5 ms·
> How do you envision energy storage of the future? Not the guy you responded to, but: a combination of traditional batteries, molten metal batteries, liquid a
by audunw 4y ago
> How do you envision energy storage of the future?
Not the guy you responded to, but: a combination of traditional batteries, molten metal batteries, liquid air or CO2 storage, pumped hydro, gravity storage, stored thermal energy, and more. All of which are around the commercial demonstration plant phase right now.
> Where are you going to get the metals?
Many of these don't require much metals. Molten metal batteries use metals that are extremely abundant.
> How much fossil fuel is going to burn in order to extract it?
In a decarbonized world? Zero. What? You think climate change can be solved without making mining zero-emission? If you're wondering how this will be done, it'll be battery/hydrogen/ammonia/e-fuel for mining equipment, trucks, ships, etc. We have to do that no matter what, otherwise we've just postponed climate change, not solved it.
> How else do you provide base load requirements?
Personally I believe a good share of base load will be provided by nuclear in many countries. I have nothing against nuclear. But I also think the base load problem can be solved without nuclear quite easily, assuming we actually solve CO2-emissions. This is because solving CO2-emissions means we'll produce batteries/hydrogen/ammonia/e-fuels on the same order of magnitude needed to balance renewables to provide baseload.
If you want to dive into more detail, look at Marc Z Jacobsens studies: https://www.sciencedirect.com/science/article/abs/pii/S0960148121016499?via%3Dihub https://www.sciencedirect.com/science/article/abs/pii/S09601...
I think advanced geothermal may become a significant part of renewable base load in the future. It would be a huge hail mary for the climate change cause, because it'd make it SO much easier to get political willpower and investments from the whole oil/gas-sector. Check this out: https://www.youtube.com/watch?v=n2P2stuQ_KY https://www.youtube.com/watch?v=n2P2stuQ_KY
> It's naive to think we can transition to a "green grid" without nuclear.
Optimistic, but not naive. There's a clear path. Difficult, but not much more difficult than rebooting the nuclear energy industry.
And you have to be optimistic to think we can get to zero CO2-emissions anyway.
- catiopatio 4y agoThis is science fiction-thinking. Not just optimistic: absolutely naive. We have the tech for nuclear, today. In fact, we could have switched the entire country over to nuclear 30+ years ago. Instead, we’ve been burning fossil fuels for decades because, for so-called environmental activists, an impossible perfect solution is the only thing they’ll accept.
- epistasis 4y agoRun the numbers, time tracking throughout the day, on how you're going to run an entire grid on nuclear. Calculate how much storage you need in order to convert baseload into something that matches demand. Do the same calculation for renewables. Both need storage, and renewables need a bit more storage, but their primary energy is also 5-10x cheaper than nuclear. Calling something "naive" or "fantasy" requires evaluating the current state of the tech, and where the tech is going. From that perspective, especially with the data coming from the nuclear build at Vogtle and Summer, thinking that nuclear GenIII+ reactors have any place on the grid is completely unrealistic. We can not even build four of these nuclear reactors . We started plans to build about a dozen, started on only four, and had to abandon two mid-build. Nuclear is not a good fit for advanced economies, anymore than complex Victorian style wood carving is a fit for advanced economies. Nuclear requires way too much skilled labor, too much construction versus manufacturing. We no longer live in the 80s, we have much better tech, 40 years of advancement, and we need to use the best tech, not the one that was best in 1980.
- Dylan16807 4y ago> Run the numbers, time tracking throughout the day, on how you're going to run an entire grid on nuclear. Calculate how much storage you need in order to convert baseload into something that matches demand. Very tiny amounts are needed. It's pretty straightforward to make reactors that can ramp along with daily power use.