To me, this kind of blocking out of massive areas of unused space and dedicating it to energy production is a very likely future. The Empty Quarter is larger than France, and it really is empty - there really isn't much of a lot of consequence going on there in terms of life, certainly nothing whose disappearance is likely to adversely affect anywhere else.
I did some relevant maths
here.
Renewables currently supplies about 1% of global energy demand.
Set aside, for a moment, the inconvenient fact that most of the things we need energy for (food production, and the global transportation subsystem required to sustain the industrial manufacturing system renewable energy system manufacture depends on) don't and can't run on electricity, and that this is a hypothetical question with no practical relevance to the civilisation we happen to live in.
Existing renewable generating capacity would have to double in size seven times. But the power generating capacity necessary to power the manufacturing system to provide that generating capacity would have to increase 2,560 times (assuming a solar EROEI of 3 - this number rises to infinity as solar EROEI falls to zero, which it does as its own gross input energy demand rises as manufacturing switches from hydrocarbon powered to renewable powered).
But we still need the power that has been diverted from the array into the array's own manufacturing process (it was currently being provided for free by hydrocarbon, remember). So increase the size of the array by the amount required to power its own manufacture. We are talking here about the entire industrial manufacturing system's requirement, not the trivial amounts solar enthusiasts account for in their estimates of EROEI after casually assuming the existence of the (hydrocarbon powered) industrial manufacturing system.
But that needs more manufacturing power, which needs more array, which needs more manufacturing power. Repeat that geometrically expansive recursive loop a few times in your head, then retire to have a think about how easy things were (and how small mobile phones could get) when all you had to do was stick a straw in the ground and suck out EROEI 100 hydrocarbons.
Now recall that many of the resources critical to the array's manufacture (particularly rare earth elements used in their various electrical and electronic subsystems) are already approaching exhaustion*. Expansion of generating capacity ceases at the point of exhaustion of the scarcest critical resource. Most of the systems (energy system, financial system, transportation system, manufacturing system, food system, law and order system, etc.) that would have been required to develop substitutes for scarce resources require the energy system to be functioning for their own functioning. In systems theory terminology, the array is both a "single point of failure", and a self terminating failure. The array never gets built.
Set that aside, and assume you could manufacture the array, carefully noting that none of the raw materials required by that manufacture are found near your array, and the transportation system necessitated by that fact doesn't run on the energy format of your array's output. The part that serviced our current energy demand would be a square about 100km on each side. (The part that serviced the grid's own manufacturing and operational energy requirement would be a square very much larger in the short term, and infinitely larger in the long term. Set that aside, too).
Note that power losses getting power from the array are in proportion to transport distance, and you are now transporting over colossal distances. Build some more array to cover that loss. Build some more to power the steel, copper, and aluminium mining, refining, and manufacturing systems required by the distribution infrastructure (and for the vehicles required for the ore mining, and for the factories they are built in. Oh, and for the industrial oxygen manufacturing facility for the smelting. In fact, for a bunch of stuff.) See above.
The elements of the array have to be accessible for construction, maintenance, and repair, via some sort of access infrastructure. Assume you can access elements 500m either side of a road. Then to access all the elements, build a (100km x 100km) 10,000 km road structure, associated construction vehicles and infrastructure, and access vehicles. Build some more array to power the global industrial manufacturing system required by that. See above.
Now spot that global power demand is increasing about 2% per year (unless you feel like telling the 80% of the planet that consumes 80% less than you otherwise). Do the whole thing again in 35 years time. Probably need to start that now (the existing system took 100 years to build) i.e. build two systems concurrently - one at the rate imposed by the uninvested depletion rate of the petroleum system (i.e. 7 year half life) since you can't invest in propping up the petroleum system AND its replacement simultaneously in a financial system on the verge of collapse. Meanwhile, set aside that the global system is already in energy deficit before we divert a substantial fraction into energy infrastructure replacement manufacture, noting that we have historically replaced the dominant energy system before its saturation, ensuring sufficient margin to power its replacement while sustaining the existing system.
Then build twice as much again in the following 35 years. And four times as much in the 35 years following that.
The issue isn't finding a nice, empty spot.
* Rare earth elements are not rare because they cannot be found. They are rare because they are (now) found in concentrations that require more refining energy than the energy upgrading equipment they are required for can deliver. In the parlance of resource scarcity, they lie "below the energy horizon". This is thermodynamics 101.
