How will we ship solar energy?

Oil is great, could blimps or death rays be greater?

I wrote a follow up to this idea here.

Some places are blessed with more solar energy than others. Australia and the Sahara, for example, have a huge opportunity at their hands. While solar resources further north are still substantial, specialization may mean that some of the worlds energy is produced in sunny countries and shipped elsewhere.

Source

Trading energy is pretty easy with fossil fuels, you can just put them in a tank and ship them anywhere. Taiwan gets 98% of its energy from imported oil and gas!

But how are we going to ship sunlight around the world? Solar is so much cheaper than fossil fuels that it requires a different approach. Just look at what’s happening in energy generation. Right now, panels are so cheap that their metal racks are becoming a significant fraction of costs. So now, the Big New Innovation in solar is throwing the panels on the ground without any racks.

Shipping energy has a similar problem. Even cost-effective transmission methods can limit the gains from solar. For example, Singapore and Australia are building a direct “sun line” connecting Australia’s cheap solar energy to growing markets in southeast Asia. Similar projects have been proposed to connect the Sahara to Europe.

But transmitting energy over such long distances is tricky. The cable itself is expensive to build and maintain, and stepping up the voltage from panels to the HVDC line requires expensive transformers and some energy loss. And you’ll need batteries to match supply with demand. All of these costs hurt the case for solar energy. It doesn’t matter if solar can deliver energy for free if the transmission system adds $50/MWh.

That’s why many industries and communities will generate their energy on-site rather than deal with the grid. If they can orient around intermittent solar, they can lower costs significantly.

But shipping energy will remain an important part of the global economy, even after we stop using fossil fuels for energy and transport. Let’s see how we can share the sun.

Microwave beaming

Though nobody has demonstrated the technology over long distances, beaming microwaves is surprisingly efficient in theory. One could imagine a large antenna in low Earth orbit that redirects microwave energy from sunny places to more northerly latitudes.

That being said, it’s hard to see how this is any better than building a cable. To produce microwaves, you have to convert DC electricity into AC and suffer the same drawbacks as transmission lines. But in this case, the technology is far more speculative.

Another challenge is to keep the beam intensity low enough so that it doesn’t fry birds or airplanes in its path. That requires a larger antenna and raises costs. So it looks like death rays will only be used for beaming energy to the moon. Sigh.

Battery ships

So microwaves won’t work, but what about something more established, like batteries? Imagine a massive cargo ship where each shipping container was a battery. The ship could dock at a sunny port, swap charged batteries for empty ones, and head north to supply industry. On arrival, a self driving car could park one of these batteries inside any factory that needed juice.

Shipping sunlight around the world in huge, self-driving batteries is a vivid image, but I think this idea is probably unworkable¹. Though batteries are getting cheaper, they’re extremely heavy. That brings shipping costs up by a lot. For example, Eli Dourado estimates that international shipping costs $0.01 per ton-km². The energy density of lithium iron phosphate batteries is roughly 200 Wh/kg and the distance from Darwin, Australia to Singapore is about 3000 km. That implies a cost of energy of $150/MWh even without accounting for battery costs, energy costs, or energy losses.

So battery ships are out, is there an energy source that’s significantly lighter?

Hydrogen or methane balloons

Everyone wants to make blimps happen, but the case for using blimps in bulk shipping looks pretty thin³. They carry less than a cargo ship, are slower than airplanes, and are trickier to implement.

But what if the only thing on the blimp was the lifting gas itself? Leveraging atmospheric winds, hundreds of balloons filled with hydrogen or methane could float around the world and land where they’re needed. These lifting gases could be produced directly with solar energy.

There’s a massive engineering challenge here. Each blimp would carry very little fuel, so you’d need to make sure the balloon and navigation system are cheap, otherwise your capital costs will spoil the plan. You’d also need to make sure that a substantial fraction of your blimps actually make it to their destination (and quickly). On top of all this, venting gas for altitude control would lead to significant losses. You can amortize some of these costs by making the airship bigger, but there are limits to how big it can get.

In all, blimps don’t look that promising compared to something like LNG tankers, which lose only ~5% of their fuel to boil-off. We’ll need to find another application for circumnavigating balloon swarms. Sigh.

Liquid chemicals

Now we’re back to the system we use today: putting energy-dense liquids into tanks and shipping them. Oil and liquid natural gas dominate the international energy market because they’re easy to work with, easy to move around, stable, and not too toxic.

With solar energy, we can produce our fuels from recycled CO2 rather than getting them from the ground. Since we’re making these from scratch, what’s the best chemical to ship? For example, some people have suggested methanol as a versatile, safe fuel. Others have proposed ammonia as a carbon-free hydrogen source.

After thinking about exotic fuels and the costs of transport, I’ve decided that a few principles dramatically cut down our options:

1. The fuel should only contain carbon, hydrogen, and oxygen⁴.

2. Easy to synthesize from hydrogen and carbon dioxide.

3. Liquid at ambient condition for ease of movement.

4. Nonreactive and noncorrosive.

5. High energy density.

Only oil satisfies all the conditions, but methane (liquid at lower temperatures) and methanol (slightly corrosive) are also good options.

So which of these fuels is best suited for shipping solar energy? They need good solar-to-fuels efficiency, low shipping costs, easy transport on land, low storage cost, and easy to use.

