Will Solar Come to Dominate Utility Power Generation?

Econintersect: Solar power is rapidly becoming the cheapest form of power generation, soon to drop below the nominally cheapest coal and natural gas. But the problem has been that the sun doesn’t shine at night and the sun doesn’t shine on cloudy days (although some solar energy does get through the clouds) and little power is generated when it is raining or snowing. It has often been assumed that a large stand-by capacity of fossil fuel or nuclear generation would be necessary to smooth the production fluctuations of solar. But is that about to change?

Large solar farm, location not known. Econintersect estimate: >100 acres.*

Here is what Kent Moors ( a contributor to GEI) said in a recent article:

[T]he Japanese mega-corporation Toshiba (OTC: TOSBF) recently achieved a significant advance in large-scale electricity storage using hydrogen. It just may be the solution the energy market is looking for.

Two weeks ago, the Nikkei Asia Review (NAR) announced the development as follows:
“Toshiba has developed a way to use hydrogen to store large quantities of electricity for extended periods of time, with a storage system deploying the technology potentially reaching the market in 2020.

“The Japanese company will first provide a system capable of storing up to 40,000 kilowatt-hours of electricity – enough to power 10,000 households for eight hours. The system will be made by assembling fuel cells, electrolysis equipment and hydrogen storage tanks on a 600 square meter plot. Hydrogen obtained through electrolysis will be stored in the tanks and reacted as needed with oxygen in the air via the fuel cells to produce electricity.”

The cost, according to Toshiba, of electricity storage as hydrogen is half what batteries would be and not susceptible to long term leakage to the extent that batteries are. Hydrogen fuel cells produce electricity with 80% efficiency compared to 70% for pumped hydro storage, another competing technology.

Moors says:

Given these new developments, we may be on the verge of something quite important here.

So don’t be so quick to write off the value of hydrogen to the energy markets. I’ll have more on this as it develops.

Toshiba plans to have a 350 kwh test system in operation this spring and to have large production units (40,000 kwh) commercially available in five years. The 40,000 kwh output is enough for “powering 10,000 households for 8 hours“. The footprint for the entire system (panels, fuel cells, electrolysis units to produce hydrogen and hydrogen storage will be about 600 square meters (approximately 6500 square feet or 0.15 acres)

Econintersect has not found a cost estimate for the new system.

Cost for solar photovoltaic electricity generation is now competitive with other sources of generation and is projected to become the cheapest, perhaps by a wide margin based on the most aggressive estimates. Here is the status as of 2010 from the International Energy Agency:

Click for larger image.

In the U.S. the cost competitiveness varies by state depending primarily on local electrical rates from utilities and to a lesser extent on the level of state subsidies for installation costs.

But not all is looking up for large-scale solar farms. Last year the idea was rejected in Suffolk (UK) when the British Communities Secretary Eric pickles ruled against the proposal for large solar panel farm “the size of 75 football pitches” because it was “unsightly and a waste of arable land“. Econintersect: Presumably there are desolate areas in the U.S. which might escape opposition for solar farms. But someone may find an endangered plant species which will be threatened by reduced sun exposure and make a case to block the development.

Note: It is widely assumed that distributed solar (panels on rooftops) will be an important competitor to power from the grid. But, if the centralized storage in hydrogen is much less expensive than distributed storage, the grid may still maintain an important role for producing and delivering electricity. If the solar farm and the roof top have the same low cost solar PV output**, the central generation system may still be the low-cost system for all the heavily overcast and nighttime hours if central storage is cheaper by a sufficient margin to offset distribution costs through the grid.

Weather effects can be averaged out – it may be overcast in Missouri and sunny in Oklahoma, for example. And of course, if the grid is large enough (say U.S. national coverage) then the potential hours of summer sunlight can be from 7 am in Boston to 8 pm in Los Angeles so that there can be a wider range of hours that electricity is used directly as produced can be realized, reducing the amount of storage that is needed.

The ultimate economics of solar power have many variables yet to be refined and defined. But from what is known now the potential is very promising. Rooftop solar has already been making itself felt. Giles Parkinson wrote last year in The Guardian:

Last week, for the first time in memory, the wholesale price of electricity in Queensland fell into negative territory – in the middle of the day. For several days the price, normally around $40-$50 a megawatt hour, hovered in and around zero. Prices were deflated throughout the week, largely because of the influence of one of the newest, biggest power stations in the state – rooftop solar.

It is too early to answer the question: Will solar farms come to dominate utility power generation? But there is a distinct possibility that the answer is yes and perhaps within the next ten years or so.

* At 200 acres the farm pictured could supply all the power needed by 4 million homes using the Toshiba hydrogen fuel cell facility. Of course up to 1/3 of the power would be supplied as generated and not go through the storage facility, thus lowering the average unit cost of electricity.

** Estimates are that the generation costs at utility scale (solar farms) are from 25% to 50% of the costs for residential rooftops.

John Lounsbury

Sources: