|Fig. 1: - Webberville Solar Farm Near Austin Texas from the Air. (Source: Wikimedia Commons)|
Solar energy is becoming more and more popular in the United States. With more people becoming more conscious about the effects of global warming, the interest in solar energy to replace fossil fuels has also greatly increased. In order for solar energy to achieve this feat, large solar farms, order of magnitude larger than the typical solar farm shown in Fig. 1 would need to be constructed. In this thought experiment, I want to explore the hypothesis that solar energy can replace fossil fuel as the main source of energy production in the United States.
Solar panels efficiency has been the fundamental limiting factor of solar energy harvesting. From ShockleyQueisser limit calculating the maximum efficiency of solar energy conversion to be 33.7% with a single p-n junction to the maximum theoretical efficiency cap of 68.7% with an infinite stacking of cells from skylight.  In this calculation, I will be using the current state-of-the-art of 29%. 
To be able to generate the power is only part of the equation. The next immediate need is to be able to store the energy, as a buffer, for when solar panels are not active such as at night or during rainy and cloudy days. This gives incentives to construct solar farms in areas of dry weather, long day time, and cheap land. The next part of this solar power system is the transmission of power from one or multiple central locations in the United States to the rest of the country. Some regions of the country is not suitable for solar panels, which then must rely on other states to produce and transmit power. This not only creates additional infrastructure needs but also potentially geopolitical issues between states. The last consideration is the maintenance cost of the proposed solar farm, although this could create a lot of local jobs for the locations of the solar farms. This also could be a hotly contested political issue as states which are currently big on fossil fuel electricity generation would lose a large number of jobs to states which are ideal for solar farms.
First, let's calculate roughly how much solar energy are there per square meters:
|103 Watts m-2
|× 3600 sec h-1 × 24 h d-1 × 100 days = 2.8 × 109 Joules m-2|
Given the previously quoted current best solar panel conversion rate:
And the current energy consumption of the United States, and given the hypothesis that solar energy will be the dominant source of energy generation, we will assume that 51% of this energy will need to be produced by solar: 
So this can lead us to how many square meters of land needs to be covered by solar panel:
|5.38 × 1019 Joules
8.12 × 108 Joules m-2
|= 6.62 × 1010 m2|
To more easily put this number into perspective in the United States, we can convert this number into 25560 square miles. To put this number into perspective, it is would be about covering about 21% of the land area of the state of New Mexico.
Now that we have calculated the land area that would need to be covered with solar panels to make this hypothesis work, let's talk about the other part of this thought experience, definitely the harder part, which is the cost. What makes this calculation hard is the uncertainty of price, especially given the scale of this project. Ignoring the cost of the infrastructure of transmitting the power to the rest of the nation, the main costs of this giant solar farm is the cost of the solar panels and the cost of the battery needed to store the energy. With current solar farms, at a large scale, usually cost about $1 per Watt it generates. 
|5.3 × 1019 Joules y-1
× $1 Watt-1
3600 J Wh-1 × 24 h d-1 × 365 d y-1
|=||$1.7 × 1012|
This calculation does not take into consideration of the battery needed to store the power for day/night cycles and potential bad weathers, land cost and the maintenance cost of the solar farm, which would be significant as well.
Let's talk about the area first. Building one giant solar farm that takes up 21% of the land area of New Mexico definitely sounds not feasible. However, this definitely would not be the way of possible implementation. Most likely multiple mega-large scale solar farms would be constructed at multiple states with a lot of sunlight hours and clear days, such as Arizona, California, Nevada, New Mexico, Georgia, Colorado, Idaho, Kansas, Oklahoma, Utah, Wyoming and Texas. If spread out in these 11 states, we are only looking at about 2% of the total land area of these states.
The above calculation shows a lower-bound starting cost of more than $1700 billion for this project. The approved 2019 U.S. military budget of $686.1 billion.  So this project would require multiple years of budget and spending similar to the scale of U.S. military budget. However, the additional cost of upkeeping of the system, the infrastructure required to transmit and deliver the power, and the geopolitical issue between states require much further discussion.
© Tony Chen. The author warrants that the work is the author's own and that Stanford University provided no input other than typesetting and referencing guidelines. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.
 A. De Vos and H. Pauwels, "On The Thermodynamic Limit of Photovoltaic Energy Conversion," Appl. Phys. 25, 119 (1981).
 J. Temple, "These Flexible Solar Cells Bring Us Closer to Kicking the Fossil-Fuel Habbit," MIT Technology Review, 16 May 19.
 "October 2020 Monthly Energy Review," U.S. Energy Information Administration, 27 Oct 20.
 A. Sharma, "A Comprehensive Study of Solar Power in India and World," Renew. Sust. Energ. Rev. 15, 1767 (2011).
 "2019 Defense Budget Overview," Office of the Under Secretary of Defense (Comptroller), Feburary 18.