Benny
This Australian University study explains the recycled silicon further (aye no sticky stuff this time, what a difference a single letter makes!). They also claim 7 years for PV and 8-11 for total system if installed in Sydney. http://www.ecotopia.com/apollo2/pvepbtoz.htm
It is an older though, but it should point you in the right direction to find out more about the silicone recycling process. I believe it is the low grade reject ingot that is purified and recycled, and this would happen long before they would be "chipped" and packaged in plastic to be sold as ICs. As I tried to explain further up specific manufacturing data is hard to come by, as this is generally not available from companies interested in selling PV.
In regards to the electrical safety I would like to make some general comments about electrical networks, island and parallel modes:
- in a parallel network type arrangement (the most common PV installation method) the inverter physically needs to read the 50Hz from the network side, in order to produce a synchronized AC current with the grid. The inverter has no "own" oscillator to drive the inverter in sync with the grid, hence it is not possible to run such a parallel inverter without the 50Hz pulse. This is similar to old LED clock radios that always lose their time when theres a power failure or you unplug them. This type of clock does nothing else than count up to 50 AC cycles and then add a 1 second increase on the display. It then just keeps on counting for minutes and hours. Try plugging a US one in via a step down transformer and you will find that time slows down..this is the easiest time machine I know of 
BTW this is also were the old style TV sync comes from.
Another way to look at synchronization on the network is to imagine marking every generator armature and stator whilst stationary, and then start up the network. You will find that all alternators are rotating at exactly the same speed ie the marks will pass each other in synchronization with all other alternator marks(if they have the same pole configuration). A inverter needs this "mark" to synchronize to the grid otherwise it is impossible to export to the grid. Trying to feed in out of sync is the equivalent of electro-mechanically trying to force the other generators to speed up/slow down and will result in spectacular failure of the tiny inverter. Commercial non-inverter and gearboxed Wind Turbines display this effect of synchronization clearly. It is one of the reasons Enercon (whom we used to do structural consulting for 13 Years ago) decided to use inverters, without any gearboxes to fully utilize the power of the wind. The ones with gearboxes generally have two alternators that they switch between for different speeds, and will only operate at either speed regardless of the wind. My point is it will not work to mod such an inverter to make SJ's original proposal work without serious interference with the inverter itself. That is why I did not in any way promote it, I did however proceed to recommend how he could achieve the desired effect, as I described further up in this thread.
- The islanding/parallel system, that I proposed, uses both internal and external timing for synronising to either the grid or to operate itself in islanding mode, isolated from the grid. This system even includes some benefits that no-one seemed to have an opinion at all, which I find a bit disappointing. But a dual islanding/parallel system can be profitable if some intelligent manipulation of the system is used. Remember the return on investment is a function of money, not of energy. A FIT is designed to provide an economic stimulus to consumers to install renewables, I see no harm in making use of such stimulus to the fullest extent in order to provide more power to the network. In regard to using storage to take advantage of the off-on-peak periods, this is not counterproductive if alternating between tariffs. For further info on this chk out the plans of Shai Agassi and his electric car buffered, movable capacity network.
In regards to the DC transmission, these are primarily used for long undersea and out of sync connections. AC will only go 30km or so underwater. BTW I'm generally opposed to centralized generation and distribution. A locally sourced, distributed generation system requires little long distance interconnectivity. But now I'm beating my own drum...chk out Biogas systems in Germany, they achieve exactly that, with many more benefits than any other renewable I know of.
Defibrillation usually uses around 1000V and once used a single Lown waveform before they introduced Biphasic defibrillation. The original 50Hz power point versions only had very limited success, and only managed to cause serious burns and muscle damage in most patients. Frequency is key when dealing with high potentials, check out some history and achievements on Tesla the inventor of AC.
