Talk:Solar cell/Archive 5

Contradictory claims to efficiency record
In the "History" section it is claimed that

"Scientists at the U.S. Department of Energy's National Renewable Energy Laboratory (NREL) have set a world record in solar cell efficiency with a photovoltaic device that converts 40.8 percent of the light that hits it into electricity. This is the highest confirmed efficiency of any photovoltaic device to date." REFERENCE 7: http://www.nrel.gov/news/press/2008/625.html

Whereas in the section "Three generations of solar cells", under the heading "High efficiency solar cells", it is claimed that

"In July, 2007, a University of Delaware team set a new record of 42.8% efficiency." REFERENCE 15: http://www.udel.edu/PR/UDaily/2008/jul/solar072307.html

So there are two contradictory claims to the efficiency record. Any thoughts on how to improve this?

Liquidcentre (talk) 18:49, 8 December 2008 (UTC)


 * I found a third: "Yesterday, Spectrolab announced that its newest triple-junction solar cells had achieved the world record in efficiency, converting 41.6 percent of specially concentrated sunlight into electricity." "New solar-cell efficiency record set" Scientific American Aug 27, 2009--BruceGrubb (talk) 19:56, 29 October 2009 (UTC)

High Efficiency
I think this merger is necessary, as some people will want both peices of information. Mikesta178, 5 now —Preceding unsigned comment added by Mikesta178 (talk • contribs) 16:59, 5 November 2008 (UTC)

Generations
Stating that there are three generations violates the neutral point of view principles of wikipedia. It would be OK to say that there are different technologies. For the record, single junction silicon is the cost leader on a total installed cost basis, it also is more than 80% of the market, the most reliable and has the highest efficiency. This article seems to be written by people with a strong thin film agenda. Pushing specific companies like Nanosolar is especially troublesome.

The technology race is simply not over and it is too soon to say that something is the next generation if that unproven idea has not achieved success in the market. Most of the new ideas will never work. Remember when GaAs was predicted to be the technology of the future for semiconductors? It never happened, CMOS came along and silicon is still king.

The references to generations should be deleted and the article should be written with facts rather than thin film propaganda.

Cleanenergy (talk) 03:35, 8 November 2009 (UTC)
 * Please have a look at pag 7. about the fourth generation.I restored the section. Mion (talk) 10:22, 17 November 2009 (UTC)

I looked at it, it is a weak presentation most of the so called "Next generation" technologies will never become real products. Wikipedia should be about facts and honest information not speculative claims about the future. I reverted your edit Cleanenergy (talk) 04:16, 21 November 2009 (UTC)
 * Well, generation 3 is hardly about claims in the future, plz read Neutral point of view or discuss the matter on Neutral point of view/Noticeboard, Mion (talk) 22:46, 29 November 2009 (UTC)

There is an implicit value judgement in calling something 1st 2nd and 3rd. the implication is that the sucessive generation is more advanced. By which metric is thin film more advanced than crystaline silicon? Not cost, crystaline solar is the cost leader on a fully installed basis, not efficiency crystaline silicon is more efficient.

Multi junction solar cells have the world reccord for efficiency and so would have some claim to a next generation if their cost was not prohibitive. Truth is that multi junction cells are a specialty product for space and not neccesarely the future either. Rather than making a subjective statements I think the article should deal in facts.

I do appreciate you care enough to look at this however.

Cleanenergy (talk) 04:02, 30 November 2009 (UTC)
 * As science sets it, The new record also inches the UNSW team closer to the 29 per cent theoretical maximum efficiency possible for first-generation silicon photovoltaic cellsHighest Silicon Solar Cell Efficiency Ever Reached. It would be silly to remove established classifications, and dont remove the section as long as it is under discussion. Mion (talk) 12:44, 18 December 2009 (UTC)


 * I am trying to see your point of view, I certainly agree that your reference is worthwhile and added it to the reccord efficiency section. However surely you see that calling something a second generation implies that it is more advanced than the first generation. So the fact that it is silicon that continious to set all the reccords is further proof that this is the wrong terminology. It simply violates a neutral point of view when you claim one technology is more advanced than another technology especially if by every scientific measure the silicon (what you want to call the "Old" technology) is more advanced.

Silicon has * Lower weight - Silicon modules only use one sheet of glass. Thin film uses two panes of glass thus double the weight per area and 3X the weight per Watt. * Higher efficiency - Si is 50% higher than CdTe the only thin film in real production. * Lower cost - On a fully installed basis silicon is 30 US cent per Watt cheaper than CdTe.

Also the so called newer generations have been around for a long long time without gaining market share so calling them newer is not presenting a neutral point of view.

Cleanenergy (talk) 23:02, 22 December 2009 (UTC)
 * It is not relevant what you or i think about this classification, that would be OR, The fact that silicon has a higher efficiency is well stated in the article multiple times, that is has a big marketshare etc, so I don't agree with you that silicon is handled as old technology on the contrary. And sure it has a good trackrecord. Mion (talk) 23:10, 22 December 2009 (UTC)
 * Good to see grid parity getting closer, waiting for a payback time of 1 month :) Mion (talk) 23:14, 22 December 2009 (UTC)

Revert by Mion
Mion, you reverted a section on Generations of solar cells to a previous version, which was marked for citing of references. Not only is the information you reverted to original research and opinion but it is also outdated and outright wrong in some instances.

"The advent of thin-film technology contributed to a prediction of greatly reduced costs for thin film solar cells that has yet to be achieved"

"Third-generation photovoltaics are proposed to be very different from the previous semiconductor devices as they do not rely on a traditional p-n junction to separate photogenerated charge carriers."

I have reverted back, these are my reasons. If you object perhaps you could layout some reasons of your own. GG (talk) 00:38, 21 June 2008 (UTC)
 * Well, i think the original research comes from an exhibit article with "we are presenting news" (IBM), secondly wurth solar is into CIS not CIGS, not mentioning nanosolar (why?), i understand not mentioning miasole and what happend to "these new devices include photoelectrochemical cells, polymer solar cells, nanocrystal solar cells, Dye-sensitized solar cells". Dye-sensitized solar cells are in production. I think we can work this out. Mion (talk) 10:38, 21 June 2008 (UTC)
 * I removed the part about reseach and intentions from IBM, first there is a lot of market buzz, they all have intentions and currently nanosolar seems to go for 14,7 % (NREL), so what about older technologies 6-12% ? Mion (talk) 11:07, 21 June 2008 (UTC)


 * Yes, I didn't post about IBM venture or the older techs so I'll let you get away with that but... CIS is just an abbreviation for CIGS, its really the same thing. While Nanosolar have installed 430 MW of capacity they produced nothing in 2007, Wurth made up 70% of CIS sales last year so has really been the first to get mass production working well. But like I wrote, their production line did nothing for seven years before they got the factory working, 2008 will tell if Nanosolar got it right.


 * I was trying to avoid listing all the different types of technologies in the history as there is a huge section on this page about alternative PV technologies, but it is certainly relevant if you want to. GG (talk) 11:25, 21 June 2008 (UTC)
 * ok. i searched the site of photon for the article, but can't find it, it's only in the printed version ? or do you have a direct link ? would be interesting to read it. Mion (talk) 11:50, 21 June 2008 (UTC)


 * Yes its a great article but I only have the magazine in print. Its a very credible source, maybe one day I'll get around to doing them a little bio. GG (talk) 12:02, 21 June 2008 (UTC)
 * GGByte, i oppose the fact of bringing in Wurth solar again, first if you want to mention CIS, it should be Showa Shell Sekiyu 20 MW in production and 60 MW upcoming, another reason for choosing nanosolar is that there is no indication that Wurth or Showa Shell Sekiyu are getting close to the magical $1 per watt barrier and nanosolar has the production facilities up and running. Cheers. Mion (talk) 11:45, 23 June 2008 (UTC)

Agreed Nanosolar and Shell are both building new CIS plants, as is IBM but the point is that none of these companies produced anything in 2007, Shell capacity is 20 MW but production is 0 MW, Nanosolar capacity 430 MW, production 0MW, Wurth capacity 15MW, production 15 MW, with 30MW upgrade planned for this year. Producing photovoltaics on a production line is not an easy thing, to give you a sense of scale Q-Cells the worlds largest producer in 2007 produced 389.2 MW. Nanosolar cannot go from 0 to 430 MW in the space of one year, maybe (and it is a big maybe) they do 30MW this year, 100 the next and so on. As it stands the article is misleading. Historically manufacturers will upgrade factories to meet demand, Nanosolar is not the first to build a big factory and run it at 'experimental' capacity for a few years(ie Solland, Schott, Wurth). This is the point I was trying to illustrate, Wurth ran at experimental capacity (1 MW) for seven years before ramping up, there are several reasons for doing this, to name a few: I read the article about nanosolar, i don't know very much about them (as technically they are not a manufacturer yet). Wurth have been developing the vacuum sealed CIGS technology and I know this has been a big barrier to overcome, if Nanosolar have found a way around the issue then it probably won't take seven years, but it won't happen overnight either. It seems Wurth have mastered this process now and so they may be equal rivals in the future, Wurth upgrading capacity to meet demand and Nanosolar overcoming growing pains as they experiment with their new factory. The point is Wurth commercialised this technology first (and it wasn't easy) which was the point I was trying to make.
 * Costs money to run a factory
 * Costs money to produce product that doesn't work, or insufficient quality
 * Loose customers when you can't fill orders

