Advanced Search

Please click here to take a brief survey

Solar Carbon Payback
Jeremy Faludi, 18 May 09

Some naysayers argue that solar panels don't make sense because it takes so much energy to make them--mining, smelting or refining, processing, etc. Do they really save fossil fuel energy and greenhouse gas emissions over the long run? The simple answer is yes. They save a whole lot. But then the question is: what kinds of solar panels are better than others?

A recent life-cycle analysis published at the Institute of Science in Society (ISIS) showed that in a nice sunny place like Spain, PV panels reach energy payback (when they've saved as much fossil fuel as it took to make them) in about one to three years, depending on the type of panel. Interestingly, the new thin-film chemistry cadmium telluride (CdTe) fared best at 1.1 year despite having the lowest efficiency (9%), while monocrystalline silicon fared worst at 2.7 years despite having the highest efficiency (14%). This may seem counterintuitive, but the explanation is simple: it takes a lot less energy to make CdTe film. See the graph below, which shows greenhouse gas emissions from the manufacture of different kinds of PV panels (each kind scaled to the same lifetime power production). (The four bars on the right, "Case 3", is US data.)


If your climate is less sunny than Spain, there will obviously be a longer payback time; but most of the continental US gets as much solar radiation as the location in the study. (It got 1,700 kWh/m^2/year, which is 4.7 kWh/m^2/day. As the map below shows, basically everywhere in the continental US is this sunny or better, except for the Pacific Northwest, upper Midwest, and New England.)


Avoiding Cadmium emissions

You might balk at the thought of cadmium in PV panels--after all, isn't it a toxic heavy metal that's outlawed in RoHS? Well, yes. But the especially interesting thing this ISIS article says is that using cadmium solar cells actually reduces cadmium emissions to the environment. Cadmium is contained in the exhaust of burning fossil fuels, and so much exhaust is prevented by using the panels instead of fossil fuels, that it pays for itself ten times over. Of course, crystalline silicon PV panels will be even better, because they don't directly contain any cadmium (30x improvement over fossil fuels).

Death Knell for Crystalline PV

Though thin-film PV panels have been in the market for more than a decade, the new chemistries that are coming to maturity (mostly CdTe and copper indium gallium selenide, "CIGS") are fundamentally changing the game. They are still not as efficient, but they are good enough, and they are far cheaper (up to 1/3 the cost per watt), even before reaching true economies of scale from mass-manufacturing. Now we see that they are environmentally preferable as well, since they have faster energy payback times. In the years ahead, crystalline silicon will become relegated to satellites, mobile systems, and other applications where compactness is king, while thin-film PV will be the default style. Crystalline prices will stay high, but thin-film prices will continue to drop steadily until they hit grid parity (arguably they have already), at which point they will plummet because manufacturing will ramp up massively. CdTe and CIGS will ultimately be price-limited by the rarity of their ingredients; the next wave after them will be dye-sensitized solar cells, made from common titanium dioxide and organic dyes. It'll be probably ten years or more before those dominate, though, and in that much time, who knows what other chemistries will come along? In any case, keep an eye out for it: the future of PV is thin.


image credits: ISIS, NREL
front page image: flickr/kevinthoule, Creative Commons license.

Bookmark and Share


Why don't we have solar cells with no toxic materials at all? Plants generate "solar cells" (aka leaves) from soil without cadmium or telluride. I don't understand the biology in detail, but is there some reason the plant method of converting to sugar can't be used to make electricity directly?

Posted by: Dave C. on 18 May 09

Toxicity is relative and many-fold. Can solar cells be made completely non-toxic? Unlikely. (Dyes, even chlorophyll, are toxic by nature)

Can they be made using materials and processes that result in much lower waste and toxicity? Apart from know-how, and the 'business case' to make use of that know-how, yes.

See Jeremy's final remarks on Titanium Dioxide, and Sarah Kuck's article 'What Would Nature Do?'

Posted by: Tony on 18 May 09

There's a Brookhaven Labs report on lifecycle costs of photovoltaics a year or so ago. It found paybacks of 1.1 to 6 years depending upon the type of PV.

See for more.

