Myhrvold on Solar: Blue is a Kind of Black
Superfreakonomics contains a number of significant misleading claims about climate change and clean energy, but the one I found most shocking was the contention that solar panels actually make the world warmer because they’re black. Solar panels are not black. They’re usually blue. This is an easily verifiable fact. This is a photo of a company I visited in Dresden where they manufacture solar panels. Their office is covered in solar panels. Blue solar panels:
They wouldn’t let us take pictures inside their factory so as to discourage industrial espionage by the Chinese (or so they said), but I can assure you that what they were manufacturing was blue solar panels. But if you look at their website you can clearly see that they’re blue. During the tour someone even asked why the solar panels are blue. We were told that you can make them any color if there’s some particular desire for funny-colored ones, but they determined that this particular shade of blue is the most efficient one to use. And that’s why most solar panels are blue.
So two quick takeaway lessons from that chat. One is that solar panels are usually blue. The other is that contra Levitt, Dubner, and Nathan Myhrvold the guys who build solar panels aren’t idiots who’ve never considered the fact that different colored material has different light-absorption properties.
Remarkably, however, Levitt and Dubner choose not to simply admit that quoting Myhrvold as saying that solar panels are black was a sloppy error that they’ll correct in the future. Instead, they had him write a post on their blog in which he digs in his heels on the black point, insisting (really) that blue solar panels are in some sense really black so his statement that “[t]he problem with solar cells is that they’re black” was accurate even though it’s not, in fact, true that solar cells are black. Then as Nicholas Weaver points out he adds new errors:
He compares the cost of running a coal plant with the cost of building a solar plant, neglecting that we need to construct vastly more power plants to both meet growing demand and to deal with end-of-life on old, inefficient plants. Even then, the break even point is less than 3 years, by his inflate-the-cost of solar figure!
This is really insane. The obvious problem with solar power is that it doesn’t work when the sun isn’t shining. Thus barring some really miraculous developments in energy storage, we’re going to need a lot of non-solar power. But that still leaves plenty of room for the deployment of solar panels, especially in places that tend to be sunny. Germany uses a lot more solar power than we do, despite being a very non-sunny country, and nonetheless manages to exist as an advanced industrial society. There are limits to what can be realistically done in this regard, but we’re not currently pushing up against them. There’s no reason for this to be controversial and certainly no reason for people to be making ludicrous claims about the color of objects.
98 Responses to “Myhrvold on Solar: Blue is a Kind of Black”
Register
RSSLogin
Topic Cloud
ArchivesBlog RollAbout Matt YglesiasContact Me
Donate
October 21st, 2009 at 10:48 am
“They wouldn’t let us take pictures inside their factory so as to discourage industrial espionage by the Chinese (or so they said), but I can assure you that what they were manufacturing was blue solar panels.”
You are right implying they left out something. The part about how they were also frigthened about industry espionage by the Americans mainly.
October 21st, 2009 at 10:54 am
The obvious problem with solar power is that it doesn’t work when the sun isn’t shining.
True, but it’s not much of a problem. As a society, we consume a lot more energy during the day than at night. If at some point, we’re generating more solar power than what’s necessary to take care of the day/night differential in power consumption, then and only then will that amount to a genuine problem.
October 21st, 2009 at 10:57 am
Also, the point you made yesterday about the ability of a MARKET SOLUTION to determine the most efficient abatement procedure would seem relevant once again.
October 21st, 2009 at 11:03 am
The obvious problem with solar power is that it doesn’t work when the sun isn’t shining.
Let me add that, to the extent that it’s a problem now, it’s a problem because we make it so. A homeowner might want to put solar panels on her roof, but if she works during the day, she won’t be home to use the electricity her solar cells are generating. Instead, she’s probably at an office building several miles away which keeps the lights on in her office by buying power generated at a conventional power plant.
If utilities were generally required to buy such home-generated power at market rates, the homeowner would have more incentive to install home solar panels, because they would then only be paying for the net amount of electricity they obtained from the grid – and might actually get paid by the power company if they used less electricity, on average, than their solar panels generated.
And the power company would have less need to build power plants to provide for peak daytime capacity.
But first, power companies would have to buy the power you generated.
October 21st, 2009 at 11:04 am
The venality of the Freakonomics people is really starting to show in how they’re responding to their errors. It’s pathetic, really, but probably also no worse than you’d expect from any media type who’s even remotely well known.
October 21st, 2009 at 11:05 am
This is starting to look more like a defense of Brand Freakanomics and less about a defense of the quality of the social science. Freakanomics gets you invited to Aspen and on Charlie Rose as a hip intellectual. Being right has nothing to do with that. Brand Freakanomics is beginning to look to good social science what Tylenol looked to drug safety in 1982 except Tylenol wasn’t to blame for the tampering that killed 7 people and Tylenol fixed the problem anyway and Levitt and Dubner ARE to blame and won’t admit there is a problem.
October 21st, 2009 at 11:09 am
The claim that solar panels are effectively black is close to correct. The blacker the better. They work by absorbing light and converting it to useful electrical energy. The more they absorb, the more efficient they are. When they capture photons above the energy limit, they generate more excess heat. When they radiate excess thermal energy, they are esentially radiating as a blackbody, with only tiny deviations. It isn’t a big deal. It is miniscule compared to the excess heat generated by fossil fuel or nuclear plants.
Most solar panels are actually not very efficient, and do reflect some energy. That’s a bad thing, not a good thing.
There are solar cells that have very high efficiency – graded Gallium-indium nitride cells that convert the entire visible spectrum to electrons, but they’re in the research stage last I heard.
The real problem with solar is that it is monumentally expensive – so expensive that it is a waste of money that could be spent on other ways to reduce atmospheric carbon. No consumer power should be generated with solar. The Germans are wasting money that could be put to good use on some other clean power generation, or more research into solar.
October 21st, 2009 at 11:12 am
Matt says:
“Solar panels are not black. They’re usually blue. This is an easily verifiable fact.”
Cool Matt, cite your proof that most solar panels in the world are blue.
The fact that a single German company finds them most efficient is not proof of your assertion.
