not sunlight limited, then they'll have to run on diesel or coal-based electricity for financial department reasons, even if its no engineering challenge at all to run on intermittent solar power.
> Because electrical power is now super cheap, there are industries that have not put the slightest engineering effort on it... think CNC machinery, where there's no scientific reason a machine can't enter "sleep mode" within seconds, and return to work in seconds after sleeping, although in practice no one has ever made or shipped a machine like that, that I know of.
Big power consumers even now can enter "consumption reduction offer" market. They provide offers that they will reduce consumption by 1 MW between 11:00 and 13:00 at thursday, and when electric network needs rebalancing - distributor buys these offers from them (often at many times the regular electricity price) because it's cheaper to slightly reduce consumption here, than to pay someone else to power on a whole new powerplant block elsewhere and send the energy longer way around just because at one point in the network there's not enough transfer capacity for a few minutes.
There's a whole exchange market, allowing to publish these offers on consumers side, and buy these offers on distributor side, and regular traders doing brokering beetween them, even some automated trading systems, you can even buy futures based on that :). It is quite elegant system - using free market + some regulation to solve allocation and prediction problem.
This, with law support for smaller consumers (and serious automation, because consumers wuldn't want to spend their time trading energy) could make these changes (quick sleep/wake modes) profitable, even if the energy savings alone aren't enough to justify it.
In the end new washingmachine etc could be connected to internet, automaticaly publish such offers when you switch them to energysaving mode, and react when someone buys its offer.
I think EU wanted to expand these regulations to allow regular consumers to participate, but I don't know how advanced it is (I changed job and no longer work on energy trading systems).
We already have materials that can achieve 75% efficiency [1] and probably more(it's just a rough search), so the 90% figure is much less impressive . But they also claim low cost and low maintenance over other methods, which are highly valuable.
Checking out some of the Fresnel videos on Youtube really helps one appreciate the power of the Sun. If the new material can withstand 700 C then point a Fresnel at it and hook up the output to a Stirling engine for conversion to electricity.
.jpg){kind=link}
I believe the following articles are the ones on which this story is based: http://dx.doi.org/10.1016/j.nanoen.2014.06.016 http://dx.doi.org/10.1016/j.nanoen.2014.10.018 (behind a pay-wall unfortunately)