It turns out that there aren’t big differences here. For instance, all three fuels require hydrogen and carbon monoxide for synthesis in roughly the same ratio. Their solar-to-fuel efficiencies depend on producing those precursors more than producing those particular fuels. All three are easy to turn into syngas, from which you can make everything else.

Their synthesis from CO2 and hydrogen requires similar reaction conditions, so I expect the reactor costs per MWh of fuel produced to be similar. Estimates from this paper imply that they would all provide energy at a around $0.10/kWh. At the destination port, converting them into energy can employ similarly-efficient combined-cycle gas turbines.

The deciding factor is transport costs. Because oil is liquid and noncorrosive it’s cheaper to ship internationally and over land (see appendix). It’s also the cheapest to store long term.

So I’d guess that synthetic oil will become the way we ship solar energy across the world, with natural gas or methanol taking up some market share.

Conclusion

It was fun to speculate about which chemical will reign supreme in a solar-dominated world. Can blimps make a comeback? Will the methanol economy happen?

But remember shipping fuel can’t be cheaper than getting your energy locally, so most places will opt to build solar panels rather than buy energy abroad. E-fuels will be a small market compared to local solar. But imported e-fuels can help stabilize local grids and provide supplemental energy to land-constrained or high-latitude locations.

Rather than provide fuel, places like Australia or the Sahara can use cheap energy to transform matter at scale. They can capture carbon, desalinate water, recycle atoms, and ship platform molecules to the rest of the world. Local economies can transform platform molecules into whatever they need. Opportunity costs mean that energy-poor economies will specialize in making sophisticated products.

As a society, we’ll bottle the sun by rearranging atoms into standardized, useful molecules. Everyone will be able to get these molecules for cheap and turn them into products. This transition will take decades, but will eventually lower the cost of virtually every material, turning sunlight into abundance.

Appendix

Overland transport

Overland transport is a bit different than shipping, since there are more options. In general, the cheapest way to move stuff is via pipelines, then rail, then roads, with air freight being the most expensive⁵. These papers consider the cost of moving different e-fuels via pipelines. The main takeaway: transporting electrons is much more expensive than a pipeline.

Pipeline costs depend on how difficult the chemical is to work with; hydrogen is difficult for a bunch of reasons, natural gas needs to be compressed, and alcohols are mildly corrosive⁶. Oil pipelines win out because oil is liquid at ambient conditions, noncorrosive, and we already have the technology figured out.

But the pipeline cost is only one part of the equation. At one end of the pipeline, energy and chemical reactors are used to make the fuel, and at the other end, reactors convert the fuel into useful products. If the fuel is expensive to make or use, then it would be silly to transport in the first place.

I looked at the process conditions for producing methane, methanol, and oil using CO2 and hydrogen, and they look pretty similar, with methanol requiring higher pressures and oil having more issues with coking. So I expect the capital cost per MWh of fuel to be similar. My guess is that the round trip efficiency differences are probably pretty small because they all require hydrogen and carbon capture and they all can be burned in a combined cycle gas turbine. Even with a round trip efficiency of 10%, they still beat power lines on cost.

So it really comes down to what you need on the other end of the pipe. If you’re providing transportation fuel, just make oil. If you need natural gas for a city, send methane. For chemicals production, methanol is a good option. But overall, oil seems like the safest bet. It’s cheap to transmit and easy to use for energy, heating, transportation, and chemicals production. And on top of that, it’s cheap to store⁷ and ship internationally.

DAC vs. CO2 recycling

Direct air capture is expensive, would it make sense for countries buying e-fuels to ship the CO2 they generate back to sunny places?

I looked into this a bit and came to the conclusion that shipping CO2 over long distances is more expensive than simply capturing it from the air. CO2 is hard to store and carbonates add a lot of mass. CO2 pipelines aren’t established yet, but would probably be much more expensive than their corresponding fuel pipeline.

The only argument I can make for shipping CO2 is the fact that an empty oil tanker could backhaul some useful stuff to its port of origin. They also require some mass on the return as a ballast and have water tanks for balance that could hold carbonate solution instead. So perhaps carrying carbonates back to Australia to be turned into fuel could make sense. But it won’t be enough to supply all the CO2 needed to make the fuel.

1. Though the case for powering cargo ships with batteries looks a lot better.

2. Which is similar to this estimate.

3. Though blimps for air freight looks more promising.

4. Because these are abundant elements with well-understood chemistry. More energetic materials like borohydrides are simply too expensive and dangerous to manipulate. This rule also excludes amines, which are mildly toxic, corrosive, and have slightly lower energy density than hydrocarbons.

5. But it also depends on the situation. If you don’t have the capital to build a new pipeline, putting your product on existing rail capacity can be better.

6. These papers ignored coal. Can coal be transported via pipelines? Yes, coal slurry is coal powder mixed with water. The problem is that the coal powder is abrasive and needs to be dried after delivery. A better option is to make logs of solid carbon and push them through a pipe using water, which requires less water and less drying overall. The impression I get is that coal pipelines are not as cheap as other fuels. For instance, this study suggested that gassifying the coal would be the cheaper than transporting it by rail or by wire.

7. Only coal is cheaper to store, since it can literally sit outside.