Google “black solar panel” and you get 25% more hits than “blue solar panel”.
The Chinese make ‘em black.
Lot’s of people it seems make ‘em black.
The silicon crystals themselves are grey to black depending on purity.
They are NEVER blue.
So once again please cite your “easily verifiable” proof that most solar panels in the world are black.
Or have you once again lied?
October 21st, 2009 at 11:12 am
Matt you should start your posts on this subject “DON’T BUY FREAKANOMICS IT’S A RIDICULOUS BOOK” Lest their be any doubts.
Or read it the way you would read “the Da Vinci code”.
October 21st, 2009 at 11:17 am
The problem with solar is that we don’t use enough of it to supplement fossil fuel-based sources. What I found retarded in the piece was the implication that solar solutions involve massive central plants. The other thing I found retarded was the inefficiency/cost benefit conundrum in which the nay-sayers routinely fail to factor in the true cost of fossil fuels.
Jumping Jimenez, let’s follow the German model and get going with it.
October 21st, 2009 at 11:21 am
Correction:
So once again please cite your “easily verifiable” proof that most solar panels in the world are blue.
October 21st, 2009 at 11:22 am
Hate to step into another idiotic American renewable energy discussion – talk about your long running quagmires – but Germany doesn’t simply manufacture and install PV cells, but also is quite advanced in solar water heating systems.
Not that I expect a former Microsoft genius executive of the IP troll variety to actually know much about the real world, especially that concerning an advanced manufacturing economy, but he would have just the hint of an escape if he starts talking about solar water heating – those collectors are as black as possible.
Of course, those collectors also replace burning fossil fuels to heat water at an efficiency level considerably higher than ca. 12%, making his general point as idiotic as most discussion coming from the upper levels of the American business echelon.
October 21st, 2009 at 11:23 am
I read the NYTimes article. Everything he says is accurate. Matt, your knowledge of economics, finance, politics and foreign policy dwarfs your knowledge of science.
October 21st, 2009 at 11:23 am
Like most of your commenters thus far, I really don’t know much about this issue, but your position in this post is juvenile at best.
Harvard, philosophy. Generally a bright, thoughtful guy. You might want to try again.
October 21st, 2009 at 11:30 am
I’m moderately skeptical about photovoltaic power other than on a micro-level (ie individual houses/businesses in regions with the right climate). But solar thermal power seems pretty promising for large scale power production, and guess what? The reflectors used to concentrate the solar power are, well, highly reflective.
October 21st, 2009 at 11:30 am
This is ridiculous. It doesn’t matter what color they appear to us. What matters is the mount of light absorbed. If they absorb 95% of light evenly across the spectrum they look black, if they absorb 95% of light with lower absorption toward the shorter wavelengths, they’ll look blue. It doesn’t mean a damn thing. If they emit the majority of their radiation according to blackbody statistics, they are, for all practical climatic purposes, black.
October 21st, 2009 at 11:32 am
I hate to agree with Njorl (and the feeling is probably mutual), but, really, Matthew is just displaying his ignorance here. He ought to stick to topics where he has vastly more knowledge – like finance.
Black is not a color, it is a statement about the amount of light reflected from an object. Blue, on the other hand, is a color – which can be measured by the wavelength of reflected light.
Properly put, solar panels are both black and blue. That is, they reflect little light (thus, “black”), but the light they do reflect is more in the blue wavelength (thus, “blue”).
However, for purposes of Myrhvold’s discussion, only the black part really matters – the blue part is almost completely irrelevant.
October 21st, 2009 at 11:32 am
I have a factual question. I, like many others, had always thought that solar power did not contribute to climate change at all b/c it doesn’t generate any greenhouse gasses. Now I’m hearing, for the first time, that this is not true, because the color of solar panels (whether black or blue) causes something or other to happen that also contributes to climate chage. Bottom line: is that true or not? If true, how seriously does that reduce solar’s promise as a non-climate-change-contributing energy source?
October 21st, 2009 at 11:34 am
It’s like How much blacker could these solar panels be? And the answer is None. None more black.
October 21st, 2009 at 11:37 am
You know, if solar panels were 100% efficient, they’d be black. All visible light would be absorbed and converted to electrical energy.
The converse is not true; i.e., black is not necessarily efficient. If they were black like asphalt, they’d absorb visible light and re-radiate infrared. That would be bad.
But being black by itself is not always bad. And, in any case, I doubt that there will be an order of magnitude of more solar panel area than there is parking lot asphalt, so who cares about re-radiating solar panels?
October 21st, 2009 at 11:38 am
I usually don’t post in the comments, but this post brought me here. I find that my points have already been made, and probably better than I would have made them. Thanks Njorl and Al.
October 21st, 2009 at 11:38 am
Also, the storage of power with Solar Thermal is rather non-magical. its the future.
October 21st, 2009 at 11:41 am
They reflect power onto an absorber, so the heat imparted to the earth from the light captured is greater than it would have been, but that’s good, not bad.
All energy use heats the planet, except hydro and wind. You burn coal, the planet gets hotter. You absorb light for solar rather than letting it get reflected back into space, you heat the planet. However, the waste heat generated is a small problem compared to the alteration of the optical properties of the atmosphere caused by increasing CO2.
October 21st, 2009 at 11:42 am
BTW, I’ll add that Nicholas Weaver’s comment (”He compares the cost of running a coal plant with the cost of building a solar plant”) is completely false. Myhrvold compares that cost (in CO2 terms) of building a solar plant with the cost of building a coal plant. Period. He goes on to say that, despite that fact that it costs more to build the solar plant up front, that will recovered over time running the plants (about 3 years).
I don’t know why the anti-Freakonomics people get saddled with these morons like Nicholas Weaver, but there it is.
October 21st, 2009 at 11:42 am
Let me get this straight — people are criticizing solar panels because they are have a low albedo? Isn’t that kind of the point (as others have mentioned)? For light to be absorbed and converted to energy?
