5 ms·

Here's the simplest way to get to net zero: solar, wind, batteries, natural gas, then hydrogen. Solar and wind are cheap, but intermittent. Batteries can help

Here's the simplest way to get to net zero: solar, wind, batteries, natural gas, then hydrogen. Solar and wind are cheap, but intermittent. Batteries can help store energy from day to night. To help store energy from summer to winter, they need to become 1000 times cheaper, which is not going to happen. For that you need peaker plants. Initially methane-based, later they can be converted to hydrogen. It's as simple as that.

fivea 5y ago
> Solar and wind are cheap, but intermittent. Batteries can help store energy from day to night. You don't need batteries per se, only energy storage systems. Pumped storage, pressurized air, flywheels, ultracapacitors, and of course chemical batteries. Pumped storage and compressed air systems store orders of magnitude more energy and power than batteries.
- xyzzyz 5y ago
  Pumped storage has a serious problem: you can’t build it just about anywhere. There is very limited number of sites, and the most practical ones are already built out. This also ignores the issue of environmentalists being almost certain to block destroying entire valleys or mountain ranges.
  - Taniwha 5y ago
    what you can do is retrofit existing hydro dams, add more generation capacity (penstocks/turbines) and then let them fill when the sun shines/wind blows and empty when it doesn't, this is actually more efficient than pumping water uphill
  - ncmncm 5y ago
    There are hundreds or thousands of times as many places to store pumped hydro as there are places to put hydroelectric dams. The latter needs a river valley and a watershed. The former only needs an elevated basin. An elevated box canyon would do; you just build a dam across the end. Sometimes, you dig a pit, or build a dam all around a flat hilltop. Furthermore, underground reservoirs have been demonstrated, in both natural and abandoned mining cavities. A hollow sphere anchored to the sea bed can have the water pumped out with surplus power, and allowed back in to generate power. Draining water from an elevated reservoir to a deep underground cavern can multiply the energy storage capacity. Height is strictly limited to hill height, but underground cavities may have any depth, to the limits of drilling.
- credit_guy 5y ago
  > You don't need batteries per se, only energy storage systems. That's true, but there are synergies. We will need lots and lots of batteries for electrifying the transportation. There's lots of R&D money in battery research. Much less in other types of storage. There's all the reasons to believe batteries will become cheaper in the long run by a factor of 2, or even more [1]. There are also a few advantages that batteries have over other types of storage: they don't have moving parts, they have quite high round-trip efficiency, their maintenance is cheap, the failure mode is not that bad. [1] https://www.nrel.gov/docs/fy21osti/79236.pdf https://www.nrel.gov/docs/fy21osti/79236.pdf
  - Ekaros 5y ago
    Realistically you want different types of batteries for grid-scale or even home scale storage and transportation. With static installation mass and volume are little of concern, where as with transport they are critical. On other hand scale is also big difference. And lithium batteries in current form do have many risks. I can't even imagine what sort of project it would be to even control a fire in grid-scale installation...
    - credit_guy 5y ago
      > I can't even imagine what sort of project it would be to even control a fire in grid-scale installation... You don't need to imagine it. Such a fire already happened. It was brought under control. https://www.cnbc.com/2021/08/05/tesla-megapack-fire-highlights-early-stage-issues-with-big-batteries.html https://www.cnbc.com/2021/08/05/tesla-megapack-fire-highligh...
    - 41b696ef1113 5y ago
      >I can't even imagine what sort of project it would be to even control a fire in grid-scale installation... That is going to be true regardless of storage technology. Concentrating energy into a tiny volume is effectively making a bomb.
      - ncmncm 5y ago
- tialaramex 5y ago
  You need systems with different performance for different problems. In particular although spin-up time for pumped storage is much better than for fossil fuels (pumped storage may be able to go from nothing to 80% power in under a minute, lots of coal stations take an hour, combined cycle gas might be 20 minutes) you can't sit around for a minute or the lights go out, batteries can do that same-cycle balancing you need to give other systems time to come online. Pumped storage in particular is also geographically dependent. If you've got lots of spare lakes on mountains, pumped storage is just what you needed, if you're the Dutch not so much.
