Editor's note: A previous version of this article stated that it would require 214 million Gemasolar stations to meet the world's energy needs, when in fact it would only require 214,000.

Solar power holds the theoretical potential to solve all the world's energy problems. For example, in just one hour the sun shines enough light on the earth to power the world for a year.

The problem is harnessing this nearly limitless, clean, and renewable power when and where we need it. I recently wrote about Japan's innovative efforts at solar space-based power. This article brings the concept of solar breakthroughs down to earth and looks at two exciting technologies that could allow solar power to come into its own as a practical future power source.

Solar power at night
One of the major weaknesses of solar power is that it's intermittent. Clouds, storms, and the setting of the sun can lower or eliminate power output. This means that traditional power sources such as gas, nuclear, or coal are required to maintain a base load. 

Concentrated Solar Power, or CSP, potentially offers a way around this issue by combining solar power with mirrors and a liquid energy storage medium. 


Source: Torresol Energy 

Terrosol Energy, a Spanish concentrated solar company, has designed the Gemasolar station to generate 20 megawatts of power -- enough for 30,000 homes. 

Mirrors, a total of 2,650 of them, concentrate sunlight by a factor of 1,000 onto a tower where molten salts are heated to 1,050 degrees Fahrenheit. The molten salts can be stored in tanks and used to generate steam to spin electricity generating turbines up to 15 hours after the sun has set. This technology allows for solar power to be generated around the clock and has a capacity factor of 75% (generating maximum output 75% of the time). This is in between gas-fired power plants, whose capacity factors range from 25%-50%, and nuclear plants which run at 90% of maximum capacity.

All told Gemasolar can produce 60% more power than a similarly sized solar panel installation, but there are two problems with the design. First, the size of the installation is enormous, requiring 480 acres of land. Given that in 2012 global power demand was 23,500 TWh, this means to generate all the power demanded by the world using this design would require around 214,000 Gemasolar stations and take up 103 million acres of land.

Second, the cost of the station is estimated to have been $260 million, or $13 billion per GWh of capacity. To put that into perspective it's 2.7 times as expensive as traditional solar power and 13 times as expensive as gas-fired power.

Gemasolar is expected to achieve a positive return on investment in 18 years, but Terrasol Energy is attempting to bring down the cost by scaling up concentrated solar with its Valle 1 and 2 plants, which together generate 100 megawatts of power, enough for 80,000 homes.


Source: Torresol Energy

Unlike the tower design of the Gemasolar station, the Valle power stations use a parabolic trough design. Sunlight is focused onto a glass

parabolic concentrated solar source Torresol Energy

tube containing thermal oil, which is then heated and used to generate steam. Like the tower design, this technique can generate power when the sun isn't shining and the cost is 44% less per gigawatt of capacity than the Gemasolar station. Unfortunately this method also has major drawbacks. 

For one, the cost, though lower, is still $7.3 billion per gigawatt. Second, the area required is mind-boggling with a total of 1.1 million square meters of solar collectors. Finally, the heat storage capacity of this system is just 7.5 hours, meaning it can only operate around 4,000 hours per year, vs 6,000 for the solar tower design of Gemasolar (and 2,000 hours for traditional solar technology).

Concentrated solar may offer a way for solar to provide all day power, but its enormous land area requirements and uneconomical cost make it unlikely to ever solve the world's power needs.

Massive solar cost savings needed
A Durham, North Carolina start-up called Semprius is working on a way of stacking miniature solar cells (each just a millimeter across) together to achieve unheard of efficiency and cost savings. By growing layers of different semiconductor materials, each capturing a different frequency of light, the company has been able to build a solar cell with 44.1% efficiency. Semprius thinks it can eventually achieve efficiency beyond 50% (double current solar cell efficiency) within three to five years.

According to Scott Burroughs, vice president of technology at Semprius, with sufficient economies of scale his company's technology could result in solar power costing just five cents/KWh. That's 22% less than the cost of a new gas-fired power plant (6.4 cents/KWh) according to the U.S. Energy Information Administration.

Solar power, despite its amazing potential, still has a long way to go to overcome its two largest hurdles -- constant output and high cost. However, advances in innovative solar engineering techniques and material sciences are helping to solve those issues and may one day usher in an era where the sun may be all we need to power the world.