The problem with low albedo surfaces is that, instead of reflecting light, they absorb light and convert the light to heat energy. But if solar panels are truly efficient, then they are converting light into energy other than heat, so they aren’t really functioning as low albedo objects as far as the global warming concern about such objects are. So what’s the big deal about what color they are if they are?
Of course, solar panels don’t convert all absorbed light into electrical energy but rather convert some into heat. So they might not be as good at keeping things cool as whitewash. But then, they offset other heat producers, so maybe there is still a net gain (as others above have pointed out)?
BTW, I don’t see the weakness as being one of freakenomics in particular — there are plenty of other brands of economists who regularly succumb to the same sorts of fallacies.
October 21st, 2009 at 11:50 am
+1 to Njorl and Al. This blue/black controversy is painful to watch if you’ve got any physics background at all. Matt needs to crack a book and figure this out, because he’s a good advocate for this sort of thing and he’s damaging his own credibility.
October 21st, 2009 at 11:50 am
Njorl and others… of course in discussing Myrhvold’s point 1) he is correct and solar panels are “black” and 2) their perceived color is irrelevant to the larger point.
However Matt has a habit of making factual claims without the factual part to back them up.
And to add stupidity to foolishness he deploys these “facts”
to refute those he accuses of lying.
In that respect only it is important that Matt be able to back up his “easily verifiable” factoid.
October 21st, 2009 at 11:53 am
READ NJORL’S COMMENTS! EVERYONE! Apparently s/he is the only person who knows anything about science in thread.
Matt: This pretty poor post on the substance.
October 21st, 2009 at 11:54 am
“Thus barring some really miraculous developments in energy storage, we’re going to need a lot of non-solar power.”
Matsushita’s home fuel cells are supposed to go on sale in Japan this year. (They may already be available.)
Other manufacturers are expected to follow, and tax incentives are I believe already available in California for them.
October 21st, 2009 at 11:55 am
Al at 24:
Myhrvold compares that cost (in CO2 terms) of building a solar plant with the cost of building a coal plant. Period.
Read it again. It looks to me like Myhrvold is comparing the cost of building a solar plant with the cost of operating (not building) a coal plant over time.
October 21st, 2009 at 12:00 pm
‘I read the NYTimes article. Everything he says is accurate. ‘
Because he is careful to frame his statements – for example, comparing the amount of CO2 burned in a coal plant to the amount of CO2 used to manufacture a solar cell. A very clever way to frame the issue, since the amount of fossil fuel used to actually burn coal is then conveniently ignored, the same way that the amount of CO2 used in manufacturing the coal railcars, the 150 ton trucks, the massive shovels, etc.
Basically, anyone connected to Microsoft learned a long time ago that a certain number of people are simply incapable of actually thinking for themselves – which is a very profitable source of revenue for a company that continues to churn out software that is as demonstrably inferior. Still buying anti-virus software? – I don’t bother, but then, that is the cost of using free software. Which oddly seems to reduce the revenue stream of a company that has fostered an ecosystem devoted to the user experience – that is, getting the user experienced in paying, and paying, and paying. Somewhat like the process involved in burning fuel, compared to paying the expense up front, and then reaping the benefits for decades.
The resemblances are plain to the American arguments explaining why renewable energy isn’t workable – in a country that needs to import growing amounts of oil from such steadfast defenders of the free market and freedom as the Kingdom of Saudi Arabia or Venezuela. There is a lot of money to be made being the monopoly supplier of a necessary good or service, which is why those monopolies tend to react in a manner that reveals that the only interest they truly serve is their own.
October 21st, 2009 at 12:01 pm
There’s no reason for this to be controversial and certainly no reason for people to be making ludicrous claims about the color of objects.
Indeed, this is probably the closest we’ve come to the fabled “Democrats say sky is blue; Republicans disagree” story.
October 21st, 2009 at 12:03 pm
@18
The answer is that you can’t just look at the final situation for any means of generating power to see it’s carbon footprint.
It requires energy to make solar panels. Right now, most energy production comes with carbon emissions. Since energy is fairly fungible, even if you use windmill energy to make your solar panels, you are adding to greenhouse gasses. Because you’re using the windmills, someone else is using a coal plant.
The carbon generated in production should be assumed to be emitted over the life of the solar panel. This means, for practical calculations, that solar panels are not carbon free energy. However, as our energy sources generate less carbon in general, this effective carbon emission will shrink. It will require just as much energy to build the panel, but that energy will result in less carbon.
Next, there is the much smaller effect of the excess heat they generate. In a coal plant, they try to make as much electricity as possible from the heat they generate by burning coal. They turn about 30% of that heat into electricity. Some plants make use of some of the other heat, but most of it is just wasted, warming the planet.
Solar panels look worse at first glance. They absorb light, and turn about 14% of that into electricity, and the rest is waste heat that warms the planet. The caveat is that the light was going to hit the planet anyway. If the light hits the solar panel instead of snow, it’s causing more warming. If it’s hitting the solar panel instead of a black tar roof, it’s preventing warming. I don’t know anyone who wants to put solar panels on pack-ice.
The desert is an interesting case. The ground is fairly reflective, so your solar panel is generating more waste heat by intercepting light that would have hit the ground. On the otherhand, your solar panel is generating more electricity in the desert, so it is defraying the carbon used to manufacture it better.
You’d have to crunch all of the numbers accurately to see the net benefit of solar panels, climate wise. I think you’d find that they were a large net benefit, but probably not as much as some people claim.
October 21st, 2009 at 12:03 pm
Let’s postulate for a moment that solar panels are in fact always black, evilly black, light-absorbing and heat-generating black. Putting solar panels on roofs across America would have the deleterious effect of … wait a minute … but …
Most of my life has been spent in buildings with dark roofs.
Oh no!
October 21st, 2009 at 12:06 pm
Hum – no. He is comparing the cost of building solar vs the cost of building coal. In CO2 term, building coal is much less costly than building solar. But then he compares the cost of running solar with the cost of running coal – and states that over time (3 years), you can “break even” in CO2 terms – that is, the benefits of running solar over running coal equal the cost of building solar over coal.
