Polysilicon recycling does indeed require HF, for cleaning of dopants and etching.
Environmentally "Benign Solar Cell Manufacturing. In any case, chlorosilanes are potent carcinogens, and phosphorous oxy-chlorides are extremely dangerous chemicals as well, as is this dangerous desiccant P2O5. All are products used in the solar cell industry. In addition tetrafluoromethane is one of the most persistent greenhouse warming gases known, since there is no known mechanism for its depletion. Sulfur hexafluoride, also used in solar cell manufacture, is a very persistant greenhouse gas, with a half life of hundreds of years in the atmosphere. Both of these gases are enormously more potent greenhouse agents than is carbon dioxide, sulfur hexafluoride in particularly being the target of emission restraint considerations right now. (It is used in transformers and other electrical applications.)
The fact is that the solar industry PV industry, which produces very little of the world's energy output (is it up to 1% yet) escapes environmental scrutiny precisely because it is so small. It is so small because it is economically not viable. Scale it up to an appreciable fraction of the world's energy demand, comparable to nuclear even, and it will seem less benign.
(I note that if the solar cells contain aluminum supports, this will increase the use of HF, although I certainly do not expect to hear from religious nuclear opponents about this increased risk. To wit: You want to shut the nuclear industry based on 3% of the HF demand (all of which is recyclable) but don't give a rat's ass about the aluminum industry or refrigerant industries which combine to use more than 75% of the world's HF.)
Now, we haven't even touched the environmental cost of
storing solar energy.
Your claim about solar cells storing CdTe is valid in the sense that the solar cells will actually reduce the output of Cadmium into the environment since Cadmium is just one dangerous output of coal fired power plants. However, you expect me to accept on faith that thirty years of weathered solar cells will not ultimately become point source pollutants for Cadmium. This is hand waving nonsense. I have been trying to recycle my old fluorescent bulbs (which necessarily contain mercury) and can find no one to take them up. The claim that solar cells will be recycled is dismissive hand waving. We know for certain that point source polluters (the majority of whom are home owners) are very, very, very loose in compliance. Thus, assuming we could get around the rather horrid economics of solar cells, and the environmental (and economic) abyss of storing the energy in a 100% solar environment, we will only place the responsibility for dealing with millions of metric tons of CdTe on future generations. Note that I am leaving out the health costs for workers working with these materials.
How much cadmium telluride are we talking about? This hand waving link
Cadmium Facts gives a figure of 70 Metric Tons of Cadmium per Gigawatt. Given that the solar cells will operate (on average) for 1/3 of the day (allowing for night, rain, clouds etc), and that we will have to store the energy for the other 2/3 of the day at (an overly generous) 50% efficiency, we should really multiply by 5 this number to give 350 m metric tons of Cadmium Telluride for 1GW (1000 MWe), the output of a typical nuclear plant. Now if we build enough plants to merely equal the output of nuclear energy in the United States, we will quickly see that we will require 35,000 metric tons of processed Cadmium Telluride alone just to manufacture the cells to provide 20% of US energy.
Now, I'm sure that since you're a good guy, you have absolute faith that none of the Cadmium or hydrogen telluride (one of the more toxic gases known) used to manufacture these cells will harm a single human being during the manufacturing process, because the solar industry is uniformly populated with noble people and the nuclear industry is uniformly populated with evil people. I'm not so sanguine. I think they'll be some cost cutting going on here and there and maybe the violation of some environmental and health laws.
Probably it is true that once manufactured, the solar cells Cadmium Telluride will not leach out for centuries after the cells are discarded, but this is really not particularly different than the claim made about so called "nuclear waste," should it be defined as "waste," in the end. Interestingly this is rather close to the amount of so called "nuclear waste" produced in 30 years of producing nuclear energy, so we can say that based solely on toxic Cadmium Telluride, discounting HF, Arsine, chlorosilanes, sodium hydroxide, phosphorous chlorides, oxides, lead, and other interesting materials involved in solar cell manufacture.
Of course, as I've indicated many times before, nuclear materials differ from point source pollutants like Cadmium in that, because of the equilibrium condition (they decay proportional to the quantity available even as they form), they have accumulation maxima. For instance, the troubling fission product Cesium-137 can (under circumstances under which nuclear power was providing 100% of the worlds energy needs in 2050, 1000 exajoules) accumulate to the tune of 19,000 MT. It will do this slowly but it is true that after 100 years of such a program, the amount of new Cs-137 produced will be only 50 MT/year greater than the amount that spontaneously decays to non radioactive Barium 137. In 200 years, that figure will fall to 4MT/year. Cadmium Telluride, and the other waste products of the solar industry will remain forever distributed as point source (and therefore difficult to control) potential pollutants. Moreover, the amount of these materials produced and processed will NOT fall as amount of energy produced increases, since solar cells will need to be produced to replace those worn out thirty years earlier. Undoubtedly, a wise solar industry, will recycle some of this material, but even at a 90% recycle rate, we will have hundreds of metric tons of point source tellurium compounds spread around the planet.
Still, on purely environmental grounds, discounting economic grounds, entirely, PV cells are definitely preferable to coal in particular and fossil fuels in general.
I think it behooves us to advance PV and other solar technologies and to address these serious risks associated with their expansion. I think the solar industry will have a very very very long way to go to match the safety profile of the nuclear industry, but it is necessary to assume the added risk simply because nuclear resources are in fact exhaustible. It would be very safe for the next several 100 generations to avail themselves of nuclear resources, but I think that each generation, including our own, as a responsibility to use as much truly renewable energy as possible, so that the Uranium and Thorium are available for tens of millenia as opposed to the three millenia they would be available if they simply replaced all current energy output.
I have long stated here and elsewhere, that a wise distribution of these energy sources (especially on environmental grounds because of the tremendous environmental cost of storing solar energy) would be to use nuclear as a base load power source, and solar as a peak load source. This seems to be the best option if we are to survive the next several centuries.
As always, the best energy strategy of all is simply not to use energy in the first place. Conservation is the safest, cleanest, and most ethical of all energy approaches.