Posted by: gmoke on 18 May 09

Hi All,

This carbon payback article is totally overstated and absolutely worthless discussion because in 1-2 years ALL manufacturing and the processes to make solar cells will be using the sun's energy directly to make these products with NO carbon footprint. A totally green process is all in the news RIGHT NOW as one solar company after another announces installation of their own solar panels for their total energy manufacturing consumption. The solar panel manufacturing companies and chemical processing people are no dummies. They know what going on and are rethinking their entire processes to go totally clean energy using the very solar panels they manufacture to supply the energy.

So this article is just blowing in the already cleaner and greener wind. Just give them a little time folks and the entire process will be green from start to finish even the huge mining trucks are being retrofitted to all electric, but we have to settle for hybrids for now, just as our car manufacturers are doing to us.

My two cents are on the thin film guys with a great foil printing process. Just wish they would hurry up and flood the market so pv/watt prices would drop and every household and commercial building in the world would be clammering to sign up for a roof retrofit. Making that buzz happen quickly will be the only thing that will save this planet from drastic climate shifts. It would then drive the buzz and supply the extra electric needs for all electric cars, all electric manufacturing, and all electric processing of every commercial and residential sector.

We are all participates in the dawning of a new era that will be talked about in history books for thousands of years to come, for better or for worse. Let the BUZZ continue, only louder PLEASE until every man, woman, and child knows these things as intrinsic truths embedded in their memories!

For a better greener tomorrow,

Robert Grothe
TCM & Associates, Inc.

Posted by: Robert Grothe on 19 May 09

I'm wondering if the Jeremy's previous "peak gallium" articles impact on the solar article above?

Surely we need not just renewable energy making the next generation of renewable energy, but also from renewable "cradle to cradle" materials as well.

Posted by: Eclipse Now on 20 May 09

Hi Robert Grothe,
yes it will be a wonderful day when green energy makes the green energy, because it will finally dispel one of the pessimist's arguments which is that the ERoEI of renewables does not stack up. How many times have I heard 'doomers' say that solar PV is not even an energy source, but just fossil fuels in disguise (because it presumably took SOOOOOO much energy to make). It's like they assume engineers just cannot ever count all the energy inputs. But the studies all include their list of included energy costs in making these, so I don't know what their problem is?

Worldchanging is in my top 3 favourite sites on the net, and as such I know that they were trying to dispel some of these doomer myths.

Keep up the great work guys!

Posted by: Eclipse Now on 20 May 09

To respond to:
Why don't we have solar cells with no toxic materials at all? Plants generate "solar cells" (aka leaves) from soil without cadmium or telluride. I don't understand the biology in detail, but is there some reason the plant method of converting to sugar can't be used to make electricity directly?

Plants do need metals to collect light, in the form usually of Magnesium and Manganese. The problem with plants is that they aren't very efficient, usually 1% in most plants (it can get close to 10% in some species). Also they do require lots of water to keep their electron cycles flowing... so while we could use plant-type cells, it would be a very bad idea. Solar cells are the better of the two.

Posted by: KingKaid on 20 May 09

Regarding titania/dye-sensitized solar cells or DSSC's, the major problem with these materials is that TiO2 (a great photo-catalyst) + sunlight + organic dyes = chemical reaction. Meaning, that the dyes decay upon exposure to light, a big problem for a PV material. These things have short life-times. I think it's unrealistic to expect this long standing fundamental issue to be solved to the point of rolling out large scale commercialization of DSSC's for power generation within 10 years. In my opinion these materials will find a niche market where limited life-time can be tolerated.

Posted by: Dr Pete on 20 May 09

A particular PV may well be energy positive and carbon negative over its life time. However, this doesn't solve the problem of large scale renewables ramp-up exacerbating energy scarcity in the short-term, in a peak oil context. Renewable energy cost are loaded on initial stage while payback is over the lifetime. Can or will sociey as a whole afford to make these investments? Further invaluable discussion at


Posted by: Simon Tegg on 20 May 09

This is a highly informative article which gives me potentially very useful information that I'm glad to have learned.