- buu700 5y ago
  I'm personally a huge fan of carbon capture -> synthetic hydrocarbons + fossil fuel plants, in theory. Maybe a mix of storage systems will ultimately be the best solution for whatever reasons, but this idea makes so much sense to me because of how it takes advantage of all of our existing infrastructure. Just stand up renewables at existing plants, use spare capacity to power capturing CO2 from the output and/or the air, formulate the captured CO2 into a usable fuel on site, and then swap out the dirty fuel supply for the synthetic one. Now you have the functional equivalent of renewables + batteries — with the advantages that the plants are already built and wired into the grid, battery supply is no longer a bottleneck to rollout, and a lot of jobs are saved. So as a whole, as I see it, this would be faster, easier to plan, less wasteful, and more politically viable, as compared to replacing existing plants with brand new renewable ones that use battery storage. The same concept applies to vehicles. Our entire infrastructure around gasoline is already sitting and waiting to be used as a massive energy storage system. If carbon could be captured from the air using renewables and centrally processed into a backwards-compatible fuel, it would no longer be necessary to push so heavily for a transition to BEVs. Instead, consumers could choose solely based on the costs and benefits to themselves; maybe gas cars would be less efficient than BEVs (like the equivalent of running an app in Wine or Rosetta), but that would be reflected in the price of the fuel if so, with consumers eating the full cost rather than passing part of it along to society as a negative externality. Ditto for existing gas-powered stoves, heating, etc. Additionally, once it's running at scale, this system would naturally lend itself to the next step: transitioning from net-zero emissions to net-negative emissions. Governments could allocate funds toward buying carbon capture fuel and simply sequestering it away for emergencies, at least until atmospheric CO2 drops to pre-industrial levels. Maybe the same method would prove useful for terraforming other planets with inhospitable atmospheres as well. I'm very interested in seeing how United's experiments with synthetic fuel go for these reasons. Last time I brought this topic up on HN, someone shared Prometheus Fuels with me: https://en.wikipedia.org/wiki/Prometheus_Fuels https://en.wikipedia.org/wiki/Prometheus_Fuels. Does anyone know how far along this kind of technology is? What factors have prevented the idea from already being implemented in traditional power plants at scale?
  - aaronblohowiak 5y ago
    Imagine a lossless conversion from electricity to making hydrocarbons from atmospheric co2. Car engines are very very inefficient (about 35% of chemical energy converted to mechanical, out of a theoretical max of 46% or so iiuc.) comparatively, electric motors are like 80% efficient. Now it’s true that bev have transmission and battery losses, but synthetic gasoline is going to have inefficiencies and cost to transport, too. The case wher e it makes sense is where the weight of batteries matter, ie: jet packs and airplanes.
ipaddr 5y ago
Nuclear power can offer consistent base loads to power grids.
- pfdietz 5y ago
  Renewables + storage can provide synthetic baseload, and likely will be able to do it more cheaply than new nuclear plants would. https://model.energy/ https://model.energy/
  - edelans 5y ago
    One problem of renewables is the energy density. 1 GigaWatt nuclear reactor takes 10 acres. 1 Gigawatt equivalent of solar panels is 2500 acres. All those solar panels require much more metals and cement than the nuclear plant. And the lifetime of solar panels is also twice shorter than the nuclear plant.
    - pfdietz 5y ago
      Utility scale PV doesn't require any cement/concrete at all. The mountings can be anchored to the ground with steel earth anchors. This is typically cheaper than using concrete, and faster to install. The area requirement ignores that land is extremely cheap, as little as 1% of the cost of a utility-scale PV field in the US. Building an expensive nuclear plant to save cheap land is deeply foolish.
      - edelans 5y ago
        > Building an expensive nuclear plant to save cheap land is deeply foolish. it depends. When it competes with other usage of the land (crops, biodiversity in forests...), that's not foolish. Agree with you that each country has a different situation though.
- credit_guy 5y ago
  I'm pro nuclear. But the fact is right now nuclear is too expensive. It just can't compete. With the new generation of small modular reactors, it might become competitive, or it might not. We just don't know. We simply can't bet the house on nuclear. But if the private sector wants to pursue nuclear (be it fission or fusion), I'm all for it.
