'Artificial leaf' concept inspires research into solar-powered fuel production | Hacker News Reader

'Artificial leaf' concept inspires research into solar-powered fuel production

phys.org

60 pointsdnetesn37 comments

What's the efficiency of using actual plants to produce fuel? Either through refining e.g. corn oil directly or using plants to feed bacteria that produce a fuel. That's a carbon neutral way to convert sunlight to fuel.

https://en.wikipedia.org/wiki/Photosynthetic_efficiency:

  Quoted values sunlight-to-biomass efficiency

  Plants, typical     0.1% [3], 0.2–2% [4], 3.5-4.3% [5]
  Typical crop plants 1–2% [3]

Solar cells can turn about 10% of incoming sunlight into electricity (cells are more efficient, but they don’t cover the whole ground, typically aren’t kept optimally oriented, etc)

Plants are much easier to recycle as well.

Biodegradable by definition.

The maintenance is much more expensive and time consuming though. Most solar panels are install and forget for their lifetime measured in decades, with some small failure rate in the meantime. Plants however typically require maintenance multiple times a year and most have usable lifetime measured in months, not decades.

There are more types of plants than just corn. Trees are much easier and cheaper to establish than solar for example.

Trees are the lowest of that low range that start at 0.1% efficiency.

And if you collect only the wood, it's a couple of magnitude lower than it.

But delivering the energy from those trees to consumers is harder and more expensive than for solar cells.

Also, if your desired output is electrical power, you’ll lose more energy converting your trees to electricity.

On the plus side for trees, storing the energy they stored is cheap, and can be done for centuries.

The net effect is that PV cells are more economically viable than using photosynthesis to catch solar energy (at least for now; there’s work being done on improving photosynthesis; I don’t see that beating PV soon, though, certainly not because PV cells keep improving fairly rapidly)

Trees are often turned into heat and energy. Obviously they have very different requirements from PV cells and overall probably aren't as good they are certainly viable in some places (e.g Italy).

Somehow it just doesn’t feel right to burn trees to produce clean or green energy. It might work on a small scale but not sure about doing it on an industrial scale.

It might be renewable but it still makes air pollution. On certain nights when everyone in the neighborhood decides to have a fire the air quality is terrible - almost as bad as during summer forest fire season.

and plants are not only energy storage. They are great water storage, air cleaners, area temperature regulators. Not even counting as habitat for rich fauna fungi...life.

Still waiting to hear a politician running on the platform of more plants.

Plants are enormous consumers of water. The water argument is strongly in favor of PV.

There is a lot more to water than consumption. Overall trees usually have a positive effect on water flows.

Solar cells are more efficient and last longer, but don't grow themselves from seeds produced by the previous generation. That makes it much more expensive to cover large areas with them.

Algae and Cyanobacteria tend to be substantially more efficient than whole plants (6-8% [1]). There are some caveats though. Generally the industry has found it hard to grow algae at commercial scale cost-effectively for several reasons including (but not limited to):

- Contamination by outside organism kills your algal culture

- If you create a closed system, the algae will stick to the glass, fouling the surface and reducing light transmission

- Extraction of materials from the algae is hard, you end up with a lot of water weight you need to get rid of somehow

There are also some counterintuitive aspects in that apparently dropping the efficiency of the antennae (light collection) actually makes the overall culture more productive because excess light makes it past the initial cell layer and therefore other cells can benefit from the light. Otherwise the initial cells grab all the light they can, and waste whatever light energy they don’t need.

I think it’s an interesting field, but so far it’s been a hard road to make it actually work at scale cost effectively.

[1] https://www.sciencedirect.com/topics/chemistry/photosyntheti...

Since lichen are a symbiosis with algae, cyanobacteria, or both, it's possible that we could create lichen-like habitat for both or either and extract the sugars for fuel.

Just a few years ago we realized a pretty profound misunderstanding of how lichen function, which partly explains why we have trouble propagating them in a lab. Perhaps there's enough new information there to warrant some more research.

During the dot-com boom there were people cooking biodiesel from algae but I suspect they couldn't get the embodied cost of the system down. Either materials, or maintenance. I'll bet microbial mats in particular foul the system over months if not weeks.

Edit to add:

Whereas the earlier work was on 'digesting' algae and extracting the lipids, within a symbiotic structure (lichen, coral, root nodules, mycorrhiza) the host is exchanging sugars for minerals. You wouldn't have to 'crack open' anything to get to them, if you can find a way to collect them, which I suspect puts us deep into nanomaterials territory.

I worked on phycology for an algae biofuels startup for the better part of a year.

Just chiming in to say that this is a good post, and I don't even have much to add to it.

I quit once I realized that paving over desert ecosystems with huge raceway ponds filled with glyphosate, wasn't solving any problem I wanted solved.