Overall if there are objections to Wurth's presence in the article, it is not necessary but it shouldn't be replace by wrong information. GG (talk) 01:05, 24 June 2008 (UTC)
 * sorry, bizzy at the moment, i'll get you a proper answer to that in the weekend. CheersMion (talk) 18:46, 25 June 2008 (UTC)
 * Hi GG, I moved three generation out of history as a classification is not per def the same as a history, Wurth is back, maybe the 30 mw update for 2008 should also be mentioned, as well as Global Solar which opened 40mw on CIGS??Cheers Mion (talk) 00:08, 4 July 2008 (UTC)
 * "Historically manufacturers will upgrade factories to meet demand" its not until 2010 that production capacity will meet demand. Mion (talk) 00:17, 4 July 2008 (UTC)

First generation solar not dead yet!
I was amazed to read the oppinion that "... with high production costs are unlikely to achieve cost parity with fossil fuel energy generation". The theoretical efficiency keeps increasing, costs keep decreasing, and the overall installed system cost is always a consideration. Second generation cells are far from acheving widespread market success for one very good reason- Efficiency!! -You need more installed area per watt- and it all costs money- The frames, supports, wiring, glass, etc, etc. Furthermore, with the cost of fossil energy increasing and the cost of first generation solar cells decreasing due to new supplies of cheap silicon, it's a bit rich to make such a bold prediction. I have made appropriate modifications to the text. AD —Preceding unsigned comment added by Adinov (talk • contribs) 10:00, 1 July 2008 (UTC)
 * I don't really agree, the first generation has a different production method which partly explains the high cost, and please reference the statements you make. Mion (talk) 10:31, 1 July 2008 (UTC)
 * Good reference for the efficiency, thanks, i think we have something about it Multijunction photovoltaic cell. Cheers Mion (talk) 11:46, 1 July 2008 (UTC)
 * Since First Generation Solar cells achieve such high efficiencies, their use is usually cost effective where the cost of space, glass, and other materials is considered. ? reference? pointing to the most expensive solar cells on earth so maybe "where space is an issue and cost is irrelevant (in space) ? Mion (talk) 12:41, 1 July 2008 (UTC)

I'm not sure I like these changes either. The bulk material and the energy required to purify the material are the main barriers to cheap first generation solar cells and not much can be done about it. Multi junction cells are third generation devices and are discussed in their relevant section, some estimates put the limiting efficiency of multi junction devices at 85% so comparing the limiting efficiency of a single junction silicon device to that of GaInP/GaInAs/Ge multi junction cells is not at all fair. No first generation isn't dead but it is unlikely that it will reach cost parity with fossil fuels, hence the investment into second and third generation. GG (talk) 07:19, 2 July 2008 (UTC)
 * ok, i think we agree GG, Squirmymcphee, and me, that the former state was better for the first generation, so ill set the first generation back as it was.Mion (talk) 22:50, 3 July 2008 (UTC)

The last clause in this section claims that first generation cells "achieve cost parity with fossil fuel energy generation after a payback period of 5-7 years." At the very least this need clarifying - it may be true in Hawaii, but not anywhere else as far as I'm aware...? Please correct me if I'm wrong. Reinforcing my suspicion that this claim is spurious is the reference: [9], a US DOE document about energy payback that makes no mention of generation costs. If you have a reference that does show this result, a similar one, I'd love to see it! --122.160.208.114 (talk) 06:59, 6 October 2008 (UTC)


 * The NREL link provided for the 5-7 year payback period claim does not include installation costs. In fact, the payback period is much longer (12.9 years here for a residential system in a sunny location with lots of government incentives).  Also, the payback period is mentioned twice in the article (once in First generation and once in Solar cells and energy payback) with the same reference, yet the payback period is stated differently in those two sections; this needs to be corrected.  Also, I've never heard of any real project expecting to get payback with flat PV in less than 10 years (after factoring in all relevant factors including installation cost, rebates, taxes, etc); if someone has a reference to such a project, I'd like to see it.  Bjp716 (talk) 18:15, 6 November 2008 (UTC)


 * Google claim payback of costs in 7.5years 149.117.65.27 (talk) 10:42, 29 May 2009 (UTC)

Three generations of solar cells
I have a few thoughts on this section. I haven't edited anything there because my thoughts aren't terribly coherent at the moment, so instead I'll toss them out there and try to come back in a few days.

First generation: The "relatively high energy and labor inputs" passage most definitely needs to be referenced if it is to stay -- for crystalline silicon, at least, labor inputs are quite low (around 10% of the cost of a PV module). Energy costs are on par with labor. Raw material costs are some 60-70% of the cost of a finished module. I can provide references for those figures if need be. You also have to note that this is purely a matter of accounting -- when people throw these figures around, purified silicon and purified indium (for example) each count as raw materials. If the point is to assess the total cost of energy and labor in producing PV modules from ore then that's an entirely different animal.

Also, the increased sophistication of first-generation solar cells has not led to a rethinking of theoretical efficiency limits. The 33% figure cited is applicable only to single-junction silicon, and that is all that it has ever been applicable to. The higher limits for multijunction cells, as well as theoretical limits for single-junction cells of other materials, have been well known for quite a long time. This sentence makes an apples-to-oranges comparison so that it looks like scientists have discovered some magic trick to exceed thermodynamic limits. The have not.

Second generation: Processes like vapor deposition can reduce high-temperature processing, but most of them also require very high vacuum. This is expensive and often slow, plus it forces the plant into more expensive batch processing (as opposed to continuous processing). That's not to say that it doesn't have advantages or that it is never used -- most crystalline silicon solar cells use vapor deposition to apply antireflection coatings -- but it is not fundamentally low-cost.

I have a hard time accepting amorphous silicon as a second-generation technology, at least in its current state, considering that it was invented and commercialized about the same time as crystalline silicon and had 30% market share by 1990 (which obviously has dropped, in part because of scaling issues and high capital costs related to vapor deposition processes).

I think it is difficult to justify the statement that "there is certainly a trend" toward second generation technologies by major manufacturers: Amorphous silicon is all produced by the same companies that have always produced it, there's virtually no CIGS on the market, and CdTe only became available in significant quantities about 5 years ago and is almost all produced by a single manufacturer. And Nanosolar most definitely did not produce 430 MW in 2007 -- it produced a plant that it says has a capacity of 430 MW. Last I heard, I think they were aiming for 100 MW in 2008.

It has been awhile since I read Martin Green's seminal paper on the three generations of photovoltaics, but I don't recall it being divided so sharply as this article on the basis of underlying material system. I'll have to have another look....--Squirmymcphee (talk) 22:32, 3 July 2008 (UTC)

Fair comments. Yes I was referencing both energy requirements in purifying silicon and high temp processing, since more silicon (mass) is used in wafer more energy is required. However not so much the cost of energy but rather energy payback times, perhaps not clearly expressed. |"First generation refers to high quality and hence low defect single crystal photovoltaic devices" this definition says that amorphous silicon is second generation but it also says that multicrystalline wafers are too. If you can find a better definition I am happy to discuss this. I believe that it is commonly anticipated that second generation will eventually replace first, which in turn will be replaced by third generation. My statement infers that this transition is happening at present which is a crystal ball prediction and hard to prove. But I believe it is valid, Q-Cells have acquired several thin film assets and while their main workhorse remains the wafer it should be recognised as a trend, the success of First Solar and the ambitions of Nanosolar support this too. GG (talk) 00:55, 4 July 2008 (UTC)

3 generations of solar cells
I don't think it is appropriate to place so much emphasis on space application solar cells. They represent a very small share of the market and are really not relevant in the discussion to the evolution of solar cells. I believe a few errors have been made with these recent edits (see discussions above) particularly relating to the presence of multijunction devices in 2nd generation (they are referenced as third generation see http://www.pv.unsw.edu.au/Research/3gp.asp). There is an appropriate section for the discussion of extraterrestrial solar cells, these comments should be placed there. Also comments of cost parity have nothing to do with a 5-7 year energy payback, the cost of PV devices over their life do not generate electricity to cover their expense. GG (talk) 01:29, 12 July 2008 (UTC)

Merger discussion
A merger discussion has been started regarding the organisation of Photovoltaics, Photovoltaic array, Solar panel, Photovoltaic module, Photovoltaic system and Solar cell. If you would like to contribute to this discussion please click here. GG (talk) 08:20, 25 July 2008 (UTC)

rollability
The introduction to this "solar cell" article currently states: "Crystalline silicon will never give tightly rollable devices let alone transparent ones". If rollability *is* relevant, why is it never mentioned again in the rest of the article? If rollability is *not* relevant, why bother bringing it up in the introduction? --68.0.124.33 (talk) 05:57, 27 July 2008 (UTC)


 * Some believe that roll-to-roll manufacturing processes contribute to reduced manufacturing cost, though that has yet to be proven. And the idea of a transparent solar cell is laughable, since a transparent medium by definition absorbs no light. The reference on the passage appeared to be linkspam -- it went directly to a page advertising a rather expensive report on thin film solar cells -- so I deleted the whole passage.--Squirmymcphee (talk) 15:45, 28 July 2008 (UTC)


