Submitted by Taps Coogan on the 19th of November 2019 to The Sounding Line.
Enjoy The Sounding Line? Click here to subscribe for free.
Enjoy The Sounding Line? Click here to subscribe.
Due to the fact that the sun doesn’t always shine and the wind doesn’t always blow, schemes to power America’s electricity grid exclusively on wind and solar inevitably end up relying on very large battery backup systems.
While daily wind generation data for the entire US is not readily available, UK wind generation data illustrates the point that it is not unusual to have windless days over large geographic areas. It is typical to have at least one day a month with negligible wind generation across the entire UK, including its numerous off-shore wind farms. It follows that, when windless periods occur during cloudy periods or at night, wind and solar generation could be negligible for much of the US.
So, how much would it cost to purchase enough lithium-ion batteries to power the US for a typical one hour period? A back-of-the-envelope answer is actually fairly straightforward to calculate:
Accordingly to the IEA, the US used roughly 11.2 Quadrillion Watt-Hours of electricity in 2018 (38.3 Quads), or roughly 1.27 Trillion Watt-Hours of energy per hour (TWh). Accordingly to the National Renewable Energy Laboratory, the cost of grid scale lithium ion batteries in 2018 is reportedly around $209 per kWh. For simplicity sake, let’s ignore the other costs associated with installing battery storage system, which can double or triple that price.
The results: Based on a battery price of $209 per kWh, it would cost just over $250 billion to purchase enough lithium ion batteries to store enough energy to back up the US electric grid for one hour. To backup the grid for one windless night (12 hours) would cost just over $3 trillion (probably a bit less because energy demand is lower at night). To provide 100% backup for just 1% of the year (3.6 days of consecutive backup) would cost about $22 trillion. While three entirely cloudy windless days in a row across all of the US is pretty unlikely, something approaching it is not unlikely at all. At times, North American can be a cloudy place.
The IEA estimates that lithium-ion battery prices may fall to roughly $117 per kWh by 2040. If that optimistic forecast is correct, in 2040 it would cost roughly $140 billion per hour, $1.7 trillion per night, or $12 trillion per 1% of the year, to backup the US electric grid with lithium ion batteries (assuming no growth in electricity demand). Remember, that excludes all of the ancillary equipment and installation costs which can double or triple those prices.
Keep in mind that those batteries typically only have a 10 to 20 year lifespan and, to accommodate for the large seasonal variations in wind and solar generation, there would need to be enough storage to power a significant portion of the grid for months at a time.
Spending tens of trillion of dollars on batteries, only to replace them every ten to twenty years, is simply not going to happen, and probably wouldn’t be great for the environment if it did. That doesn’t mean that wind and solar won’t continue to grow in importance. They can provide a large minority of the nation’s power, without needing battery backup. However, 100% wind and solar backed-up by lithium-ion batteries remains science fiction.
Would you like to be notified when we publish a new article on The Sounding Line? Click here to subscribe for free.
Two words… molten salt…
Indeed, they do look really promising for concentrating solar (CSP). CSP is about three times more expansive than photovoltaic though.
http://euanmearns.com/concentrated-solar-power-vs-solar-pv-an-update/
Interesting that in that link they only consider PV not PV+Storage (and spends plenty of time in the aricle and comments dismissing storage), where CSP includes storage… so you have to add your $117/kWh (in 2040) to the 3x 0.2/kWh, where as there are CSP plants being built now (in 2019) with install costs at $71/MWh [0]
[0] https://www.nrel.gov/docs/fy19osti/74585.pdf