Let’s take an example:
Let’s say that it costs 10 units of CO2 to build a solar plant, but it only costs 1 unit/year to run a solar plant. It might take only 4 units of CO2 to build a coal plant but it costs 3 units/year to run it.
Thus, on day 1, when the coal and solar plants both open, you’ve spent 10 units on solar and only 4 units on coal. Only after 3 years are they the same – you’ve spent 13 units for each (solar is 10 + 1/yr * 3 yrs; coal is 4 + 3/yr * 3 yrs).
The issue that Myrhvold is bringing up is that, if you are ramping up solar, your short term costs (in terms of CO2) are high, so you are not likely to see much short term benefit from solar. It is only long term that they help.
October 21st, 2009 at 12:09 pm
Much of this discussion of solar power is frustrating because it focuses solely on photovoltaic solar and pretends that solar thermal does not exist. I especially found this to be the case in the Nicholas Weaver comment Matt linked to.
Photovoltaic is still expensive and storage of the generated electricity for nighttime/cloudy day use remains a problem. Large-scale solar thermal is fairly cheap–nearly competitive with coal and gas and MUCH cheaper than nuclear–and the energy storage problem was solved decades ago. These plants don’t use special expensive materials like photovoltaic cells.
October 21st, 2009 at 12:12 pm
And here is the quote – ‘The point I was making to Dubner and Levitt is the following: when you build a solar plant it costs you energy. Lots of energy. Pacca and Horvath, in a 2002 study, found that the greenhouse gas emissions necessary to build a solar plant are about 2.75 times larger than the emissions from a coal plant of the same net power output (1.1 * 1010 kg of CO2 to build the solar plant versus 4 * 109 kg of CO2 per year for coal).’
Like I noted at the start, just a quagmire, filled with people churning the mud as quickly as possible, because they have nothing valuable to offer at all, except for blinding everyone to the truth that the pit was dug for the benefit of others, and not those mired in it.
October 21st, 2009 at 12:14 pm
What Myhrvold is completely eliding over is the CO2 effect. Yes, solar power generation means that some waste heat is emitted into the atmosphere. But that extra waste heat pales in comparison to the extra heat trapped by CO2 emitted by coal plants. The solar panel emits some extra infrared radiation, but then it’s gone. The CO2 emitted by the coal plant remains in the atmosphere. A single ton of coal may only emit a moderate amount of waste heat when burnt, but the CO2 produced will trap hundreds of kilowatts of heat over its lifespan in the atmosphere.
October 21st, 2009 at 12:17 pm
not_scottbot:
Huh. Haven’t checked it myself, but if that’s right, it would mean Al is a gigantic lying hack. I just can’t see it.
October 21st, 2009 at 12:19 pm
One wonders why Myhrvold couldn’t say ‘black body’ anywhere in his rebuttal, which is I think his basis for the claim. Which is a stupid claim.
The reason it is stupid is because…
1) It is one thing to have lowered albedo, but for what fraction of the earths surface.
2) When talking about global warming, thermal heat is a ’second order’ effect – it happens once when you generate the electricity. The trapping of sunlight by carbon is something that keeps on happening forever.
October 21st, 2009 at 12:24 pm
Al, I don’t read the passage I have in mind that way (it’s the one quoted in #37 above) — but I do agree about what Myhrvold’s overall claim is (solar is more carbon-costly up front but then catches up over time), so it probably doesn’t really matter. Whether Myhrvold is pulling a fast one as per 31/37/38 is something I don’t feel qualified to address.
October 21st, 2009 at 12:25 pm
I don’t think this makes Myrvold wrong. He never says solar is bad or coal is good. He should have included more depth on the analysis of the coal costs, but I don’t think he was trying to make a comprehensive cost benefit analysis. Instead he was trying to show that a complex cost/benefit analysis should be done, and makes a tiny step toward doing so. He picked a few facts to illustrate that solar power was not carbon free. He could (and should) have mentioned some of the things that add to the carbon costs of coal, but it wouldn’t have been very instructive. The difference between “none” and “some” is bigger than the difference between “a whole lot” and “even more than a whole lot”.
October 21st, 2009 at 12:29 pm
Always, always go to the sources when dealing with people discussing the broad issue of climate change. And then, think for yourself.
For example, one point often heard from those scoffing at climate science is how flawed the current models are. Which is true, actually – since observational data collected in real time shows that even the most pessimistic consensus models are not actually keeping up with the measured changes around us. That’s right – the models are flawed. They are flawed because they are far too optimistic, which would not seem to be the conclusion that those making this argument are interested in having anyone draw.
Basically, there is a group of people who have little interest in reality, and yet they loudly claim the mantle of being ‘realists,’ unlike those they so steadfastly oppose. Which is just a symptom of something much deeper than any interest in observing the world around us, and the interactions which define it. Especially since they seem unable, or unwilling, to grasp that reality is not about ideology.
October 21st, 2009 at 12:31 pm
The best red herrings usually are.
But I see that what Myhrvold left out, for whatever reason, is already under discussion both here and in the NYT blog comments section. It’s nice that Myhrvold came back to explain himself, but he actually wastes most of his article bitching about other people’s tone, only to be picked apart by people more calm than himself.
Makes for good reading.
October 21st, 2009 at 12:35 pm
It works on the small scale too, but maximum efficiency requires incorporating it from the point of home construction, and only constructing these homes at normal replacement levels.
October 21st, 2009 at 12:40 pm
‘I don’t think this makes Myrvold wrong.’
Well, he isn’t comparing the cost of building and fueling a coal plant with that of building and then using a PV panel, which already makes his comparison, charitably, somewhat dishonest. In other words, the coal plant didn’t magically spring into being, and the coal didn’t mystically arrive at the furnace, without a fair amount of fossil fuels being burnt, before the coal itself is burnt.
Reasonable people may disagree, but this argument is not really based on reason, this is a carefully constructed smokescreen. Possibly for no greater purpose than to increase book sales, though such willful construction of a seemingly reasonable framework which any thinking person can shred so easily upon reflection probably has another goal in mind.
For example, to increase the amount of mud being thrown in all directions, so no one can see anything clearly.