 * No, no, no. It has nothing to do with manufacturing costs; The reason rollability is important is that it means that in the distant future, it will be trivial to install all over the roofs of buildings everywhere; it enables the installer to purchase a roll that can be easily transported to the roof and simply unrolled, like the way insulation is installed today.
 * Also, obviously where it applies to solar power "transparent" doesn't refer to all light. Absorption of either UV or IR could generate small amounts of energy compared to normal solar, but if a technology were developed that were very cheap and transparent, it could be overlaid on windows. —Preceding unsigned comment added by 70.162.242.234 (talk) 17:07, 4 October 2008 (UTC)


 * Um, yes, it has everything to do with manufacturing cost. It is true that some companies are promoting rollable solar panels in the way you describe, but in truth they are only practical on a limited number of surfaces. Anywhere a roof protrusion gets in the way, a rollable panel has no advantage over a standard module since it cannot be cut to go around the protrusion without destroying the panel. And you might think roof protrusions are not that common, but try talking to a solar installer or an architect -- the problem comes up a lot more often than you might realize. For that matter, try talking to manufacturers like United Solar -- the motivation for rollable panels was the anticipated reduction in manufacturing cost; other perceived benefits, like ease of installation, were secondary to their thinking.
 * As for "transparent" solar cells, what you say is true, but with an emphasis on the if. Only 3% of the solar spectrum is in the UV range, so the installed cost of UV-only solar cells would have to be an order of magnitude less than current solar cells to compete on an even economic playing field -- and that's assuming they can reach the same efficiency as a crystalline silicon solar cell. In reality, a semiconductor solar cell that absorbs only in the UV region would produce a high voltage, but almost no current, so you're talking about extremely low efficiencies. And assuming you can find a semiconductor material that absorbs IR without absorbing visible light, you'll have the opposite problem: high current with very little voltage (and thus also low efficiency). I won't say never, and perhaps somebody will invent such a transparent solar cell, but with respect to the article such a thing is pure speculation and is therefore not verifiable and should not to be included, particularly in the lede.--Squirmymcphee (talk) 21:44, 26 November 2008 (UTC)

4N
I would find it helpful if the article explained the term "4N gallium". I believe this refers to gallium that is 99.99% pure, and the terms 6N, 7N, etc. have similar meanings. I am not sure, however.--User:Mateat 18:39 July 29 (UTC)

Merge from High efficiency solar cells
I propose to merge the High efficiency solar cells article into this article.Beagel (talk) 15:40, 3 August 2008 (UTC)

See also Talk:Photovoltaic module. GG (talk) 06:10, 5 August 2008 (UTC)


 * I agree. There is nothing in the high-efficiency solar cells article that would be inappropriate in this one, and there is nothing notable enough in the high-efficiency article that it should stand on its own.--Squirmymcphee (talk) 15:10, 25 September 2008 (UTC)


 * Another vote for a merge here; when someone is ready to create an article about a highly efficient type of solar cell, it should probably be an article devoted to that type of solar cell, not all high efficiency solar cells. For example, dye sensitized solar cells might be worthy of an article of its own, but I wouldn't want it called high efficiency solar cells either. If it needs its own article someday, how about thin film solar cells?70.162.242.234 (talk) 17:13, 4 October 2008 (UTC)

Possible plagarism
The following text looks like it has been copied verbatim out of a book, since there is no "Fig. 1" in on the page: "The solar constant is the total integrated irradiance over the entire spectrum (the area under the curve in Fig. 1 plus the 3.7% at shorter and longer wavelengths." —Preceding unsigned comment added by 217.44.226.237 (talk) 14:16, 29 August 2008 (UTC)
 * Confirmed and removed .Mion (talk) 14:38, 29 August 2008 (UTC)
 * Came from Sun unit. Mion (talk) 14:55, 29 August 2008 (UTC)

hello
can anyone give me information about the advantage and disadvantage of solar cells, this is for my science lessons?? Thanksss —Preceding unsigned comment added by 212.219.188.3 (talk) 09:13, 16 September 2008 (UTC)
 * Wikipedia isn't a forum; talk pages exist for the purpose of discussing how to improve articles; they are not mere general discussion pages about the subject of the article, nor are they a helpdesk for obtaining instructions or technical assistance. If you wish to ask a specific question on a topic, Wikipedia has a Reference Desk, and questions should be asked there rather than on talk pages. —Disavian (talk/contribs) 22:49, 24 September 2008 (UTC)

Photovoltaic effect
Photovoltaic effect used to redirect to solar cell until I repointed it to photoelectric effect. Photovoltaic cell and photoelectric cell still point to solar cell. Any thoughts? --Rumping (talk) 12:59, 9 October 2008 (UTC)


 * Strictly speaking the photovoltaic and photoelectric effects are not quite the same thing, but it looks like this has been addressed by the editors of the article on the photoelectric effect. Seems fine to me.--Squirmymcphee (talk) 18:00, 19 October 2008 (UTC)

Cadmium telluride photovoltaics

 * Cadmium telluride photovoltaics, needs a cleanup. feel free. Mion (talk) 01:43, 10 October 2008 (UTC)


 * I'm not an expert in the field, but the statement  sounds very strange to me. Skimming the source given, I couldn't find it there. Danmichaelo (talk) 15:25, 3 November 2008 (UTC)

Solar cell equations
I added a new section on the solar cell characteristic equation to replace the section that had been there previously. The equations I deleted were actually a system for empirically calculating mathematical coefficients from measurement data to estimate how a particular PV cell or module will respond under changing weather conditions. The coefficients are purely mathematical, however, and do not correspond to anything physical. The equation I replaced them with is based on fundamental physics and has much greater instructional value and is the one that appears in every solar cell textbook on the planet. I am concerned I may have been too wordy or provided too much detail for an encyclopedia entry, though, so I would be interested in hearing people's thoughts (or seeing their edits) on that. I am also interested in comments on the graphics that I added, particularly the labeling of the curves in the charts -- I feel like they're pretty decent, but could still use some improvement. And I just realized I forgot to leave comments on my edits -- sorry!--Squirmymcphee (talk) 18:11, 19 October 2008 (UTC)


 * Some of the variable names contrast with those used previously in the text. For example, it is not explicitly clear what VOC and ISC refer to. PowerWill500 (talk) 11:51, 7 May 2009 (UTC)


 * Thanks for noticing -- took care of it. Any others you caught?--Squirmymcphee (talk) 13:42, 31 May 2009 (UTC)

Unverified 33% theoretical efficiency for silicon PV?
This article gives the theoretical efficiency as 33% and references a paper by Martin Green. I read his paper quickly and it seems to suggest a theoretical efficiency of 31%, referencing a paper by Shockley. Carts-Powell also cites the value 31%, referencing back to the same paper by Green.

I then quickly read the Shockley paper that Green referenced and it actually seems to suggest a theoretical efficiency of 30%, not 31%. The paper from Shockley and Queisser is from 1960 and it doesn't aim to be definitive. There is probably a better modern reference to use. Until someone supplies a better value (with reference), I would suggest using 30% not 33%. —Preceding unsigned comment added by 124.169.232.19 (talk) 12:51, 31 October 2008 (UTC)


 * The Shockley and Queisser paper is about as definitive as they come (though that's not to say it's the final word on the subject). You see different numbers because the exact value of theoretical maximum efficiency that they calculated is dependent upon the intensity and spectrum of the incident light. Assume black body radiation with a temperature near that of the sun and you get one value; assume the AM1.5 spectrum and you get another; assume concentrated sunlight and you get still another. Off the top of my head I want to say that 33% is the value that goes with AM1.5 illumination (32.8% is the number that pops to mind, actually) and that it is a bit lower with blackbody radiation (which is probably the assumption used by Shockley and Queisser), but I will have to check on that later. The Shockley-Quiesser limit is a true theoretical limit -- it uses very few assumptions and takes very little accounting of the practical physical limitations of silicon. Therefore, it is pretty well immune to advances in technology (as a theoretical limit should be) and there is not much room to improve it. Nonetheless, there are a lot of publications that do attempt to improve it, though generally by taking account of technology-controlled physical limitations of silicon (and other materials).--Squirmymcphee (talk) 23:31, 3 November 2008 (UTC)

Contrary?
I removed the below from page, as it doesn't really *say* anything -- it says "energy use in the production of solar cells is a concern", then it goes on to say "Hang on, let me work out the numbers here, hmm,  no its not a problem after all!". This is a bit silly to include in the section, perhaps one sentence saying "Utilising energy production from fossil fuels to transfer from fossil to solar has been deemed to be of little concern . Also why are we referring to  a web page, when the author has a journal article? User A1 (talk) 23:18, 3 November 2008 (UTC)

''A related recent concern is energy cannibalism where technologies such as solar cells can have a limited growth rate if climate neutrality is demanded. Many energy technologies, such as solar cells are capable replacing significant volumes of fossil fuels. Unfortunately, neither the enormous scale of the current fossil fuel energy system nor the necessary growth rate of these technologies is well understood within the limits imposed by the net energy produced for a growing industry. This technical limitation is known as energy cannibalism and refers to an effect where rapid growth of an entire energy producing or energy efficiency industry creates a need for energy that uses (or cannibalizes) the energy of existing power plants or production plants. In the case of solar cells, the payback time is low, which enables a fast climate neutral growth rate. ''

"Three generations": Biased POV
I think the article should be worded so that the classification of solar cells into three generations is presented as an opinion, not a fact (albeit a well-known opinion). There's no way to use the terms "1st generation", "2nd generation", "3rd generation", without strongly implying that the future will be 3rd generation, not 1st or 2nd generation. It's no coincidence that the person who invented this terminology is working on 3rd-generation solar cells! Of course a lot of people feel that the first or second generation is ultimately going to triumph, and the third generation is a silly sidetrack. Anyway, I think this article should try to stay neutral, by not using the "generation" nomenclature at all, except as an attributed opinion within the text. Anyone object? :-) --Steve (talk) 07:03, 3 January 2009 (UTC)

Small change to references.
It seems that reference number 72 no longer hyperlinks to the article on the transparent ultraviolet solar cell. Due to restructuring of that website, the new link to that article seems to have been changed to .