Do you have any information, or can you direct us to such information, on the cost / benefit statistics of thin solar panels in terms of the amount of fossil fuel conserved by their use throughout their probable lifespan, versus the amount of such fuel required to manufacture and distribute a given unit of solar panels? Rather than unsupported statements like "so much exhaust is prevented by using the panels instead of fossil fuels, that it pays for itself ten times over," I would like to see hard data that lets me work out the math for myself.

In those terms, and in terms of electrical productivity relative to cash outlay, it would be interesting to see a comparison of thin panels with the solar shingles I recall being touted a couple or three years ago.

What is the current availability of thin solar panels for the ordinary person? At this time, would I have to order a large quantity from some centralized distributor, or could I economically purchase a small amount from some local retailer? Can I stick a few square yards of such panels in the back yard of my rented home for personal household use, or would I necessarily have to incorporate it into the design of a new building or major retrofit? If these panels are not currently available on such terms, can you say how long it's likely to be before they become so?

About solar panels with dyes in them: I trust these dyes will by thoroughly lightfast? Many are not, and rapidly fade with exposure to sunlight. Durability in the face of continuous solar radiation would seem to be an essential quality in solar panels.

Not trying to be negative. I'm just a "show me" kinda guy.

Posted by: Kevin on 22 May 09

Great article. A minor point about cheaper and less efficient panels: "Now we see that they are environmentally preferable as well, since they have faster energy payback times." Given a finite place to put panels (my roof), the cheaper materials pay back faster but the more expensive & more efficient may still pay back more per roof over the long run.

Posted by: Stephen Cataldo on 22 May 09

1) In 1979 I had a biochemistry course that outlined the detailed of photosynthesis - all the steps, all the enzymes and co-enzymes. They understood it all. But in the next 30 years, with all our super-computers and amazing technology, we still can't reproduce it. To say it's not efficient enough is pure stupidity. Ever see an Agave flower grow 15" a day? Ever watch bamboo shoots grow? Ever wonder how a 150' tall tree gets water to every last leaf? (Through capillary action it moves hundreds of gallons a day with NO pump).
And to the guy who wishes Solar PV was on 'every home and building in the whole wide world' - well, if the roof faces north or isn't big enough or is blocked by trees or other buildings? It isn't right for every situation.
Plus, it's only good for one-story, low use buildings - schools, warehouses. 2 stories - it only takes out 1/2 the bill. 3 stories - 1/3rd. 4 stories - 1/4. Whose going to spend $1 - $2 million to COVER their whole roof and only take out 20-25% of their bill? No one.
Why not put vast solar arrays in southern CA, most of southern AZ and NM and TX and then transport that across the country via DC (much more efficient that AC) and have local inversion stations to deliver the AC. It would cost $300 to $500 billion - a fraction of the corrupt government bailout and we'd have free, carbon-free electricity for everyone.
You want to take out your electric bill? Go LED lighting to reduce your demand dramatically - then a horizontal, roof mounted wind turbine that rides on magnets and accepts wind from all directions - at a much lower price than solar. And solar generates electricity 6 maybe 8 hours a day and only when it's sunny. This wind turbine generates all day - building up credits at night when electricity is cheap for the daytime heavy load.

Posted by: David on 26 May 09

Great article with a lot of research. I especially appreciate the map with KWh/day/meter^2 ratings (much higher than I thought). One question:

Has anybody done this net-net (energy in vs. energy out) for coal and oil to electricity. These technologies also require energy to mine, transport and convert to usable forms (ultimately electricity). Also the energy-resources that go into reclamation after extraction, the amount of energy used to construct power-plant, rail-tracks, and other infrastructure all leans against the break-even payback. It would be helpful when doing these analysis for green tech to have a baseline against which to compare.

Obviously, there are other issues that make these technologies more desirable. Imo we should be working to get these things on everybody's roofs. But when doing the analysis, we should probably keep those qualitative issues separate, as each individual will assign his or her own level of importance to those.


Posted by: Stosh on 29 May 09

CSP is simpler and cheaper to implement on a large scale. A 600km square grid in Arizona could power the whole US. And the heat can be stored in tanks and used to generate electricity at night.

Posted by: Robert on 29 May 09

Seems to be a lot of industry cheerleaders here.

While I'm sure we will be happy to spend the world's mineral and energy wealth building an alternative we can merely plug into the space left by fossil fuel, I don't think it is wise.