October 21st, 2009 at 12:40 pm
… or maybe not so accurate.
and
and
and
and
There are many more, but I think we can dispense with “technically accurate” as a description of Mr. Myhrvold’s article.
October 21st, 2009 at 12:46 pm
[...] Matthew Yglesias » Myhrvold on Solar: Blue is a Kind of Black [...]
October 21st, 2009 at 12:52 pm
The difference between “none” and “some” is bigger than the difference between “a whole lot” and “even more than a whole lot”.
What? If you’re talking about carbon, I would say that’s probably not true. Even if solar panels generate “some” carbon dioxide in the construction (which seems to me like its too obvious to even bother pointing that out), that would seem to be irrelevant when compared to the “even more than even more than a whole lot” generated by coal.
Many arguments against alternative energies are valid, but some are just ludicrous. “Solar panels are black so they have a lower albedo” seems to be of the ludicrous variety. It would likely only be of any concern if we decide to cover 10% of the earth’s surface with solar panels, in which case I say we can cross that bridge when we come to it. Someone who can crunch the numbers, figure out the lowering of the earth’s albedo by solar panels, and compare that to the lowering of the earth’s albedo by polar ice melt caused by greenhouse gases, and we’ll see what the real concern is.
October 21st, 2009 at 12:55 pm
Most States have “net-metering” laws which make the “no power at night” issue moot. They are not usually reimbursing at full-retail rates though.
And as others have said, if a tiny portion of current black tar roofs are covered up with solar panels, you reduce the heat energy reflected back – not increase it. And my German-made solar hot water system kicks ass no matter how you measure things.
October 21st, 2009 at 12:59 pm
Romm, so far, has romped in this discussion, which has been framed of course by the Superfreaks to divert the conversation from CO2 – thus, the attention paid to the thermal heat effects of solar power, and the complete lack of attention to the CO2 put into the atmosphere by the coal plants. I know the latter is so technical – in fact, so technical that it is the subject of study by Dubner and Levitt’s other supposed source, Caldeira – the man who says, according to D and L, that co2 is not the villain here – and who says, on his website, that CO2 is the villain here. Funny – why didn’t they invite him to write on their blog?
Actually, using the NYT to mount this all week attack on Romm should have more consequences than making D and L a hero of the denialist crowd – who’ve now become concern troll denialists, still irrational but 50% more hypocritical (and by the way, that hypocrisy takes a lot of carbon to build – you have to pump the oil, you have to pay the bills for AEI and Heritage and the GOP operatives, they have to hire albots, etc. It all adds up). One consequence for sure is that they should force D and L to invite Romm to write on the blog, and thus make this a fair fight.
October 21st, 2009 at 1:02 pm
At this point I might just be adding my voice to the chorus, but let me reiterate that black has nothing to do with color. When physicists say something is black, they mean it is a perfect absorber of light. Solar panels are close to perfect absorbers, so they are black. It doesn’t matter what color they are to our eyes, it only matters what their absorption properties are.
So, solar panels are black.
October 21st, 2009 at 1:10 pm
[...] tearing the offending chapters apart, with Brad DeLong and Matt Yglesias chiming in over the color of solar panels and geoengineering: And I definitely don’t think people should misinform their readers by [...]
October 21st, 2009 at 1:20 pm
‘it only matters what their absorption properties are.’
Which is most certainly true in terms of solar collectors used for heat, but takes on another perspective when talking about a PV panel. My panel (I run my router, speakers, laptop off it, using salvaged UPS batteries for the night) is ca 21% efficient – German made, though not blue tinted, nor is it silica based. How should one measure the 21% percent of converted solar radiation – as electricity, or also as heat, which is what it becomes, eventually.
Even more interesting, one of the features of this type of PV cell is its ability to efficiently convert (scavenge isn’t too bad a term) low levels of light – that is, the cell will actually work better on a late afternoon on a cloudy day than on a clear day when in shadow, as the amount of solar radiation is actually higher in terms of ambient light – though of course, it would be difficult to claim that the PV cell is an efficient black body in such circumstances. Myhrvold’s ignorance, to be charitable, of the sorts of technology likely to installed in much larger amounts that what has been installed to date is just another aspect of making sure that the discussion remains mired in distractions.
October 21st, 2009 at 1:30 pm
I, like many others, had always thought that solar power did not contribute to climate change at all b/c it doesn’t generate any greenhouse gasses. Now I’m hearing, for the first time, that this is not true, because the color of solar panels (whether black or blue) causes something or other to happen that also contributes to climate chage. Bottom line: is that true or not? If true, how seriously does that reduce solar’s promise as a non-climate-change-contributing energy source?
It’s true but, so far, trivial. Solar power does not involve burning fossil fuels, but dark* objects absorb more heat and light energy than light objects, and once absorbed by objects on earth such as solar panels, that energy generally stays within the atmosphere. This makes the earth warmer than it would if all our solar panels were a light color but magically otherwise identical**.
We can observe this effect in similar situations. Big cities are often a couple degrees hotter than nearby undeveloped areas, partly because miles upon miles of black-or-just-dark pavement and roofs soak up heat instead of reflecting it back into space.
As Myhrvold himself says in the blog post Matt links to, this does not mean that solar panels are a net loss as a way to fight climate change. And in current use of solar panels – a roof here, a single panel on a traffic sign there – it really doesn’t matter. When/if we blanket hundreds of miles of desert with solar panels, then it will alter the weather patterns in that area. It would be irresponsible not to consider that when someone designs one of those solar power plants, but it is irresponsible to act like this has any significance for the merits of solar power in general.
* The argument by ChooChoo! and others about whether solar panels are “really” “usually” black or not is stupid at best, but simply calling them “dark” should be safe…
** And it would take magic, since the whole point of solar panels is to absorb solar energy.
October 21st, 2009 at 1:39 pm
Myhrvold is missing the forest for the trees. Putting up a solar panel is like painting a very tiny percentage of the earth black. CO2 emissions are the equivalent of painting the entire surface of the earth black. The waste heat from a solar panel pales in comparison to the heat trapped by CO2 over the entire earth.