Perhaps reference 72 might be changed to read:

"Transparent Ultraviolet Solar Cell Developed at Japan's National Institute of Advanced Industrial Science and Technology (AIST)",  —Preceding unsigned comment added by Jkollar (talk • contribs) 06:14, 21 January 2009 (UTC)

Shadows on a cell
I have heard that for some technologies, if a very narrow shadow falls on a cell, covering maybe just 5pct, then the entire cells output falls to the effect of the shadowed area. Is this true? And if yes, for which types of solar panels does it apply. Velle (talk) 07:23, 7 April 2009 (UTC)


 * I think, whether you realize it or not, you are talking about two separate things. If you shade a solar cell its output falls more or less in proportion to the amount of the cell that is shaded, at least until you have shaded quite a large portion of it. This is more or less true for all solar cell types. Where you encounter problems is when you interconnect the solar cells as in a PV module or array. There is a brief description of the issue in photovoltaic array, but it doesn't really cover the technical details. Basically what happens is this: When you have a cell (or module) in an electrically connected string of cells (or modules), a small shadow can cause a shift in voltage such that the other cells (or modules) dissipate their power in the components affected by shading. When this happens, a portion of the electricity generated by the other cells is dissipated in the elements affected by shading instead of being supplied to you to do something useful. Adding strategically placed diodes to the PV module can cause shaded cells to be bypassed when this happens, so the problem can be mitigated. It is still a problem, though, and it can happen with a very small shadow. Crystalline silicon cells are more susceptible to it than most thin films, but all solar cell technologies are affected to one degree or another.
 * This really ought to be covered in the photovoltaic module article. Perhaps I will write something up at some point....--Squirmymcphee (talk) 10:54, 11 April 2009 (UTC)


 * I think the power optimizer article is talking about the same thing. --68.0.124.33 (talk) 02:02, 10 December 2009 (UTC)

Graetzel cells and luminescent solar concentrators
Graetzel cells and and luminescent solar concentrators are nowhere mentioned. If they are a different kind of cell, please clarify in the article


 * Found it in article, see Low Cost Solar Cell section.

Updated section, needs to be lifted nonetheless, it looks like it is not a different type of cell which it actually is. Lift and add --Types-- section to article

Model circuit diagram.
Is the diode in the model in the correct direction? It seems to me that the diode shorts out the current source. —Preceding unsigned comment added by 199.3.21.130 (talk) 07:14, 26 April 2009 (UTC)


 * It's correct as per the source. And it makes sense, provided you realise that it's a model, that is, an equivalent circuit that, if it were made of 'ideal' (perfect) circuit elements, would show similar behaviours to a real solar cell with illumination. You never would try to make it out of real components: if you did it would be very wasteful of power generated by the current source, and the diode would get very hot in some circumstances. It's just a model to use to derive some equations from.  The maximum voltage it can generate is roughly the forward 'on' voltage of the diode, which is what you would expect. --Nigelj (talk) 11:04, 26 April 2009 (UTC)


 * Just to elaborate a bit on what Nigelj said, in the equivalent circuit model some of the current produced by the solar cell is diverted through the diode, some is diverted through the parallel resistor, and some is dissipated in the series resistor. The voltage across the output terminals is roughly equal to that across the diode (assuming the series resistance is small), and the current through the diode is governed by this voltage. When the solar cell is open circuit, the diode does indeed short out the current source and no current flows out of the cell. When the cell is short-circuited, the voltage across the diode is zero and no current at all flows through it. At voltages in between, the amount of current diverted through the diode varies. This is what gives the solar cell I-V curve its characteristic shape.
 * If the diode were reversed, electrons and holes in the solar cell could not be separated by the electric field at the p-n junction and no current could flow. Solar cell operation relies on the diode "shorting out" the current source to control the output voltage and current.
 * The equivalent circuit model is actually a pretty good representation of the physical structure of a solar cell. The semiconductor absorbs light and converts it electron-hole pairs -- this is the current source. The p-n junction built into the cell is the diode. The silicon and metal the current flows through have resistance, which is where the series resistor comes from. And the parallel resistor represents imperfections that provide resistive -- rather than diode-like -- current paths through the p-n junction exclusive of the diode.--Squirmymcphee (talk) 16:21, 26 April 2009 (UTC)

Cost-benefit Analysis
This is briefly mentioned, but perhaps a cost-benefit analysis of solar output could be presented with current and/or projected energy costs. This, after all, is the real factor in solar being a viable energy alternative. PowerWill500 (talk) 08:12, 9 June 2009 (UTC)


 * Easier said than done, I think, at least until somebody publishes a comprehensive study on the matter. It is impossible to fix a single number as the cost of PV and a single number as the cost of other energy sources -- they both depend as much on geography, politics, and local economics as they do any technological considerations. A cloudy place where energy prices are high is frequently a better place for PV, economically speaking, then a sunny place where energy prices are low -- but not always. Then there's the question of whether you're billed a flat rate or according to your time of use, or whether you are talking about generating electricity for a utility, all of which has its own energy cost/benefit considerations. Besides, I think such a discussion would be more appropriate for the photovoltaics or solar energy articles.--Squirmymcphee (talk) 21:10, 10 June 2009 (UTC)

There was a nice article on it in IEEE Spectrum. It takes and average solar cell about 2 years to generate and "pay back" the energy consumed in its production. Bottom line: Nothing is free! —Preceding unsigned comment added by 12.229.112.98 (talk) 17:34, 23 November 2009 (UTC)


 * It is true that there are a great number of variables to consider prior to purchasing a PV system. Most people have access to their energy bills to determine their current electricity costs.  One source that might benefit consumers on this matter is the following "neutral" website "Solarbuzz.com"  that provides links to participating international PV system manufacturers thereby allowing the consumer to price out a system and determine if this investment makes sense for them. The current trends are favorable for the consumer. JamAKiska (talk) 14:44, 3 December 2009 (UTC)


 * I think the original comment here was about economic payback, not energy payback. At least, that was my impression. Based on your comment, energy payback is what the IEEE Spectrum article covered (I did not see the article myself), and a 2-year payback time is certainly reasonable in many locales. However, economic payback -- that is, getting solar that is cheaper than grid power -- is far more difficult to achieve with currently available products. There are a few places where it is possible without subsidies, many more where subsidies make it economical (at least for the end-user), but in general, cost parity with grid power has not yet been attained. It's close, but not quite there yet.--Squirmymcphee (talk) 17:08, 6 December 2009 (UTC)

Rename

 * The following discussion is an archived discussion of the proposal. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section. 

The result of the proposal was Not moved. Splitting the article, undoing previous merge, etc. are reasonable editorial issues that don't seem to need admin assistance. Feel free to continue that discussion. DMacks (talk) 16:54, 1 July 2009 (UTC)

Solar cell → &mdash; rename the article as solar cell implies that the cells are used in all solar conversion panels. This is not so —Preceding unsigned comment added by KVDP (talk • contribs) 11:45, 19 June 2009 (UTC)
 * Oppose. Perhaps you meant "used only in solar panels?" Solar cell is the most commonly used name, by a ten to one margin. It is a technicality to say that a photovoltaic cell is not a solar cell, as I would guess that all photovoltaic cells would provide an output if placed in the path of solar "rays", even if they were intended to respond for example, to an LED laser, e.g. for intrusion detection. 199.125.109.99 (talk) 14:31, 19 June 2009 (UTC)
 * Comment: And you guess wrong. Some of the more sensitive types would simply be destroyed. Andrewa (talk) 19:33, 19 June 2009 (UTC)
 * Not if you put a filter in front of them. I do recall some sensors that count individual photos, but I do not recall if they produce a voltage output. 199.125.109.99 (talk) 02:18, 22 June 2009 (UTC)
 * That's a bit of a stretch. Surely you're not saying that they'd be acting as solar cells if you did this? No, a solar cell is a means of capturing energy for use, and it's the most newsworthy sort of photovoltaic cell just at the moment. But it's not the only sort. Andrewa (talk) 03:25, 23 June 2009 (UTC)
 * Support . The article as it stands is about all photovoltaic cells, and photovoltaic cell redirects here. A better solution would be to reverse the merge of photovoltaic cell with solar cell, they both deserve articles. But if we are to have only one aticle, it should be correctly named. Yes, most people call solar cells "solar cells", but that's not the topic of this article as it stands. Andrewa (talk) 19:33, 19 June 2009 (UTC)
 * The point is that it is the topic of this article, as the common name for any type of photovoltaic cell is solar cell. 199.125.109.99 (talk) 02:18, 22 June 2009 (UTC)
 * Disagree with that analysis. Obviously some people do, wrongly, assume that all photovoltaic cells are solar cells, because they don't know the other types exist. But it's a bit of a stretch to say that these people are therefore calling these other types solar cells. They don't call them anything. They don't know they exist. That's enough reaon in itself to have an article on photovoltaic cells, called photovoltaic cells, so they can easily find this information. Andrewa (talk) 03:25, 23 June 2009 (UTC)
 * I'm withdrawing my vote in view of the fact that there is no likelihood of consensus to move. See below. Andrewa (talk) 01:49, 29 June 2009 (UTC)


 * Opppose. Disagree that the term "solar cell" implies anything about solar panels, plus it is more common than "photovoltaic cell." However, I find the discussion above a bit confusing -- is this proposal about confusion of solar cells with solar panels or with photodetectors?--Squirmymcphee (talk) 18:42, 24 June 2009 (UTC)

Comment: Much of the discussion above seems to be making the point that solar cells are more commonly called solar cells rather than photovoltaic cells. But I don't think anyone is disputing this.

