Direct conversion drawback
At the time of writing, silicon solar cells are not the answer to the prayer for cheap (free?) energy. They are expensive (typically about £100 per installed watt for commercially produced units). Although the running cost is negligible and the energy free, the plant cost is high and the 'breakeven' point after which power is produced at an economic level of cost is between 15 to 20 years from the date of installation. This represents a long-term investment of capital (although on equipment which will not 'wear out'), and, for this reason, the accountants (not noted for economic foresight in an engineering world!) will probably veto a solar-power scheme.
Aside from the economic considerations, the limiting factor is energy storage. The Sun does not shine every day—neither does it at night! Direct conversion is dependent on the weather and time of day (at ground level at least). This dependence requires a storage system to run the load at night and during periods of no or low insolation. Sufficient capacity must be installed to cater for the excess energy generated in the summer and to store and then to release this in the winter when daylight is short and the Sun low in the sky. The cheapest way of doing this at present is by lead-acid batteries.
The power from arrays is generated as direct voltage. Unless it is chopped or converted distribution becomes unattractive (as far as paralleling the solar produced power to an existing grid distribution is concerned).
It would be foolish to make categorical predictions as to the future of direct conversion of solar energy for terrestial use. As far as solar cells are concerned, cost is probably the biggest single factor against their employment as power sources for any large-scale distribution scheme. Their low efficiency compared with nuclear or conventional power plants is a further point against them. However, a sobering thought is that 17 years of available solar energy falling on the Earth is estimated to be equivalent to the world's total supply of fissionable nuclear resources. This figure becomes more significant when it is realised that it is about 35 times the lowest estimate of the world's energy needs 100 years hence. This supply of low-grade energy will certainly become increasingly important as a means of reducing the load on power-generating plant.