Talk:Crystalline silicon

External links modified
Hello fellow Wikipedians,

I have just modified one external link on Crystalline silicon. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20120321042800/http://sunlab.site.uottawa.ca/pdf/whitepapers/HiEfficMjSc-CurrStatus%26FuturePotential.pdf to http://sunlab.site.uottawa.ca/pdf/whitepapers/HiEfficMjSc-CurrStatus%26FuturePotential.pdf

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 04:05, 15 August 2017 (UTC)

Energy payback time
I first thought this looks like a bunch of misleading nonsense when I read the FI document and looked at the map, but this is actually a very interesting aspect of the solar industry. Obviously, if it takes more energy to create, transport, install, dismantle and dispose of the panels than they will generate over their lifetime, that sucks. The problem is clearly not that bad, but it appears significant. The point is that solar power generates a large amount of unsustainable GHG emissions, just a lot less than power generation from fossil fuels, and that's not something that will change anytime soon.

First off, this entire section pertains specifically to (the GHG emissions of) photovoltaics, and needs to be moved to that article. What needs to remain are facts about silicon -manufacturing, energy intensive, made in China.

Secondly, the map needs to go. Shitty, cause it looks like someone put some work into it, but it's clear they didn't entirely understand it. Basically, it is a pointless and misleading exercise to manipulate statistics to make a what-if? map showing lots of patently incorrect but flattering-looking nonsense numbers. Rajasthan may have a lot of sun, but its not in the EU and that's not likely to change soon, and its not useful to use the figures from the best top-of-line systems, which have only a small market share. So we know the actual EPBT numbers for the map are significantly worse, but the FI data is too poor to say how much. FI calculate an efficiency increase of some 300% over the period 1980 to 2010, applying that to Alsema et al.'s worst scenario (they looked at top-of-the-line and low-end rooftop systems) gives you maybe a 3yr EPBT (in Germany (I think)).

We can try to compare the data sets only regarding the EU calculations, none of the other numbers are accurate or useful (except China, but we have no previous data). If the 1997 top-of-the-line silicon-based rooftop system connected to the grid had a EPBT of 3.5 yrs, of which the polysilicon alone accounted for 3 yrs, and a similar system 1.28 in 2019, that actually gives you a lower EPBT efficiency increase rate, a low-end system would have a EPBT of roughly 5 yrs. To put it differently, if we switched all energy generation to solar, it would produce much less GHG emissions, but still get us to the dreaded 2 degree rise by around 2600 (worst case, very rough calculation), instead of 2100 if we switched to natural gas, or 2050 if we do nothing.

This is all quite unfortunate, because we only have two data sources here, FI from 2019, Alsema et al. from 1997! And the data sets aren't directly comparable, as different methodologies were used, as well as a slightly different definition of EPBT, and that's also making a number of assumptions which aren't specifically stated. Very rough calculations, the thing about the EPBT efficiency increase rate for example, FI calculates their energy use breakdown differently, but it does appear as if silicon is progressively becoming a lower percentage of the total energy cost of PV arrays. An actual life-cycle analysis would also be a bit more useful than this limited metric, the EPBT only looks at production, and only of the array, not the mounts. And its all very unclear, take the coke needed to smelt the sand to make the silicon, is the diesel being calculated needed for the transport trucks from mine to furnace? What about the GHG emissions from workers commuting to the factories? I've seen multiple very detailed anti-beef calculations, we could do with one here.

Funny, considering life-cycle, it also true that, if we accept these EPBT efficiency increase rates and backtrack to the seventies or eighties, then there was a period of time when using (some) solar was actually net worse for the environment than using fossil fuels! Cheap calculators? 86.83.56.115 (talk) 09:23, 10 June 2021 (UTC)