The problem is rather, not all photovoltaic cells are solar cells, and there is a great deal of encyclopedic information about these other cells that should be somewhere in wikipedia. Some of these were developed and in regular use long before solar cells, the selenium cells used at one stage in nearly all photographic light meters for example.

So where should we put this information about these other photovoltaic cells? Andrewa (talk) 01:39, 25 June 2009 (UTC)
 * Which type? Photodetector has a long list of types. Put them in whichever article is appropriate. 199.125.109.102 (talk) 04:46, 25 June 2009 (UTC)


 * The selenium cell used in my classic Weston exposure meter, for example? Oops, I see we don't have an article on Edward Weston or the Weston Electrical Instrument Company either! See http://www.diaxa.com/weston/thecompany.htm for a start.


 * The more I look at it, the more of a mess our coverage of this topic becomes. We have articles on photovoltaics, photovolaic array and photovoltaic module, largely repeating the same information. And yet we have no article on the basic component, the photovoltaic cell. This appears to be at least partly because many and perhaps most of the authors of this material were unaware that not all photovoltaic cells are solar cells.


 * And that's not a criticism of them. We are all learners. But it's time to fix this. Andrewa (talk) 10:02, 25 June 2009 (UTC)


 * I am afraid I don't understand the "fix" you are suggesting. Do you propose merging all of the articles on devices that use the photovoltaic effect into a single article? That, in my opinion, would be insanity -- much like merging the articles on diodes, transistors, and integrated circuits because they all operate on the same principles. While photosensitive devices may operate on similar physical principles, they are all constructed, packaged, and used in very different ways. For example, very little of the information on the photodiode page is useful in discussing solar cells because each of the device types is built and optimized for completely different purposes.--Squirmymcphee (talk) 16:04, 27 June 2009 (UTC)


 * Agree that the radical merge you describe would be insanity! I'm afraid I consider this a straw man.


 * There are several things that need fixing, but the most urgent IMO is still to correct the apparently widely and strongly held but mistaken supposition that photovoltaic cell is just another name for solar cell. It's a bit like redirecting home to house, which was also once proposed, or something like it. But not all people live in houses!


 * Another way of achieving this would be to redirect photovoltaic cell to the existing article at photovoltaics. I don't think this is the best solution, but it's an obvious alternative. That's the main reason for mentioning the other articles at this stage. The other is to flag that in the longer term some rationalisation of all this duplicaton would be good.


 * The lead to the photovoltaics article currently reads Photovoltaics (PV) is the field of technology and research related to the application of solar cells for energy by converting sun energy (sunlight, including sun ultra violet radiation) directly into electricity. If that is true (I'm sceptical, but English is not always logical so perhaps it may be) then the redirect would not work. Andrewa (talk) 20:50, 28 June 2009 (UTC)


 * Hmmm, not as bad as I thought... we do have an article at Edward Weston (chemist) on the Edward Weston in question, with a redlink to the company, but none yet for Edward Faraday Weston, his son who also worked on photoelectric devices. Andrewa (talk) 10:35, 25 June 2009 (UTC)


 * Oppose. This is the correct name for the infromation on solar cell. I do think a seperate photovoltaics article would be usefull and several item from this article could be used in that in a more generic way. This Solar cell article should than keep focussing on the solar electricity producing variant. Anybody could remove the redirect and start an article on Photovoltaics anytime using some of the infromation in this article as a basis. hAl (talk) 21:16, 28 June 2009 (UTC)
 * Ihere seems no likelihood of consensus to move, so I have withdrawn my support vote and let's move on. Do we, however, finally have consensus that there is an electronic device called a photovoltaic cell of which a solar cell is currently the most common example but which also occurs in other notable forms and uses? Andrewa (talk) 01:49, 29 June 2009 (UTC)
 * The above discussion is preserved as an archive of the proposal. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.

Split/rewrite/distinguish

 * spawned out of the above discussion...

Do we, however, finally have consensus that there is an electronic device called a photovoltaic cell of which a solar cell is currently the most common example but which also occurs in other notable forms and uses? Andrewa (talk) 01:49, 29 June 2009 (UTC)

In the currently very newsworthy context of energy sources, photovoltaic cell and solar cell are indeed synonymous. This appears to have led to a widespread belief that the terms are synonymous in general. Many websites give definitions of photovoltaic cell reflecting this; Most but not all of them are in some way connected to solar energy.

But we are a general encyclopedia, and do need an article on photovoltaic cell in the more general sense, and in time I'll create one if nobody beats me to it. Question: What should it be called? Andrewa (talk) 01:59, 29 June 2009 (UTC)


 * Photovoltaic cell would seem appropriate enough. hAl (talk) 11:42, 29 June 2009 (UTC)


 * To me, that's the obvious name. My only reservation is that this would be reversing a previous merge. As such it's good to discuss it as fully as possible first. Andrewa (talk) 21:15, 29 June 2009 (UTC)


 * That was a very crappy little article. If you would create a crappy article like it used to be then I would suggest you do not bother. If you make make an article that differentiates between the use in Photovoltaics in different types of cells then I think it will be fine. hAl (talk) 21:30, 29 June 2009 (UTC)


 * A good stub would be far preferable to the current misleading redirect, which simply supports what is evidently a common error. But just for this reason, I think you're quite right. So the thing to do is probably to work up a reasonably substantial article in my user space, and then propose a move at WP:RM to the article namespace. By going through this formal process, when one of the many whose knowledge of the subject is restricted to solar power applications quite rightly jumps in and restores the error, it's more easily handled. Andrewa (talk) 17:00, 30 June 2009 (UTC)


 * Another possibility is to create an article at photovoltaic cell (electronics) or using some similar disambiguator, and put a hatnote into the solar cell article to catch those who are interested in other types of photovoltaic cells. Andrewa (talk) 10:34, 11 July 2009 (UTC)


 * I would prefer a seperate Photovoltaic cell article. hAl (talk) 18:15, 12 July 2009 (UTC)


 * We can always go back to WP:RM and apply to move the article fom the disambiguated name back over the redirect. Andrewa (talk) 06:53, 13 July 2009 (UTC)

See User:Andrewa/photovoltaic cell. There's lots to do...! Andrewa (talk) 06:53, 13 July 2009 (UTC)
 * I have little expertise in this area but I have added an article lead for you in your draft. Hope that it helps a bit. hAl (talk) 12:18, 13 July 2009 (UTC)


 * Never too late to learn! I'm not professionally qualifed (unless you count first year physics at university) but my father (an electronics engineer among other things) gave me my first soldering iron when I was seven years old so I'd stop borrowing his. So electronics has been a hobby for more than forty years and I've probably learned something along the way.


 * And I've just discovered tha my father's first published paper was on a photoelectric (not photovoltaic) exposure meter for metalographic photomicrography (whew)... he's right now looking for me for a photograph of the vacuum tube photocell that it used.


 * Probably the first thing to do IMO is a rewrite of the leads to both photoelectic effect and photovoltaic effect. These seem to be accurate (and are actually exactly what I thought I remembered, but I had to do some research to confirm this, many websites... even NASA!... seem to use the terms rather loosely) but they are not very clear. Andrewa (talk) 00:50, 14 July 2009 (UTC)

Metalization lines
Making the metalization lines — which export the energy from the cell — smaller, cheaper and more efficient increases the efficiency of multicrystalline solar cells --Nopetro (talk) 13:04, 22 July 2009 (UTC)

Best Research Cell Efficiencies chart
I assist Dr. Larry Kazmerski at NREL in updating the Cell Efficiency Chart that is in the Comparison of Efficiencies section. When I get a new version with updated data, what is the easiest way for me to replace the old chart with this new one?