elihu 5y ago
Storing energy in the summer to be used in the winter probably isn't going to ever be practical with batteries. (Maybe it's practical by using electricity to make synthetic liquid fuels? That wouldn't be very efficient, though.) Batteries are useful for buffering shorter-term variations, like the day/night cycle. There are some alternatives, though. One is to just build enough spare capacity into the system that one can still run a surplus on cloudy winter days when the winds aren't blowing. Better I think is to connect power grids to even loads around. Barring megavolcano eruptions or nuclear winter, the sun is always shining somewhere in the world. And the wind is always blowing somewhere. I think the thing that's needed most right now (aside from just more solar or wind plants everywhere) is high-capacity transcontinental grid connections. The United States should be able to buy solar power from the Sahara when it's night here, and sell surplus solar to Europe and Asia when the sun is shining here. This makes renewable energy much more attractive because there's always a buyer or a seller somewhere, and batteries aren't strictly needed. Fossil fuel plants can continue to exist as emergency backup power sources in case something happens to the HVDC links. Ideally we'd get to the point where they never need to be used. Apparently Chile and China are working on a plan whereby Chile supplies China with solar power via a trans-pacific HVDC line [1]. We need more of that: connections between North America and Europe, connections between South America and Africa, and so on. [1] https://news.ycombinator.com/item?id=29230066 https://news.ycombinator.com/item?id=29230066
- pfdietz 5y ago
  > Maybe it's practical by using electricity to make synthetic liquid fuels? Hydrogen, which can be stored underground like natural gas. https://power.mhi.com/regions/amer/news/20210511.html https://power.mhi.com/regions/amer/news/20210511.html "The project’s salt caverns will be capable of holding more than 5,500 metric tonnes of hydrogen. From an energy storage perspective, one cavern holds the equivalent of 150 gigawatt hours (GWh) of carbon-free dispatchable energy and/or decarbonized fuel that can be used in other industries. By comparison, a U.S. Energy Information Administration (EIA) 2020 report estimates the current installed base of battery energy storage across the U.S. at 1.2 GWh." (The formation there has room for 100 such caverns.) (Also: "metric tonnes"?!) https://www.powermag.com/large-scale-green-hydrogen-infrastructure-system-unveiled-for-california/ https://www.powermag.com/large-scale-green-hydrogen-infrastr...
goodpoint 5y ago
> they need to become 1000 times cheaper 1000 times? Citation desperately needed.
- credit_guy 5y ago
  There are 365 days in a year. A battery that you charge and discharge every day will make money daily for you. A battery that you charge and discharge once a year, will only make money once a year for you. Plus, it will also lose quite a lot of energy due to self-discharge. That's it, in a nutshell. To make it concrete, a Tesla Megapack has about 20 MWh of energy, costs about $7 MM (including maintenance) and has a lifetime of 10 years [1]. At the end of its lifetime the energy drops to about 70% of the initial charge, but I'll ignore that. So, the price per kWh comes down to 7000000 / (365 x 10 x 20000) = 10 cents. That's somewhat steep, but acceptable. The average price of 1kWh in the US last year was about 14 cents and in the state of New York about 20 cents [2]. If you manage to generate electricity with solar panels for less than 10 cents/kWh, then solar+short term batteries are competitive. If you can only sell energy once a year, then the cost comes down to 365 times higher, or about $35/kWh. [1] https://www.tesla.com/megapack/design https://www.tesla.com/megapack/design [2] https://www.bls.gov/regions/new-york-new-jersey/news-release/averageenergyprices_newyorkarea.htm https://www.bls.gov/regions/new-york-new-jersey/news-release...
edelans 5y ago
I'm curious : where does the hydrogen come from ? You mean you could use it as energy storage between summer and winter ? (in summer, excess of solar / wind production is used to produce H2 by Electrolysis (https://en.wikipedia.org/wiki/Hydrogen_production#Electrolysis https://en.wikipedia.org/wiki/Hydrogen_production#Electrolys... ), then you store H2, and then next winter you consume it to produce energy again ? Not sure it works well, because those reactions are not 100% efficient (so you have to take that loss into account, requiring more energy production) and then transporting/storing hydrogen is hard : - either you compress it and you don't transport many of if - either you cool it to make it liquid and then it takes a lot of energy

fivea 5y ago

> Solar and wind are cheap, but intermittent. Batteries can help store energy from day to night. You don't need batteries per se, only energy storage systems. Pumped storage, pressurized air, flywheels, ultracapacitors, and of course chemical batteries. Pumped storage and compressed air systems store orders of magnitude more energy and power than batteries.

xyzzyz 5y ago
Pumped storage has a serious problem: you can’t build it just about anywhere. There is very limited number of sites, and the most practical ones are already built out. This also ignores the issue of environmentalists being almost certain to block destroying entire valleys or mountain ranges.
- Taniwha 5y ago
  what you can do is retrofit existing hydro dams, add more generation capacity (penstocks/turbines) and then let them fill when the sun shines/wind blows and empty when it doesn't, this is actually more efficient than pumping water uphill
- ncmncm 5y ago
  There are hundreds or thousands of times as many places to store pumped hydro as there are places to put hydroelectric dams. The latter needs a river valley and a watershed. The former only needs an elevated basin. An elevated box canyon would do; you just build a dam across the end. Sometimes, you dig a pit, or build a dam all around a flat hilltop. Furthermore, underground reservoirs have been demonstrated, in both natural and abandoned mining cavities. A hollow sphere anchored to the sea bed can have the water pumped out with surplus power, and allowed back in to generate power. Draining water from an elevated reservoir to a deep underground cavern can multiply the energy storage capacity. Height is strictly limited to hill height, but underground cavities may have any depth, to the limits of drilling.
credit_guy 5y ago
> You don't need batteries per se, only energy storage systems. That's true, but there are synergies. We will need lots and lots of batteries for electrifying the transportation. There's lots of R&D money in battery research. Much less in other types of storage. There's all the reasons to believe batteries will become cheaper in the long run by a factor of 2, or even more [1]. There are also a few advantages that batteries have over other types of storage: they don't have moving parts, they have quite high round-trip efficiency, their maintenance is cheap, the failure mode is not that bad. [1] https://www.nrel.gov/docs/fy21osti/79236.pdf https://www.nrel.gov/docs/fy21osti/79236.pdf
- Ekaros 5y ago
  Realistically you want different types of batteries for grid-scale or even home scale storage and transportation. With static installation mass and volume are little of concern, where as with transport they are critical. On other hand scale is also big difference. And lithium batteries in current form do have many risks. I can't even imagine what sort of project it would be to even control a fire in grid-scale installation...
  - credit_guy 5y ago
    > I can't even imagine what sort of project it would be to even control a fire in grid-scale installation... You don't need to imagine it. Such a fire already happened. It was brought under control. https://www.cnbc.com/2021/08/05/tesla-megapack-fire-highlights-early-stage-issues-with-big-batteries.html https://www.cnbc.com/2021/08/05/tesla-megapack-fire-highligh...
  - 41b696ef1113 5y ago
    >I can't even imagine what sort of project it would be to even control a fire in grid-scale installation... That is going to be true regardless of storage technology. Concentrating energy into a tiny volume is effectively making a bomb.
    - ncmncm 5y ago
tialaramex 5y ago
You need systems with different performance for different problems. In particular although spin-up time for pumped storage is much better than for fossil fuels (pumped storage may be able to go from nothing to 80% power in under a minute, lots of coal stations take an hour, combined cycle gas might be 20 minutes) you can't sit around for a minute or the lights go out, batteries can do that same-cycle balancing you need to give other systems time to come online. Pumped storage in particular is also geographically dependent. If you've got lots of spare lakes on mountains, pumped storage is just what you needed, if you're the Dutch not so much.
buu700 5y ago
I'm personally a huge fan of carbon capture -> synthetic hydrocarbons + fossil fuel plants, in theory. Maybe a mix of storage systems will ultimately be the best solution for whatever reasons, but this idea makes so much sense to me because of how it takes advantage of all of our existing infrastructure. Just stand up renewables at existing plants, use spare capacity to power capturing CO2 from the output and/or the air, formulate the captured CO2 into a usable fuel on site, and then swap out the dirty fuel supply for the synthetic one. Now you have the functional equivalent of renewables + batteries — with the advantages that the plants are already built and wired into the grid, battery supply is no longer a bottleneck to rollout, and a lot of jobs are saved. So as a whole, as I see it, this would be faster, easier to plan, less wasteful, and more politically viable, as compared to replacing existing plants with brand new renewable ones that use battery storage. The same concept applies to vehicles. Our entire infrastructure around gasoline is already sitting and waiting to be used as a massive energy storage system. If carbon could be captured from the air using renewables and centrally processed into a backwards-compatible fuel, it would no longer be necessary to push so heavily for a transition to BEVs. Instead, consumers could choose solely based on the costs and benefits to themselves; maybe gas cars would be less efficient than BEVs (like the equivalent of running an app in Wine or Rosetta), but that would be reflected in the price of the fuel if so, with consumers eating the full cost rather than passing part of it along to society as a negative externality. Ditto for existing gas-powered stoves, heating, etc. Additionally, once it's running at scale, this system would naturally lend itself to the next step: transitioning from net-zero emissions to net-negative emissions. Governments could allocate funds toward buying carbon capture fuel and simply sequestering it away for emergencies, at least until atmospheric CO2 drops to pre-industrial levels. Maybe the same method would prove useful for terraforming other planets with inhospitable atmospheres as well. I'm very interested in seeing how United's experiments with synthetic fuel go for these reasons. Last time I brought this topic up on HN, someone shared Prometheus Fuels with me: https://en.wikipedia.org/wiki/Prometheus_Fuels https://en.wikipedia.org/wiki/Prometheus_Fuels. Does anyone know how far along this kind of technology is? What factors have prevented the idea from already being implemented in traditional power plants at scale?
- aaronblohowiak 5y ago
  Imagine a lossless conversion from electricity to making hydrocarbons from atmospheric co2. Car engines are very very inefficient (about 35% of chemical energy converted to mechanical, out of a theoretical max of 46% or so iiuc.) comparatively, electric motors are like 80% efficient. Now it’s true that bev have transmission and battery losses, but synthetic gasoline is going to have inefficiencies and cost to transport, too. The case wher e it makes sense is where the weight of batteries matter, ie: jet packs and airplanes.