This obsession the technorati have with continuing this poisonous paradigm, this happy motoring fiesta no matter the knock on costs to the environment, will no doubt result in the demise of many more species in the on-going extinction event, perhaps even our own.

The problem is rather much more simple than the EROEI argument presented as done and done. Even if we could blink into existence fusion reactors running and connected to a brand new grid, it does not answer the problem of failing oceans, dying farmlands, fresh water shortages, and rising populations.

You cannot engineer new bio-diverse land, resuscitate extinct species, restore the plankton, or create fresh water without creating new energy and waste sinks. Green means sustainable, not eating the earth's heart out.

Here, quoting from the Energy Bulletin, we see in the groundbreaking book, "Small is Beautiful," that "Schumacher drew a hard distinction between primary goods and secondary goods. The latter of these includes everything dealt with by conventional economics: the goods and services produced by human labor and exchanged among human beings. The former includes all those things necessary for human life and economic activity that are produced not by human beings, but by nature."

If we burn up those materials provided by nature which are not renewable, metals for example, we have burned up the capital. Like a business that relies upon its capital to support spending that exceeds its income, we are burning up the capital of the earth.

PV lasts for a little while, then it goes bad. Meanwhile the planet is degraded, the population grows, the real problem goes unacknowledged, the supposition that man's engineering prowess trumps billions of years of evolution, which is nothing less than the constant tinkering and development of the perfect biosphere for life.

I think that we need only point to a few stunning pieces of technological hubris to make that point: Chernobyl, Bopal, asbestos, Love Canal, depleted uranium, the leachate lakes from former mines, etc. I could list a couple of thousand more without breaking a sweat.

Well, I've had my say. I hope the censors don't scrub this away. I will post it on other sites with links to your site noting the comment policy. Nothing like unfettered free speech discussing the scientific merits of an idea in order to prevent catastrophe. Let's just hope the industry people don't subvert the process in order to make a few bucks.

This is simple physics. Not doomer at all. Just reality.

Posted by: Cherenkov on 29 May 09

Amen to Cherenkov post. Personally, i suspect the energy cost is not all in, just a comparison of a panel input vs output. The Goethe post waxes eloquent but methinks, ignores all the ancilliary costs such as mining, supporting steel, replacements as amorphous output degrades, etc. These are exciting times enthusiasts but before public policy mandates solutions such as the recent attempt at Copenhagen, the solutions must be in hand, or risk alienating potential democratic (small d) supporters. I worry that green fanatics may usurp democratic processes eg clause 38 of draft Copenhagen treaty, to force unworkable and unsustainable solutions too early.

My guess is that only solar shingles will turn out to be cost effective, assuming very long life and some solution to solve all those connections that may deteriorate in wet weather. How would you repair such a roof?

I also think we ignore more simple solutions. I have built 3 solar collectors that paid for themselves in less than a year and did not cost the many thousands that all the sexy models cost. The trouble with all the hype over such expensive solutions is that they ignore the political effect of forcing poor people to subsidize rich peoples toys. Ya maybe have to mull that over for awhile but it will become clear once you follow the money.

Posted by: canbyte on 25 Dec 09

Post A Comment

Please note that comments will remain open for only 14 days after the article is posted. While previous comments will remain visible, attempts to post new comments after this period will fail. This helps stop comment spam, so your forebearance is appreciated.

The Worldchanging comments are meant to be used for further exploration and evaluation of the ideas covered in our posts. Please note that, while constructive disagreement is fine, insults and abuse are not, and will result in the comment being deleted and a likely ban from commenting. We will also delete at will and without warning comments we believe are designed to disrupt a conversation rather than contribute to it. In short, we'll kill troll posts.

Finally, please note that comments which simply repost copyrighted works or commercial messages will be summarily deleted.

Yes No







MESSAGE (optional):

Search Worldchanging

Worldchanging Newsletter Get good news for a change —
Click here to sign up!


Website Design by Eben Design | Logo Design by Egg Hosting | Hosted by Amazon AWS | Problems with the site? Send email to tech /at/
Architecture for Humanity - all rights reserved except where otherwise indicated.

Find_us_on_facebook_badge.gif twitter-logo.jpg