Dielectric strength of air seems high at 3 megavolts per meter (76 kV per inch) (as per wikipedia) however that's the voltage needed to spontaneously arc across a gap. The wikipedia article also notes that "Because dielectric materials usually contain minute defects, the practical dielectric strength will be a fraction of the intrinsic dielectric strength of an ideal, defect-free, material."
The subtle point made by jazty is important: when a current is mechanically switched, there's an arc. There's always an arc; we just forget it's there because it is contained inside the case of the switch and we don't see it. If the switch isn't up to the task for the current and voltage being managed, the arc may endure longer than it's supposed to (even indefinitely). That's the principle behind dc arc welding -- strike the electrodes together, whether buzz box, MIG, TIG, whatever -- then pull back and maintain the arc. So long as the source can keep enough current flowing to keep the air hot and ionized the arc will survive even with just a few tens of volts. Anyway, that's part of why seemingly low voltage dc can still present a fire hazard, and it's why the variety of published dielectric strengths for air don't really matter to low voltage equipment.
As for converting higher open circuit voltage solar panels down for 12 volt batteries versus other voltages: I suspect step-down aka buck converters can be designed to achieve similar efficiency levels regardless of the input-to-output voltage ratio. Historical inertia more than anything else is probably what keeps us coming back to dc designs at 12 volts.
Good info & well put!!
As to the last point regarding higher voltage panels on a 12v system. That is where MPPT charge controllers shine. The last Outback controller I fooled with will support series strings of panels up to 150VDC. The MPPT controller handles dropping the voltage to the appropriate for your system (12,24 or 4