ipaddr 5y ago

Nuclear power can offer consistent base loads to power grids.

pfdietz 5y ago
Renewables + storage can provide synthetic baseload, and likely will be able to do it more cheaply than new nuclear plants would. https://model.energy/ https://model.energy/
- edelans 5y ago
  One problem of renewables is the energy density. 1 GigaWatt nuclear reactor takes 10 acres. 1 Gigawatt equivalent of solar panels is 2500 acres. All those solar panels require much more metals and cement than the nuclear plant. And the lifetime of solar panels is also twice shorter than the nuclear plant.
  - pfdietz 5y ago
    Utility scale PV doesn't require any cement/concrete at all. The mountings can be anchored to the ground with steel earth anchors. This is typically cheaper than using concrete, and faster to install. The area requirement ignores that land is extremely cheap, as little as 1% of the cost of a utility-scale PV field in the US. Building an expensive nuclear plant to save cheap land is deeply foolish.
    - edelans 5y ago
      > Building an expensive nuclear plant to save cheap land is deeply foolish. it depends. When it competes with other usage of the land (crops, biodiversity in forests...), that's not foolish. Agree with you that each country has a different situation though.
credit_guy 5y ago
I'm pro nuclear. But the fact is right now nuclear is too expensive. It just can't compete. With the new generation of small modular reactors, it might become competitive, or it might not. We just don't know. We simply can't bet the house on nuclear. But if the private sector wants to pursue nuclear (be it fission or fusion), I'm all for it.

