Constructing green charging stations is a step in the right direction, but this is a hard problem. I think one of the problems is that we don’t appreciate the scale of the problem of refueling infrastructure. Part of this is because there’s been a century of build-out for support of internal combustion engines, and this is an attempt to hit a critical mass in a much shorter time.
The Tesla supercharger stations have a transfer capacity of 100 kW and shoot for a 30 minute turnaround, which is supposed to provide 3 hours of driving. That’s 180 MJ of energy (50 kWh). How does it compare to gasoline? Gas contains around 132 MJ per gallon, and the EPA allows gasoline pumps to transfer no more than 10 gallons per minute, which gives us a transfer rate of 22 MegaWatts, though actual pumping speeds, and thus rates, are likely somewhat smaller. Still, 10-15 MW is a lot of power, and that’s what you’re transferring when you fill your tank.
One thing that electric cars have going for them is that they are significantly more efficient than gasoline — there’s an inherently higher efficiency and technology like regenerative braking, plus the ability to just turn off rather than idling. Overall, electric cars are around 5x as efficient as gasoline-powered vehicles. Good thing, too, because otherwise we’d have close to a 1:1 charge:travel time, and nobody is going to put up with that.
So we can fill up a tank of gas in just a couple of minutes, and it takes more than an order of magnitude longer for electric, for a more limited range. Let’s look at this from another perspective: waiting for a space to clear at a gas station is annoying even when it’s a few minutes, so waiting for a fill-up that takes 30 is probably a nonstarter. Which means that you are going to need proportionally more fill-up bays at each station, relative to the number of cars on the road. Right now that capacity is not a problem, with so few cars, but it’s an obstacle to wider adoption.
If you have a station capable of charging up multiple cars, you need to be able to deliver the power. For every 10 cars at once that means a MegaWatt of electricity. Perhaps that remains constant — if you can cut the charging time in half you deliver twice the power but don’t need as many charging stations, and you won’t be operating at peak capacity, so that doesn’t mean you need 1 MW coming in — you can store electricity when you have lower demand. But you have to generate all that electricity from solar, though you at least have the advantage of staying DC to charge up a car — no inverter losses as you’d have for a home system running your 60Hz loads. But how many cars are you going to handle? 100 a day? That’s 5000 kWh, and solar might generate 5-10 kWh/m^2 each day, (or even less) depending on location and time of year. That’s a 1000 m^2 solar array for the smaller value, and you probably need more in case the weather is bad for more than a day lest you tell your customers “We’re out of sun” very often. If you’re out, they are stranded, so I imagine there will be emergency generators (running on biodiesel, presumably). The array size may not be a problem for stations away from cities. You have the space, and don’t need to have a grid connection, so you are freer to put these where you want. In more occupied space, you’d tap into the grid if you needed to, though that’s not “green” and defeats (much of) the purpose of having an electric car. But being stranded is not going to be an option.