SolarVision (talk) 19:29, 3 September 2009 (UTC)
 * Make sure that your chart has a similar quantity of data, and then upload it using the upload file link on the left. ONce done, simply change the wiki code to refer to your new image User A1 (talk) 23:00, 29 October 2009 (UTC)

Improvements needed
This article is not very well written and needs to be improved. The first half is decent but the second part is mediocre. The sections from "UV solar cells" to "China" are written in a bad style. Some sentences do not even obey the subject-verb-object form. Since I was here, I made several small corrections (particularly on hyphenation and capitalization). Hopefully people will contribute to make this article a little better.

ICE77 (talk) 06:50, 28 September 2009 (UTC)


 * You're absolutely right. I read down as far as 'Antireflective and all-angle coating' and immediately found the article talking nonsense - muddling 'reflective' with 'anti-reflective' straight away.  Then I followed the two refs in that paragraph: neither of them mention anything about solar cell technologies or any relevant coatings for them. I haven't got time right now, but will try to get back to this too. --Nigelj (talk) 08:58, 28 September 2009 (UTC)

Dust often accumulates on the glass of solar panels seen here as black dots?
Enough said.Landroo (talk) 08:14, 2 October 2009 (UTC)

Are the energy payback times accurate?
The first reference listed from the US DOE does not seem to account for transportation and disposal costs. If I buy my panels from China because they are cheaper and transport them to the US, then the energy cost for that panel may be higher than locally purchased (or not). If the panels are trucked, then the energy used will be much greater than if they went by rail. If I install them on the west coast it may be less energy intensive to buy from China, but maybe German (Siemens) panels have a better $/watt purchase price, so the increased shipping cost can be justified. Dollar cost and energy payback don't match anyway. There are hidden subsidies in all manufactured goods and in the local cost of oil.

If large scale production continues, will aluminum from recycle streams be sufficient? The US DOE says it takes 20 times as much energy to make aluminum from ore as it does to recycle http://www.eia.doe.gov/emeu/mecs/iab/aluminum/page6.html That difference could change the payback time a lot.

I also see no account for loss between manufacture and end of expected life. I have personally had a new panel fail from breakage after installation, and I have bought from a local supplier who attempts repairs/rebuilds and buys same to keep the price down. Their house brand carries no UL/CSA/CEN mark, so those cannot legally be used for grid-tie, but there is still enough of an off-grid market to justify the effort.

When the panel is finally spent, how much energy will it take to legally dispose of it? Asbestos and creosote were once legal and common. Cadmium telluride may work well, but how many places will there be 20 years from now to properly recycle it? Both China and the EU refuse to allow it to be sold at home because of health/safety concerns. How far I have to send it for recycle makes a big difference. At work, we already have problems with the legal disposal of linear fluorescent tubes because the nearest trusted vendor is half a state away. Compact flourescents are easy to get rid of because a major home improvement chain will take those for free.

Tellurium and cadmium are leftovers from copper and zinc mining. Is that a plus or minus on the energy side? 68.215.36.159 (talk) 01:50, 22 November 2009 (UTC)David


 * You ask some good questions, many of which researchers are trying to address even now. However, since Wikipedia has a policy against publishing Wikipedia:No_original_research not all of them can be addressed in the article until they are published elsewhere. That said, on the subject of transportation energy, several organizations (including the UN) have ongoing programs to better quantify this. However, I once did some back-of-envelope calculations. As I recall, a large container ship uses some 500-1000 watts/ton at cruising speed (which accounts for the majority of its fuel consumption on a China-to-US haul). On a trip from Shanghai to San Francisco, that implies energy consumption of no more than 0.25 kWh/kg of gross weight -- let's call it 0.75 kWh/kg of cargo. A 200 watt Suntech panel weighs just under 17 kg and thus requires about 12.5 kWh to transport. In a sunny location, that energy will be recovered by the panel in under a week. Again, however, it is a back-of-the-envelope calculation, and I should add that my field is solar cells, not cargo ships....


 * There are a number of solar cell manufacturers looking into module recycling, and they are making good progress. However, since modules are a long-lived product and the vast majority of them are fairly new, there is not much of a waste stream to work with yet. That said, 70-90% of the energy that goes into a PV module goes to producing silicon wafers, and besides, manufacturers are trying to reduce the amount of aluminum they use. I agree that wherever they use aluminum they should choose recycled aluminum, but I do not think it will have a great impact on the energy payback time. I'm not sure how to work this into the article without violating the prohibition against original research, though.


 * Panels would have to fail in great numbers prior to reaching energy payback to significantly change the figures. Think about it: Assuming a 3-year energy payback time, if 5% of modules fail in the first 3 years, then some portion of their embodied energy must be spread over the remaining 95%. If we spread all of their embodied energy over the remaining modules, then the energy payback time increases by only 5%. It is impossible to know with any accuracy what the actual in-field failure rate is prior to energy payback, but if warranty returns are any indication then the figure is well below 5% -- if warranty returns were that high, they would put many manufacturers out of business.


 * As for disposal, that is part of the reason for starting recycling research now. Beyond that, in many places panels are required to be disposed of as e-waste (though in Europe I believe RoHS-compliant modules can be landfilled). The only major manufacturer of cadmium telluride PV panels, First Solar, accepts all of its modules back for recycling. Regardless, after much research cadmium telluride has not yet been shown to be toxic; that is different from saying that it is non-toxic, of course, but the point is that you should no more assume that it is toxic than assume water is explosive because it is made from hydrogen and oxygen. At any rate, you should bear in mind that linear fluorescents went decades without being marketed as a green product while CFLs were marketed that way from the get-go. Like the CFL industry, the PV industry has a vested interest in making its products easy to dispose of and/or recycle. I suppose some references to recycling research could be worked into the article, but it might be difficult to make them relate to energy payback period -- I'm not aware of any references that address both.


 * By the way, you are flat-out wrong about CdTe not being allowed in China and the EU. Cd is not allowed there, but that does not necessarily mean that CdTe is not. First Solar has a number of contracts to install CdTe panels in China. And not only does First Solar sell in Europe, along with numerous smaller European CdTe manufacturers, but there is a rather well known incident where a German chicken farm with CdTe modules on the roof burned down. During cleanup, the site was treated as a toxic waste site -- not because of the CdTe modules, but because of the chicken poo. As I recall, the panels were plucked out and cleaned up by the toxic waste crew, then returned to First Solar for recycling. There is a lack of clarity on how the EU will deal with CdTe in the future, but I'm betting it will eventually go on the RoHS-exempted list. I am less familiar with the situation in China, so I won't comment on it.--Squirmymcphee (talk) 23:18, 13 December 2009 (UTC)

I misread the wiki article on cadmium-telluride and didn't realize they made an exception for solar panels: "The approach to CdTe safety in the European Union and China is much more cautious: cadmium and cadmium compounds are considered as toxic carcinogens in EU whereas China regulations allow Cd products for export only. The major concern for CdTe is inevitable presence of Cd during CdTe production and processing.[7][8]"

Did the chicken farm panels have to be sent back to the US for recycle? That's the responsible thing to do, but costly in energy terms. If First Solar goes away as a company, then what happens to panels made by them when they need to be recycled?

There are still lots of nicad batteries around. Recycling them into thin film panels would be a nice way to get them out of the waste stream at least for 25-30 years.

I hope there is some easy way to know 30 years from now what a panel is made from. Not all thin-film is CadTel. I have a bunch of Harbor Freight panels that I assume are amorphous silicon. I live in south Florida. When (not if) we have another major hurricane there will be plenty of prematurely "failed" panels, some of which will never be found. The aluminum frames will be picked up for scrap, but the rest of a thin-film panel will be just a bunch of broken glass hauled away to a dump with all the other unidentified bits that are swept up afterward. The wind code is supposed to prevent that, but it never seems to work that way.

At work we replaced our T-12 75 watt 8' linear fluorescents with the 60 watt version in Feb. 1992 as part of a electric utility energy saving program. We've considered linear fluorescents to be the "green" choice at least as far as cost of operation goes, but the mercury in them makes them a legal liability. Compact fluorescents won't light a warehouse very well, and vapor arc lamps are also considered to contain "hazardous" materials as far as landfilling goes. I would very much like to see LED lighting to replace fluorescents, but LED's don't handle heat well, so they still won't work in high bays for lighting. 72.153.123.76 (talk) 02:18, 20 December 2009 (UTC)David


 * I do not know whether First Solar is stashing panels for recycling until they have enough to make them worthwhile to ship back to a plant, if they ship them immediately, or what. If First Solar is correct that CdTe modules must be accompanied by recycling programs to gain widespread acceptance, then the problem of who will be recycling CdTe modules in 30 years solves itself: Either somebody, be it First Solar or somebody else, will be around to recycle them, or CdTe will prove unpopular, unviable, unsafe, or otherwise undesirable and there will not be a significant number of the modules around. If CdTe does not require recycling programs to gain widespread acceptance, then a lot hinges on proving their safety (though questions about safety are why First Solar says the recycling programs are necessary...).