elihu 5y ago

Storing energy in the summer to be used in the winter probably isn't going to ever be practical with batteries. (Maybe it's practical by using electricity to make synthetic liquid fuels? That wouldn't be very efficient, though.) Batteries are useful for buffering shorter-term variations, like the day/night cycle. There are some alternatives, though. One is to just build enough spare capacity into the system that one can still run a surplus on cloudy winter days when the winds aren't blowing. Better I think is to connect power grids to even loads around. Barring megavolcano eruptions or nuclear winter, the sun is always shining somewhere in the world. And the wind is always blowing somewhere. I think the thing that's needed most right now (aside from just more solar or wind plants everywhere) is high-capacity transcontinental grid connections. The United States should be able to buy solar power from the Sahara when it's night here, and sell surplus solar to Europe and Asia when the sun is shining here. This makes renewable energy much more attractive because there's always a buyer or a seller somewhere, and batteries aren't strictly needed. Fossil fuel plants can continue to exist as emergency backup power sources in case something happens to the HVDC links. Ideally we'd get to the point where they never need to be used. Apparently Chile and China are working on a plan whereby Chile supplies China with solar power via a trans-pacific HVDC line [1]. We need more of that: connections between North America and Europe, connections between South America and Africa, and so on. [1] https://news.ycombinator.com/item?id=29230066 https://news.ycombinator.com/item?id=29230066

pfdietz 5y ago
> Maybe it's practical by using electricity to make synthetic liquid fuels? Hydrogen, which can be stored underground like natural gas. https://power.mhi.com/regions/amer/news/20210511.html https://power.mhi.com/regions/amer/news/20210511.html "The project’s salt caverns will be capable of holding more than 5,500 metric tonnes of hydrogen. From an energy storage perspective, one cavern holds the equivalent of 150 gigawatt hours (GWh) of carbon-free dispatchable energy and/or decarbonized fuel that can be used in other industries. By comparison, a U.S. Energy Information Administration (EIA) 2020 report estimates the current installed base of battery energy storage across the U.S. at 1.2 GWh." (The formation there has room for 100 such caverns.) (Also: "metric tonnes"?!) https://www.powermag.com/large-scale-green-hydrogen-infrastructure-system-unveiled-for-california/ https://www.powermag.com/large-scale-green-hydrogen-infrastr...

goodpoint 5y ago

> they need to become 1000 times cheaper 1000 times? Citation desperately needed.

credit_guy 5y ago
There are 365 days in a year. A battery that you charge and discharge every day will make money daily for you. A battery that you charge and discharge once a year, will only make money once a year for you. Plus, it will also lose quite a lot of energy due to self-discharge. That's it, in a nutshell. To make it concrete, a Tesla Megapack has about 20 MWh of energy, costs about $7 MM (including maintenance) and has a lifetime of 10 years [1]. At the end of its lifetime the energy drops to about 70% of the initial charge, but I'll ignore that. So, the price per kWh comes down to 7000000 / (365 x 10 x 20000) = 10 cents. That's somewhat steep, but acceptable. The average price of 1kWh in the US last year was about 14 cents and in the state of New York about 20 cents [2]. If you manage to generate electricity with solar panels for less than 10 cents/kWh, then solar+short term batteries are competitive. If you can only sell energy once a year, then the cost comes down to 365 times higher, or about $35/kWh. [1] https://www.tesla.com/megapack/design https://www.tesla.com/megapack/design [2] https://www.bls.gov/regions/new-york-new-jersey/news-release/averageenergyprices_newyorkarea.htm https://www.bls.gov/regions/new-york-new-jersey/news-release...

edelans 5y ago

I'm curious : where does the hydrogen come from ? You mean you could use it as energy storage between summer and winter ? (in summer, excess of solar / wind production is used to produce H2 by Electrolysis (https://en.wikipedia.org/wiki/Hydrogen_production#Electrolysis https://en.wikipedia.org/wiki/Hydrogen_production#Electrolys... ), then you store H2, and then next winter you consume it to produce energy again ? Not sure it works well, because those reactions are not 100% efficient (so you have to take that loss into account, requiring more energy production) and then transporting/storing hydrogen is hard : - either you compress it and you don't transport many of if - either you cool it to make it liquid and then it takes a lot of energy