October 21st, 2009 at 1:41 pm
It doesn’t matter what color they are to our eyes, it only matters what their absorption properties are.
The blue we see isn’t magically added in post-processing, it’s actually light that’s been reflected off of the solar panels. Which they wouldn’t do if they were perfect absorbers of light.
So, solar panels aren’t black.
October 21st, 2009 at 1:42 pm
I would say that he is missing the forest for a tiny, insignificant shrub.
October 21st, 2009 at 2:06 pm
This argument seemed silly to start with. Assuming we are putting solar cells on the roof, they probably are not going to promote heat gain significantly at all. Informal survey out of my window shows that 100% of the roofs in my neighborhood are dark. I am planning on putting a white roof on my house. The solar gain with the current dark roof is obnoxious for 3/4 of the year, even in dreary Central PA. Although some posters in this thread have made me consider solar hot water as well.
October 21st, 2009 at 2:07 pm
[...] the RZA’s interview on The AV Club and how the Superfreakanomics authors are liars about climate change like most libertarian mouthpieces, even those who typically make intriguing, intelligent points. It [...]
October 21st, 2009 at 2:12 pm
JM@47,
The first comment you quote is garbage. The person completely misunderstood what Myhrvold wrote. Myrhvold states clearly that in the long run the solar plant would generate less CO2. It is only at start-up, and the short period thereafter in which the solar plant has generated more CO2.
This comment:
Efficiency has barely changed. The quotes you’ll see are for panels kept at room temperature. A panel that is being hit by a useful amount of light will be less efficient at conversion. Sixteen per cent is about the best out there for a commercially available product. Manufacturing efficiency is much better, though. For cheaper solar cells, carbon payback comes in 2-5 years, depending on where they are. For more expensive ones, 2.75 years is an upper, rather than lower limit. Payback can come in as little as 1 year. It means paying about 10 times as much for electricity though. It would only double the price of coal generated electricity to capture 90% of the carbon.
The albedo points are good. I was surprised to see sagebrush albedo was down around .17.
October 21st, 2009 at 2:19 pm
I love this picture on James Fallows’ blog.
October 21st, 2009 at 2:23 pm
I was talking about perceptions. Everybody knows coal burning plants generate CO2. They don’t know how much, and if you told them, they wouldn’t know the impact of that number. If you made a mistake and told them a number that was 20% too low, it would be a meaningless mistake.
However, many – probably most – people think solar energy generates no CO2. That is wrong. Correcting that mistake has greater value because it has some possibility of making people think in a more comprehensive way.
For wonks grinding numbers to set policy, the error on the coal side probably is more important, but we aren’t them.
October 21st, 2009 at 2:23 pm
Yeah Njorl is pretty much right. And basically excess heat is not a super important issue. It is worth trading off excess heat for less CO2.
October 21st, 2009 at 2:26 pm
Nobody said they were perfect absorbers. The point is that the reflected light is small compared to the absorbed light. The fact that that tiny amount of light happens to be blue is a triviality.
October 21st, 2009 at 2:28 pm
In whose interest is it to be hyping the tiny and largely irrelevant amounts of CO2 generated by solar panels? We shouldn’t misdirect or be dishonest, but there’s no reason to be making the oil and coal company’s arguments for them.
October 21st, 2009 at 2:52 pm
If you understand the physics of black body radiation, and are chatting with other physicists, solar panels can be treated as black. If you are chatting with glib bloggers, well, duh, they’re blue.
October 21st, 2009 at 3:01 pm
Also, I invite anyone who is big on geo-engineering to watch the Futurama episode on global warming called “None Like it Hot”, where for years they solved global warming by dropping a giant ice cube into the pacific ocean. Eventually, of course, they ran out of ice. So the “solution” to that was to move the earth into a new orbit so that it was further from sun. Thus solving the problem forever… FOREVER!
October 21st, 2009 at 3:02 pm
@Njorl
“The fact that that tiny amount of light happens to be blue is a triviality.”
Of course, it is not a triviality at all, but rather a way to minimize the waste heat generated by maximizing the energy of the emitted light (E=h*v) where v is really the greek symbol for frequency, which I can’t do in comments, but I’m sure others can. blue light is the highest frequency, so making a panel blue means reflecting back high energy light (since the panel can only convert 15-20% of the energy to electricity anyway). This minimizes the waste heat. Weird that engineers who design these things would think of something that a first year physics student could anticipate, huh? Stop wasting all of your time on this, and do something productive, for crissakes (that last comment is directed at all of you (and me!) and not just at Njorl.
October 21st, 2009 at 3:03 pm
Matt did indeed mess up his science on this one. I think.
But can anyone comment — does radiated heat contribute to global warming? Those extra watts that solar generates — does that matter? I thought that CO2 was a problem because it accumulates. But heat doesn’t work like that.
If excess heat is a problem, should be be talking about painting roofs white? It just seems weird — nobody cares that shingles are black, right? It’s not considered significant? So why the concern about panels?
October 21st, 2009 at 3:08 pm
If excess heat is a problem, should be be talking about painting roofs white? It just seems weird — nobody cares that shingles are black, right? It’s not considered significant? So why the concern about panels?
People do talk about painting roofs white, and it would be useful. But compare the amount of the earth’s surface covered in roofs and asphalt to the amount covered in solar panels. The amount covered by panels is trivial. In addition, solar panels are often placed ON ROOFS, and unless I’m missing something, putting a black object on a black roof means no extra heat. If we were to suddenly decide to cover the entire Sahara in solar panels, this MAY be a concern. But to worry about the heat generated by the albedo of solar panels and ignore the effects on albedo of melting ice caps is to fret about a mosquito bite while a bear is gnawing on your head.
October 21st, 2009 at 3:14 pm
These facts about solar will have zero impact on policy. Solar will not be used because it is an economic disaster compared to other clean sources of energy. It’s about 5 times as expensive as wind power. It’s about 7 times as expensive as coal. You could tax the carbon emissions from coal and use the money to extract CO2 from the atmosphere and come out cheaper and cleaner (CO2-wise) than solar.