If we don't switch fuels, there will soon be no shortage of desert ecosystems no matter how many we pave over...

I don't want to speak for anyone else, but the point might have been that there may be other ways to get energy which accomplish the goal without requiring that sacrifice.

Correct, and while I work on developer tools these days, I had a satisfying stint at a battery-analytics company after transitioning to software development.

Raceway ponds full of glyphosate in the desert; sounds like Sapphire Energy?

For what it's worth, I think we'll be able to dispense with the glyphosate if we're more flexible about what algae we're growing, and fuzzily select ecosystems rather than strains. That way we're better able to tolerate invaders & other stresses. But that means we're not operating at peak efficiency, which means a larger footprint.

I'm more concerned about how we source fertilizer.

(For context, I'm between jobs, but my last job was as an algae cultivation technician.)

I spent a couple of years doing USDA/DOE research into oilseed biofuels. The short answer to your question is 'it depends'. Part of our work involved doing a LCA (life cycle assessment) on the oilseeds as a biofuel.

There are several considerations to be had here. Firstly, are you farming on marginal or primary farmland? Marginal land is generally marginal for a reason and will have typically lower yields. How are you farming? What intensity? Where do you have to move the product to for processing? What is the energy mix going into the processing workflow? Do you have to move the fuel again after you process it?

The relatively easy answer to solving for the above issues, is that in some circumstances, it can make sense for some farmers to generate some oilseed for the purposes of making their own biodiesel.

you tend to spend nearly the same energy to produce biofuel as you get. that could improve some but it doesn’t appear to be a real solution, at least not with 9 billion to feed too.

"Rice University researchers have created an efficient, low-cost device that splits water to produce hydrogen fuel."

We've heard that one before.

"The platform developed by the Brown School of Engineering lab of Rice materials scientist Jun Lou integrates catalytic electrodes and perovskite solar cells that, when triggered by sunlight, produce electricity. The current flows to the catalysts that turn water into hydrogen and oxygen, with a sunlight-to-hydrogen efficiency as high as 6.7%."

OK. things like that have been built before. The usual problems:

- Costs too much.

- System only works for a while before filling up with crud from the water. (This is also a big problem with fuel cells.)

- Sunlight to hydrogen to fuel cells to electricity is far less efficient than solar cells.

There are a lot of industrial processes that need hydrogen as an input, fertilizer production as possibly the most important. But yes, hydrogen to electricity is silly and I'm pretty skeptical of its use for vehicles too.

At these efficiencies (6%) it would be better to use PVC panels and use that electricity to make hydrogen.

> Sunlight to hydrogen to fuel cells to electricity is far less efficient than solar cells.

Isn’t this already obvious?

The purpose of the hydrogen is to produce it offline, via solar, wind, tide, nuclear, etc., so that it can be used at the moment you need it.

Just last week I was thinking we need solar cells wired up in a tree structure with batteries integrated into the tree so that we don't need to combine series and parallel or have the drawbacks of either. Is anybody doing that? I presume they would need to be connected with fluids, electrolyte in channels instead of with wires. Or have I just reinvented the tree?

What problem are you trying to solve?

That is always the most important question, isn't it?

I am looking at RV applications, thinking series would be ideal to reduce wiring cost, but just learned that a small patch on one panel of shade lowers current in the whole series, when wired that way. If there's some way they could be wired as a tree, with the high amperage portion of the circuit fairly short, that seems more efficient. But then I guess I have just reinvented the power grid... I'm talking about residential scale, though, or smaller.

So you're trying to connect an array of solar panels in series, such that shading one doesn't limit the current of the others. Isn't that normally solved using bypass diodes?

I guess you just solved my problem. This is a topic I'm just beginning to look into. Thank you.

But now I'm wondering why are panels made this way? I mean internally they're wired in series or at least substantially so. and I've read that shading a small portion of one panel will drastically reduce the output of that one panel. Is it possible to create a panel (e.g. nominally 12 volt system) with bypass diodes inside it? If so then why wouldn't they have been doing this all along?

Sounds similar to a redox flow battery (https://en.m.wikipedia.org/wiki/Flow_battery) which uses tanks of anolyte and catholyte solutions to store power. Energy density for Vanadium based redox flow batteries are only 20Wh/kg but might be worth it given the volume of a typical (normal sized) tree.

Interesting. It’s a hydrogen generator.

As an alternative idea, is it possible to use a laser to break the bond between hydrogen and oxygen, in a water molecule, to release the hydrogen?

Kind of how a laser was used to cool atoms, by knocking off energetic electrons from the molecule. But this time, in reverse.

Asking for a laser physicist.

Interesting suggestion.

Lasers themselves would likely be inefficient. An LED emitting at a speficic frequency tuned to the H-O bond strength ... might ... work. First question is: what is that frequency, if it exists at all?

Reply on news.ycombinator.com