Solar power is not cheaper than combined-cycle natural gas at midnight. The article that you cited does not contain the words "battery" or "storage", and one really needs to compare the all-in levelized cost of energy (and storage) over the life of the project. It does mention hydrogen, but does not go into the efficiency calculations.
There are 280 million vehicles in the US. If fifty million of those were electric vehicles, then ten years later we'd have enough already-produced batteries to power the entire United States overnight.
There were 300,000 EVs sold in the US in 2020. There were 600,000 in 2021.
This doesn't help right now, but over time, it's becoming clear there will be a flood of cheap batteries available for grid storage.
As a nice side effect, as the market for these used batteries takes off, this might help to assuage fears about expensive battery pack replacements, fueling more EV adoption.
In this dimension, EVs that charge during the midday solar peak and overnight from wind will make an important contribution to rolling out reneweables, cheaper energy and ditching oil and gas.
I'll quote a brief section:
> In general, batteries will be retired from use in an EV when the range and performance is no longer acceptable to the driver. The remaining capacity of the battery at the time of retirement will vary depending on consumer preference, but it is generally assumed to be between 70- 80%.
> Given the large capacity and high performance of modern vehicle batteries, retired batteries could still offer significant value in lower-power, secondary applications, such as storing energy from solar panels to be used in off-grid or peak demand-shaving applications.37
> A growing body of research has examined the environmental impacts and technical and economic feasibility of repurposing batteries for use in second-life applications.38–42 Since repurposed batteries are a relatively new phenomenon, data about their performance is uncertain, particularly because of the uneven degradation of battery cells over time. However, it is estimated that battery lifespan can be extended by 10 years or longer depending on the application.38,43,44
[1] https://calepa.ca.gov/wp-content/uploads/sites/6/2021/12/Fin...
All it takes is for a car scrap dealer to realise that for every EV battery he grabs out and hooks up to the grid he'll earn $200/year. Before long, every scrap dealer will be doing it to every car.
However, I suspect that small players will be barred from the electricity markets via mountains of bureaucracy and certification, so that might not happen.
This type of analysis is pretty difficult though. A modern grid with both solar and wind power is pretty resilient and calculating the amount of storage needed depends on the things like if it has long distance transmission lines (increasing reliance by increasing the geography).
I think this is why a lot of places end up having a gas peaker plant. In terms of running costs it's not great, and the climate impact is bad (although I guess hydrogen helps is that is available anywhere yet). But it's cheap to build ($85M gets you 55MW[2]). These days batteries are cost competitive with that though[3]:
> a 250MW, four-hour (1,000MWh) battery system in New South Wales would be a cheaper option for meeting peak demand than a 250MW new-build OCGT from both levelised cost of energy (LCOE) and levelised cost of capacity (LCOC) perspectives.[3]
[1] https://www.lazard.com/perspective/levelized-cost-of-energy-... (scroll down for storage)
[2] https://www.wbur.org/news/2022/04/08/peabody-peaker-natural-...
[3] https://www.energy-storage.news/battery-storage-30-cheaper-t...