Solar cost about $.33/kWh more than coal. It would cost about $.06/kWh to capture 90% of carbon emissions at the source of a coal plant, and another $.05/kWh to pull the last 10% out of the atmosphere at large. All we need is a $500/ton carbon tax. That would triple the cost of electricity, but it’s still a lot cheaper than solar.
October 21st, 2009 at 3:24 pm
First, waste heat is almost certainly a much less serious problem than greenhouse gases.
The global albedo (essentially how much light gets absorbed rather than reflected) is important, but solar panels really are a drop in the ocean compared to other effects. Sprawl, for example, means you have a lot of black roofs on 2 story houses rather than a few black roofs on apartment buildings, more streets and more parking lots. Receding pack ice would be bad because you’d be replacing highly reflective snow with highly absorbing ocean.
October 21st, 2009 at 3:39 pm
I think you’re really burning a strawman here, Matt. The author’s point is that solar panels absorb a lot of sunlight and release most of it as heat. Whether or not the panels are black or dark blue is irrelevant; both colors absorb a crapload more sunlight than, say, sand in the desert or light surfaces.
And furthermore, the authors *main* point is that solar is great, but it isn’t going to make a huge impact for another 30-50 years. And, given the amount of energy we consume now and the amount of it we get from solar, I’d say that is accurate if not optimistic. If you’re looking for shorter term climate solutions, like for the next decade, we need to look elsewhere. Is that an unreasonable point? I don’t think so.
October 21st, 2009 at 4:00 pm
It would be wasteful to intentionally reflect significant amounts of blue light. While it generates more heat per photon, it also has a higher likelyhood of generating an electron-hole pair that will not recombine before being swept out to the conductors. Amorphous silicon cells, for example, have their peak current production from blue light. The panels reflect very little blue light, they just reflect more blue than red. Some panels are designed to optimise blue light absorption because blue light is scatterred preferentially by the atmosphere and so is available when direct sunlight is not.
Many panels have UV reflection coatings, which could be responsible for the blue color if they were highly reflective in the UV, and slightly reflective in the blue.
To finish, physics is never a waste of time.
October 21st, 2009 at 4:11 pm
[...] Matthew Yglesias » Myhrvold on Solar: Blue is a Kind of Black [...]
October 21st, 2009 at 4:34 pm
Do a Google image search on “solar panels”.
What kind of an idiot do you have to be to argue with this?
I think Al has (unsurprisingly) said the dumbest thing in this dumb thread:
“Properly put, solar panels are both black and blue”
Al, those are mutually exclusive categories. That’s why they both exist.
Blue means “dominated by energy with a wavelength of roughly 440–490 nm.”
Black means “does not emit or reflect light”.
Can’t have the first if you have the second.
October 21st, 2009 at 4:54 pm
OK, this is just annoying. Here is a comment, heavily numerical, for those who understand the physics. Yes Blue and Black are about the same, and what Myhrvoid says is physically correct, but his point is just plain silly.
Myhrvoid, knows physics, but here he is just being stupid, using the correct physics. He pushed the numbers as far as possible and came up with
10 times as much heat as electricity (probably a factor of two to five too big but why quibble — we aren’t even in the ball park of interesting effects).
A CO2 molecule is estimated to trap 100,000 times its heat of formation in its lifetime. And the heat of formation is an overestimate to the electricity produced whatever made the CO2 (you need to break apart the coal, break apart the O2, convert to electricity, …).
So Myhrvoid is saying a solar plant might have a 10 times magnifying factor in contributing to global warming instead of 0 like you thought. That is to say it adds 10 times as much heat as electricity it produces. But fossil fuel magnifiers are in the 100,000’s, maybe 200,000 would be a good guess, so 10 is a huge win.
Oh and by the way a coal plant uses coal, the coal is mostly surface mined and the area of a surface exposed coal needed to supply your coal plant is about the same as the area of an equivalent solar plant. And coal is kind of black. So really the magnifier for coal ought to be 200,010 — but the 200,000 was just an approximation anyway.
October 21st, 2009 at 5:04 pm
Matt, you are off base with your BLUE stuff. What matters for Myhrvelds foolish direct heating argument is net absorption of solar radiation minus the amount that isn’t converted to heat, i.e. the useful output. Dark blue is almost (within a couple of percent) as absorptive as black. And many solar panels are now black.
But, of course the direct heating thingy is only an issue for local heating. For example a dense urban area with a lot of panels displacing power that would have been transmitted from outside the area could have a greater urban heat island effect with the addition of the panels. Of course their argument about the effect on the global temperature is still bogus, but the color of the panels is virtually immaterial to that case.
October 21st, 2009 at 5:34 pm
Why would it even matter? Solar panels, essentially, absorb the suns heat, and turn it into electrical energy. Any energy that isn’t converted to electrical energy would be radiated out as excess heat energy.
I haven’t read the book, but since I’m sure they aren’t implying that the laws of thermodynamics are being violated, they would have to show that out of the total energy hitting the solar cells, that the amount that could not be converted to electrical energy, and is therefore radiated out as heat, would exceed the radiation of whatever surface would have been there, absent a solar panel. Do they do this? Otherwise, “cuz They’re black” isn’t much of an argument.
October 21st, 2009 at 5:43 pm
Why would it even matter? Solar panels, essentially, absorb the suns heat, and turn it into electrical energy. Any energy that isn’t converted to electrical energy would be radiated out as excess heat energy.
Before you start lecturing us, you may want to learn the basics first. Solar panels don’t absorb the sun’s “heat”.
October 21st, 2009 at 5:46 pm
Even if they were right, they’d still, possibly, be wrong… The excess of greenhouse gases, would do just what their name implies, create a greenhouse effect. The “greenhouse effect” is when light getting into the atmosphere, and is reflected, can’t escape the atmosphere because it is reflected back by all the greenhouse gases in the upper atmosphere (yeah, I know this gets more complex, as it could, at various stages, prevent some light from getting in, yada yada, either way, more net heat). So as long as the heat is trapped in our atmosphere, we could be doing something useful with it, and turn it into energy… which is what solar panels would be doing.