 * As for identifying what a panel is made of, that is generally not too difficult. If you know the manufacturer it is usually quite simple; if all you have is a cell to look at, an expert can usually tell just by looking at it. Before putting a material of unknown origin into a processing stream, though, you can bet a manufacturer or reprocessor would do a chemical analysis of some sort.--Squirmymcphee (talk) 14:41, 29 December 2009 (UTC)

Simple Demonstration of Solar Cell Technology
Can someone please draw a simple picture of how good current solar cell technology is by adding to the article something like this:

'Using mid-market off-the-shelf technology, you are half way between the equator and the north pole, at sea level, on a very calm, very clear day, at 20 Celsius, the sun is at its mid-day day-light peak, on June 21st (or there abouts), with an array of solar cells pointing square-on to the sun. How many square meters of array would be required to boil the water contained in a 3 kilo watt (13 amp fused) kettle, in the fastest time, without burning its fuse.' —Preceding unsigned comment added by 86.25.200.137 (talk) 11:56, 7 December 2009 (UTC)

(Sorry for not giving the above a headline, which should be 'Simple Demonstration of Solar Cell Technology', I should also have added how much in USDs or Euros, the array would cost per meter.) —Preceding unsigned comment added by 86.25.200.137 (talk) 12:10, 7 December 2009 (UTC)


 * I added the headline, as I think you wanted someone to do. With regard to your question, having lived with solar panels for some years on a boat, that isn't the way you would use solar panels. If you installed enough capacity to do this, you would be able to boil that kettle for many hours continuously every day throughout the year (obviously for more in summer), which is unlikely to be necessary. You would use some kind of storage capacity to hold the energy gained during all those hours, with smaller usages going on, so that when you wanted to boil it the minute or two that takes could be sustained. In the home, you use the grid for that 'storage' (or averaging). At sea in a boat, you use batteries. On my small boat, we don't try to use solar electricity to boil kettles, but I do have the capacity to use a 600 W vacuum cleaner when needed. It takes enough 'top' off the two batteries' voltage that the mains inverter trips off after two or three minutes, but, on a reasonably bright day, the batteries soon recover and you can use it again within about 20 minutes. This is enough for my needs when working or living aboard. When grid-connected at home, or with a bigger battery bank, life is simpler. Those two panels amount to about 150 Wp, which is very little by 'home' standards - you would think in terms of maybe 20, not 2, of them. —Preceding unsigned comment added by Nigelj  (talk • contribs)

Thanks for that. But I'm not trying to boil a kettle on a boat, or charge a battery. I'm trying to find a way to visualize the power of this technology in a 'practical' demonstration. The location I have given is simply an average one might be at (Washington or Bejing), and the conditions are the best possible one could get at this average location. The kettle is sited, simply because it's a device that is simple and consumes more power than most domestic appliances (bar cookers, etc.) do. I should also have added that the percentage efficency of the cells used should to be stated. And what would also be helpful is a Moore's Law type graph ploting the progress of this technology from, say the 1960s, to the present day. North American's tend not to use electric kettles for some strange reason, so maybe a more universal device should be given, what that could be, I don't know. —Preceding unsigned comment added by 86.25.200.93 (talk) 01:45, 13 December 2009 (UTC)

Metamorphic multijunction solar cell
What is this passage talking about?


 * For decades, conventional cells have featured wafers of semiconducting materials with similar crystalline structure. Their performance and cost effectiveness is constrained by growing the cells in an upright configuration. Meanwhile, the cells are rigid, heavy and thick with a bottom layer made of germanium.

It is not directly cited, and what is this talk about 'decades' of germanium solar cells? And 'growing' them in 'an upright configuration'? Has there been a whole technology going on that I know nothing about, or has someone made some mistakes here? FWIW, I believe that there have been decades of silicon solar cell technology, with crystalline ingots being grown, then sliced, polished and doped to 'grow' the cell junctions, but this is a very bad description of that. I can understand, but haven't verified, the rest of the section, about growing thin-film cells 'upside down', but this introduction makes me suspicious. --Nigelj (talk) 21:26, 31 March 2010 (UTC)


 * I suspect it refers to the growth of gallium arsenide and similar thin films on gallium substrates, which is necessary for mechanical reasons. It has been used for decades in cells destined for use in space, and indeed Emcore makes cells for use in space. I agree the passage is poorly worded, though.--Squirmymcphee (talk) 12:57, 4 April 2010 (UTC)


 * Well, if that's the case, it's not just poorly worded but downright wrong: germanium and gallium arsenide are entirely different semiconductors. have we muddled them up??! --Nigelj (talk) 16:05, 4 April 2010 (UTC)


 * Sorry, there is an error in my response. I should have said, "I suspect it refers to the growth of gallium arsenide ... on germanium substrates...". It looks like the text in the article has been changed since I last visited and in the current context the intent of your comment is not clear to me, so I will just say this: Gallium arsenide (and several other thin-film) solar cells are typically made on germanium substrates. The germanium does not form an active part of the solar cell, it is there only for mechanical stability; therefore, the cells are referred to as gallium arsenide solar cells, and it would be incorrect to call them germanium solar cells.--Squirmymcphee (talk) 12:23, 18 April 2010 (UTC)

Record efficiency claims
The 42.8% efficiency claim for Delaware State University is completely useless because, according to this article, it is an hypothetical efficiency. The solar cell hasn't been tested. The number shouldn't be in the article at all.

Spectrolab claims it has a multi-junction solar cell that produces an efficiency of 41.6% and that this record was set in 2009. The information can be found at http://www.spectrolab.com/history.htm. The page also claims that this figure has been verified by National Renewable Research Laboratory.

ICE77 (talk) 02:22, 12 April 2010 (UTC)

Most Solar Cells are subset of Photodiodes
I'd like to see somewhere on this page the admission that semiconductor solar cells are usually large-area Photodiodes, simply operated in the "photovoltaic" mode. This would communicate the usual physics of the device more succinctly to some people. The differences between silicon solar cells and the devices usually called photodiodes are merely differing optimizations. The solar-cell needs to have larger area, lower series resistance, high open-circuit voltage, and is allowed to be higher in leakage. The photodiode needs to have well-defined smaller area, higher speed, and lower leakage, while open-circuit voltage and series resistance are allowed to be less optimized. Differences only of degree. A modern triple-junction solar-cell is three photodiodes of three different materials layered vertically onto a single wafer. I acknowledge that wet-chemical solar-cells may not be exactly photodiodes, but they are not commonly used technology yet anyway.

jimswen (talk) 19:03, 21 April 2010 (UTC)

Solar cell manufacturing energy self sufficiency
Perhaps someone can comment on when the solar cell manufacturing industry will reach energy break-even in their own manfacturing plants. I would hope that the cost of solar cell manufacturing would be declining in part due to the industries ability to power its own needs by generating and consuming its own low cost solar energy. What percentage of the costs of making a solar panel is the energy cost to manufacture monocrystaline silicon wafers?

Is the US government providing financial incentives to encourage the expansion of silicon wafer production facilities that are energy self-sufficient, and if not, why not?

Dwight.klaus (talk) 07:55, 19 July 2010 (UTC)
 * I don't think it works (or will work) like that - i.e. isolated solar cell factories with their roofs covered in solar panels. The purification and production of the raw materials and their processing into the finished product are likely spread over several manufacturing facilities. Equally, the energy supply will become more 'renewable' when renewable energy production is fed into the grid (from wind, solar and other sources) over wide areas too, perhaps via continent-wide super-grids. --Nigelj (talk) 08:23, 19 July 2010 (UTC)


 * As I recall, the energy cost to manufacture monocrystalline silicon wafers for solar panels is on the order of $20/square meter. At current manufacturing costs, that is in the neighborhood of 10% of the cost of the finished module. Processing the wafers into cells uses far less energy than that. Wafers tend to be produced in areas with reliable, low-cost sources of electricity (which tend to be places served primarily by hydropower). There are a few PV companies that have solar arrays on their facilities, but often they're more for marketing and public relations than to do any serious offsetting of the company's electric bill. For one thing, as you can see, even if the cells produce electricity for free they can't reduce production costs too terribly much. The PV manufacturer is often better served by selling the cells, rather than keeping them, so as to profit from the sale and have money to invest in making more solar cells. Finally, whether the PV manufacturer keeps or sells the cells, roughly the same amount of solar electricity will be produced by them, so why not sell them and make the aforementioned profit?


 * As to the US government providing financial incentives to make wafer production facilities energy-self-sufficient, well, most wafer production facilities make wafers for both the PV and the microelectronics industries. Most of them are not renewable energy companies per se; the one major one that is, is a Norwegian company (albeit one with some manufacturing sites in the US). Some PV manufacturers buy refined silicon and make their own wafers, but only about 7% of the global PV market is located in the US, and two of the biggest US manufacturers are thin-film companies that don't use silicon wafers at all. Thus, even if the US government institutes the policy you advocate, its effect would be quite limited.--Squirmymcphee (talk) 21:56, 21 July 2010 (UTC)

Geography and sunlight
I think it would be useful if the article introduced the reader to the concept of the relationships between the amount of sunlight incident upon a solar cell and geographic location (latitude, amount of sunlight available, time of year, etc). This aspect seems relevant insofar as one might try to access this article in an attempt to evaluate the feasibility of a system utilizing solar cells (panels/modules/arrays etc). It is not necessary to cover this aspect entirely - the article might point the reader to other resources that might adequately answer these questions. 24.222.87.186 (talk) 19:41, 21 October 2010 (UTC)

The relation of sunlight and geography is covered in Solar radiation and Insolation. Joja lozzo  00:44, 22 October 2010 (UTC)

Warranty example
To source information about panel warranties we have used an actual example but some of us have been concerned about accessing this as an external link to a commercial web site. Does anyone have any ideas about how to source warranty information without sending readers to a commercial site? Joja lozzo  18:55, 11 November 2010 (UTC)
 * Answered my own question (http://www.toolbase.org/Technology-Inventory/Electrical-Electronics/pv-systems#warranty), but if anyone can improve on that, please do. Joja  lozzo  19:01, 11 November 2010 (UTC)

Link to place in article
I am not sure, that this link may be placed directly to main article.

http://drugoi.livejournal.com/3411895.html An article about producing solar cells for Russian space ships. Sergei Frolov (talk) 10:31, 12 November 2010 (UTC)

Cost
The "Cost" section in the article about solar-cells contains an interesting discussion, but no real information about cost. The section should contain information about actual cost, in dollars, of solar cells currently available for purchase.

