Can Solar replace all other energy sources?

I spent some time calculating how many solar panels we would need to produce to replace all of Americas electrical energy with solar energy.

Here area the results

On average we use 5 trillion kWH of energy every year.  That works out to 570,000,000,000 watts of power needed every hour.  Your average 16 square-foot panel produces on average 100 watts of power.  This means we would need 570,000,000,000 / 100  panels = 5,700,000,000.  Each panel is 16 sq-ft so they would cover 9827  sq-mi.  or a roughly a 100 mi x 100-mi area.

Now that seems fairly small and do-able.  So what's stopping us?

Let's look at the costs.
   First each panel today costs about $1000 to produce, takes up about 16 square-feet and generates about 100 Watts of energy.

   Second, we will need to generate enough electricity during the daylight hours to have enough stored up energy for the nighttime.  Given approximately 8 hours of good daylight to work from we will need about 3 times (24/8) the size of grid to do this. (I will not be adding in the cost of the storage facility)



Therefore the overall cost (not including support, assembly, wiring, control) is almost 17 TRILLION DOLLARS.  This is 1 TRILLION more than our current US debt.  And that is just for 1 GRID.  If we needed more than 1 for redundancy or for bad weather, the cost would double or triple. 

There are other hidden dangers and costs of solar energy as well.

1) Solar panels will have an effect on plants and animals
    Light energy used for generating electricity cannot be used for anything else.  Wherever we put
    these panels, the plants that need the light and the animals that feed off the plants in that region
    will disappear for good.  That 100x100 mile square area will become a total desolate wasteland in
    which no living thing would exist.

2)  Solar panels will change the climate too.
    Light reflected off of these panels will get reflected back up into the atmosphere and will change
    the weather patterns (imagine a 100 mile x  100 mile mirror reflecting light.. you don't think that
    will have some effect on our weather?)

NOTE: There are 9 states SMALLER than 10,000 sq-miles.  The closest to this size is New Hampshire or Maryland and Vermont with sizes of 9775 and 9249 respectively. Just to give you
some scale as to how large this area is.
 

Unlike most solar power enthusiasts and environmentalists would tell you, solar power energy is NOT FREE!   It is VERY expensive which is why our power companies have not adopted it as a good energy source.   In my opinion the cost of solar energy needs to drop by 100X so that a solar panel costs $10  and not $1000.  Then and only then will solar energy become a cost effective source of energy.   We will also have to live with the fact that our new method of "clean energy" will have effects on land, animals and maybe even our climate.

Is this possible?

Probably not.

Unlike computer transistor technology where the reducing of the size of transistors allows more capability at lower cost,  photo-cells used in solar cells to convert light into electricity, do not have that same luxury.  The main factor in solar cells is not the transistor-size, but instead the effectiveness of the cell itself in converting light to energy.  Over the past 30 years solar cells have gained in efficiency, going from 10% efficient to now up over 45% efficient.  Therefore even if we could achieve 100% efficiency, that would only gain us about a 2X improvement in cost (not 100X that we need).  The only other lever to pull is the COST lever.  Here in lies the problem.  To improve efficiency, photo-cells need to employ either the use of more rare elements (which are more expensive) or more elaborate production technology.  Either of which drives UP the cost of development.   Other technologies using "thin film" which are cheaper in development have plateaued at 15% efficiency (and little sign of improving). This means that they require 3 times the areas to achieve the same amount of wattage as silicon-based photo-cells, so our grid would have to be the size of Vermont, New Hampshire  Massachusetts and Connecticut put together  ( or 1/4 of New Mexico ).  To reach our 100X cost improvement, they would have to be 300-400X cheaper than silicon-based photo-cells and currently they are only about the same price.

I don't want to be one of those people that say "We will never ...." because often they are proven wrong over enough time.  But I think that in the very near future (< 20 years), we need to be realistic and see that there are major hurdles to replacing our fossil fuel energy usage with solar energy.