If using solar panels would offset enough carbon usage, then the greenhouse could disapate, hopefully to the point that we could start worrying about their radiant heat… which engineers already do, as they strive for more and more efficient solar cells.
Until then, fossil fuels are still a much larger concern.
October 21st, 2009 at 5:57 pm
Yes, thank you Aqua Regia, for being one of those assholes. I’m well aware that it’s EM radiation, but most non-physics majors just think of it as “heat”, which is generated by the excited state of molecules in a system, which is precisely what is leveraged to generate the voltage transfer in a photovoltaic cell.
October 21st, 2009 at 6:04 pm
I think even non-physics majors can deal with the difference between “heat” and “light”.
October 21st, 2009 at 6:07 pm
Questions like Jay’s (@70) shows how this debate obscures the forest for the trees. Waste heat does contribute to warming, but in an utterly insignificant manner compared to CO2. The CO2 produced in generating waste heat is, according to the figures I’ve read, 200,000 times more significant in contributing to warming. Seriously. 200,000 times.
If that is so, then you can imagine how dishonest it is to raise the issue of whether or not solar panels are black as a subject of discussion on the issue of AGW.
October 21st, 2009 at 6:08 pm
er, sorry for yelling…
October 21st, 2009 at 6:58 pm
[...] Matthew Yglesias » Myhrvold on Solar: Blue is a Kind of Black [...]
October 21st, 2009 at 7:17 pm
@58 Aqua Regia: “I would say that he is missing the forest for a tiny, insignificant shrub.”
That is what we are good at in this country. Focusing on the shrubbery. I can see us arguing about black and blue for years because two clowns figured out a way to make some money.
Good link, too, Aqua Regia. That’s what solar is all about right now. Solar as a practical way to heat water. Micro solutions. Solar panels in cars to power electric engines. Etc.
Then, as you continue to build a smarter electric grid, power is ALLOWED -by government fiat- to flow to the grid from various micro sources. Then, as inevitable advances in technology make solar an extremely practical energy source, you proceed to build hundreds of giant solar plants and eventually eliminate the need to use coal altogether.
Of course, we know there is no country on earth mighty enough to have any such plans. Or do we?
October 21st, 2009 at 11:32 pm
I have some of the ‘black’ or really dark solar panels and can attest while they get hot, they do not actually have a low albedo because most of that energy is turned into electricity and the red and IR light mostly passes through them.
What a crazy, crazy argument that solar panels will heat up the planet. Yeah, they would, but on the other sense, they also provide shade. It’s like saying trees absorb light, and therefore heat up places!
October 21st, 2009 at 11:36 pm
What do you mean by “allowed by government fiat”?
Micro-generation, solar panel use, and general efficiency would all be orders of magnitude higher in an unregulated power market.
October 22nd, 2009 at 2:21 am
I feel embarrassed for this Yglesias character when he tries to write about science and doesn’t know what he’s talking about. Policial blogs — ROFL!
October 22nd, 2009 at 2:22 am
Political blogs — ROFL! I meant
October 22nd, 2009 at 2:49 am
So the assertion that Superfreakonomics makes, wrongly, is that black solar panels actually increase global warming because they are black, and the color black absorbs heat. That couldn’t be more wrong, and here’s why: the POINT of a solar panel is to collect solar radiation (light/heat) and convert it into electricity. So it’s not as if that heat soaks into the atmosphere or something, it is converted into electricity.
Not only that, but that electricity which comes from solar power then offsets the need to generate an equal amount of electricity by doing something polluting, like burn some coal, which in turn would release more carbon into the air. Let’s not forget that the thing about carbon in the atmosphere that’s bad is that excess light/heat which already bounced off the earth’s surface and is on it’s way back out into space where it’s not our problem anymore rams into a floating speck of carbon, and shoots back to earth, where it heats things up more than Mother Nature intended. So by converting whatever light/heat the panel absorbs into electricity we not only don’t send the light careening back towards space in the first place, we prevent more carbon from being put out there for light that DID bounce off something to smash into. And that means it’s a double win for Mother Earth.
But there is one part here that is correct… black absorbs light/heat, white reflects it. So if you’re going to have any part of your roof any color other than white, which you really SHOULD do, that part had better be the solar panel. One more thing, there are many types of solar panels but nearly all have something in common: crystals. If you take them apart, they have lots of crystalline structures in them, usually right on the surface. Crystals reflect certain wavelengths of light, and absorb others. So when you compare that to a gravel & tar shingle, it doesn’t take a super freakin rocket scientist to figure out which will reflect more light, and thus heat, out into space.
This book and all it’s anti-global warming BS is brought to you by: America’s Coal & Oil Industries.
October 22nd, 2009 at 5:08 am
@90 Sycophant of the Bourgeois: “What do you mean by “allowed by government fiat”
Sweeping and binding Federal legislation, passed by Congress and signed into law by the President. I’ve made personal decision to use the word “fiat” to describe any piece of legislation that benefits the nation as a whole. In other words, legislation that is not completely destroyed by special interests.
“Micro-generation, solar panel use, and general efficiency would all be orders of magnitude higher in an unregulated power market.”
The power producing companies and the entities that own sections of our power grid will strive to prevent decentralization and independent power production -a National Smart Grid will cut into their EASY action. Plus, decentralization will require a great deal of coordination at the regional and national levels. Meaning: a Smart Grid can only work with government in complete control. Hence, the word fiat.
October 22nd, 2009 at 10:53 am
The obvious problem with solar power is that it doesn’t work when the sun isn’t shining.
The sun is always shining.
October 22nd, 2009 at 10:53 am
[...] colour are solar panels, really? (Matthew [...]
October 22nd, 2009 at 11:55 am
It’s actually not accurate to say that solar panels don’t work when the sun isn’t shining. In fact, they do work, just less effectively. I’ve designed buildings that have incorporated solar panels and, even on overcast days, the meter clicks forward as the panels generate energy.
October 22nd, 2009 at 1:44 pm
[...] Matthew Yglesias » Myhrvold on Solar: Blue is a Kind of Black [...]