The article states that cost for solar cells is given per unit of peak electrical power, but it seems to me that cost per square meter would be a much more useful parameter. DougGeo (talk) 21:32, 12 February 2011 (UTC)


 * Cost per unit of peak output is the standard way of talking about cost in the PV industry. The reason for this is that ultimately, buyers of solar cells are interested in how much energy they will get for their dollar, not the number of cells or the number of square meters, and in this context, peak power is a reasonably good and easy-to-understand proxy for energy (there is a debate about whether this can be improved, but that is for another discussion). As a result, low-efficiency solar cells tend to cost much less to manufacture and sell (per cell and per square meter) than high-efficiency ones, but they are roughly equal on a cost-per-peak-watt basis. Therefore, the cost per-cell or per-square-meter is strongly dependent upon cell efficiency, while the cost per peak watt is independent (or, more accurately, only weakly dependent) upon cell efficiency.--Squirmymcphee (talk) 16:35, 19 February 2011 (UTC)

Three generations
Thin films, the second generation, had 40% market share in 1988 and it has steadily declined since then, there was a brief resurgence with CdTe which ended last year when the market share of thin film again dropped by thirty percent to less than 13%. More than eighty seven percent of commercial solar cells are made with crystalline silicon and thin film is very much yesterday's technology. (For the data see March issues of Photon International for each of the last ten years) The three generation discussion is misleading and is a technology judgement that is neither objective nor accurate.

Organic solar cells could be the third generation, well maybe. Plants have an efficiency of 1% this is an order of magintude lower than today's solar cells. There will be other solutions to better solar cells. Maybe someone might invent an inexpensive tandem cell. A photo converter that converts a blue photon into two red photons that are well matched with the PV bandgap might become the next best technology. Who knows, I don't and the three generation people don't know either it is premature to crown either a second or a third generation.

I deleted the text three generation text with the intend to make wikipedia better.

Cleanenergy (talk) 21:58, 11 June 2011 (UTC)


 * I agree 100% with deleting it. "Three generations" is sometimes used for one person's opinion of the future of solar cells, and other times used as a vacuous marketing slogan. :-) --Steve (talk) 22:05, 11 June 2011 (UTC)

Merge from Fill factor to Solar cell
The Fill factor article has a concise discussion of the term (and a citation) as used in photovoltaics and then mentions a couple of other uses of the term in other areas. My proposal is to merge the PV portion into Solar cell and leave the existing article as a disambiguation page for the general term. Joja lozzo  14:56, 19 June 2011 (UTC)
 * I agree: Fill factor, in this context, is only about efficiency, which can be handled in a small paragraph here.Peter Chastain (talk) 17:39, 3 August 2011 (UTC)
 * Done. We didn't need nearly as much math as it had, and defining terms in English instead of math subscripts is probably a good plan everywhere in an encyclopedia. --Wtshymanski (talk) 18:21, 4 September 2011 (UTC)

Solar thermal resource in BusinessWeek
Solar Panels Start to Outshine Mirrors; As prices for photovoltaic cells tumble, developers are abandoning solar thermal plants October 13, 2011, 5:00 PM EDT by Ben Sills 97.87.29.188 (talk) 23:38, 18 October 2011 (UTC)

Crap state of the article
Andy is right. current article This article is an internal memo for some firm on a planet far far away. Go nuts on the hack and slash, if anyone wants or needs a third, I'll be happy to help. You could machine-gun this article with templates there are so many sections that are rubbish in so many ways. Someone needs to learn how to plagiarize their own companies brochures for an encyclopedic audience. Penyulap  talk 16:16, 24 October 2011 (UTC)

If not crappy, then unintelligible in places
"Other technologies have tried to enter the market. First Solar was briefly the largest panel manufacturer in 2009, in terms of yearly power produced, using a thin-film cell sandwiched between two layers of glass. " Why is this a different technology..... even if the content in the article is ok (and it does need some work--took me a while to figure out where the history of development material was ) it reads as if it hasn't been proofread by the last person who edited it...commas and verb tenses count...the preview feature is there for a reason...please read your edit after you post it to make sure you are not obfuscating a clear point or concatenating stupidly. Eliminate run on sentences....etc etc, etc... Avram Primack (talk) 15:57, 9 November 2011 (UTC)

To consider for inclusion
&mdash;User:Ceyockey ( talk to me ) 01:55, 24 January 2012 (UTC)
 * lay summary
 * peer-reviewed publication

This page disagrees with another wiki.
Please note that the efficiency section states that "Single p-n junction crystalline silicon devices" are nearing 37.7% efficiency as is in-line with the Shockley Queisser limit but the Shockley Queisser limit page (linked below) states a maximum of 33.7% (which is the only reference of the claim).

http://en.wikipedia.org/wiki/Shockley%E2%80%93Queisser_limit

Either one of the numbers on one of the pages are wrong or the Shockley Queisser limit is incorrect, which I strongly doubt. The link in the Shockley Quesser page directs the user to the below linked source which says 30%. Since I don't know which number is right, I'll leave it to others to make the appropriate correction.

http://jap.aip.org/resource/1/japiau/v32/i3/p510_s1?isAuthorized=no — Preceding unsigned comment added by 199.44.8.162 (talk) 15:15, 3 February 2012 (UTC)

Confusing image caption – "Mono" and "Multi"
The caption in the image in the section Solar_cell is confusing. What is "Mono" and "Multi"? "Monocrystalline" and "Multicrystalline"? "Mono-Junction" and "Multi-Junction"? Tony Mach (talk) 10:00, 27 April 2013 (UTC)
 * You're right, it is confusing. I tagged the image page itself with (I'm not sure if you can do that to an image page, but hey-ho). The user who created the image doesn't seem to have contributed to Wikipedia since 2011, and their talk page is mostly a reminder to cite sources. --Nigelj (talk) 13:13, 27 April 2013 (UTC)

current events section
the current events section references 2011, it's 2013 and the prices have been moving hard. It may be worthwhile to update this.--Patbahn (talk) 21:40, 27 April 2013 (UTC)

Efficiency - Shockley-Queissner
In the 'Efficiency' section, there is:

"Single p–n junction crystalline silicon devices are now approaching the theoretical limiting power efficiency of 33.7%, noted as the Shockley–Queisser limit..."

But in http://en.wikipedia.org/wiki/Shockley%E2%80%93Queisser_limit (the 'Shockley-Queissner limit' article) there is:

"The limit places maximum solar conversion efficiency around 33.7% assuming a single p-n junction with a band gap of 1.34 eV (using an AM 1.5 solar spectrum).[1] That is, of all the power contained in sunlight falling on an ideal solar cell (about 1000 W/m²), only 33.7% of that could ever be turned into electricity (337 W/m²). The most popular solar cell material, silicon, has a less favourable band gap of 1.1 eV, resulting in a maximum efficiency of 29%."

So it sounds to me that in this (Solar cell) article, the single p-n junction crystalline SILICON solar device would have a theoretical limit of 29%, not 33.7%. The 33.7% limit applies to some other semiconductor with a 1.34ev band gap, (NOT silicon, which is 1.1eV).

Right? 71.139.169.27 (talk) 00:54, 24 October 2013 (UTC)

- Not quite. Auger-recombination limit 29%. S-Q limit 33.7% — Preceding unsigned comment added by 14.200.69.23 (talk) 09:05, 28 December 2013 (UTC)

Electrical Circuit Diagram Symbol ?
Please would someone add the electrical circuit diagram symbol for a solar cell? Thanks ! Darkman101 (talk) 19:58, 29 December 2013 (UTC)

Photovoltaic cell a type of solar sell?
On the Dutch wiki it is said that a photovoltaic cell is a type of solar cell. Here it is stated that these are exactly the same things. I am confused on what is correct. Can someone point out some literature to put some light on this? Kind regards, Timelezz (talk) 21:12, 25 February 2014 (UTC) The Dutch wiki says: Er zijn twee soorten zonnecellen. De bekendste is de geheel uit vaste stof bestaande fotovoltaïsche cel, die met vele tegelijk wordt gemonteerd in zonnepanelen. De tweede is de foto-elektrochemische cel, welke terug te vinden is in foto-elektrochemische generatoren. There are two types of solar cells. The most famous is fotovaltaïc cell which is completely composed of solid matter, and is installed with many in solar panels. The second is the photo-electrochemical cell, which can be found in photo-electrochemical generators. Kind regards, Timelezz (talk) 21:16, 25 February 2014 (UTC)