Hacker News new | past | comments | ask | show | jobs | submit login
Solar foods (solarfoods.fi)
137 points by EastLondonCoder on Nov 1, 2018 | hide | past | web | favorite | 53 comments

The website is a bit sparse, I wonder where they get elemental hydrogen and nitrogen from. Does anyone know anything more about their technology, e.g. how they actually construct the proteins (I'm guessing either in cells, or using some enzymes outside of cells - but then what produces the enzymes?).

Edit: Ah, an article posted by another comment includes more information:

> This sounds like science fiction, but it is already approaching commercialisation. For the past year, a group of Finnish researchers has been producing food without either animals or plants. Their only ingredients are hydrogen-oxidising bacteria, electricity from solar panels, a small amount of water, carbon dioxide drawn from the air, nitrogen and trace quantities of minerals such as calcium, sodium, potassium and zinc. The food they have produced is 50% to 60% protein; the rest is carbohydrate and fat. They have started a company (Solar Foods) that seeks to open its first factory in 2021. This week it was selected as an incubation project by the European Space Agency.

> They use electricity from solar panels to electrolyse water, producing hydrogen, which feeds bacteria that turn it back into water. Unlike other forms of microbial protein (such as Quorn), it requires no carbohydrate feedstock – in other words, no plants.


Edit2: I wonder how this (water + CO2 + solar power) compares with just growing algae in water in the sun, in terms of energy efficiency and process stability.

Here's some information about it: https://www.lut.fi/web/en/news/-/asset_publisher/lGh4SAywhcP...

They say that compared to photosynthesis, it's 10x more energy efficient.

Interesting, thanks! I wonder if that's algae photosynthesys, or plant photosynthesis - arguable the latter is "less efficient" because we eat only parts of the plant.

In any case, I'm surprised that electrolysis is so efficient. I remember reading that it's so inefficient that the primary source of hydrogen is currently actually natural gas! And I wonder how sustainable is the process - doesn't electrolysis deplete the anode and catode, which need to be made of expensive materials for reasonable efficiency?

> arguable the latter is "less efficient" because we eat only parts of the plant

Solar cells are more efficient than photosynthesis.

“For...sunlight, where only 45% of the light is in the photosynthetically active wavelength range, the theoretical maximum efficiency of solar energy conversion is approximately 11%. In actuality, however, plants do not absorb all incoming sunlight (due to reflection, respiration requirements of photosynthesis and the need for optimal solar radiation levels) and do not convert all harvested energy into biomass, which results in an overall photosynthetic efficiency of 3 to 6% of total solar radiation“ [1]. Meanwhile, “solar cell efficiencies vary from 6% for amorphous silicon-based solar cells to 44.0% with multiple-junction production cells and 44.4% with multiple dies assembled into a hybrid package” [2].

[1] https://en.m.wikipedia.org/wiki/Photosynthetic_efficiency

[2] https://en.m.wikipedia.org/wiki/Solar_cell_efficiency

To make the comparision meaningful, you need to compare the efficiency of the whole cycle (sunlight -> solar power -> electricity -> electrolysis -> hydrogen -> bacteria -> biomass) to photosynthesis (sunlight -> plant -> biomass).

> you need to compare the efficiency of the whole cycle

Top line, plants are taking 11% while we’re taking—on average—double that. We’re trying to turn the energy into stored calories. Plants are trying to plant. First-order estimates rarely present a more-compelling case.

Cool, interesting and fascinating. I can't wait until this is more wide-spread, both for terrestrial and space solutions. You seem to work in this industry, or at least very knowledgeable about it - any other companies you recommend I follow?

I'd say for total transparency, you need to go even further back and calculate the energy and materials cost for making the solar panel. Plants reproduce themselves, and solar panels do not.

I think the problem with Algae is that you won't get as high a yield of proteins, because the plant cell is producing other chemicals as well.

Compared with animal cells that are producing a protein specifically and can be engineered to produce more.

>I think the problem with Algae is that you won't get as high a yield of proteins, because the plant cell is producing other chemicals as well.

And algal-derived protein and fats seem to cause digestive issues with some individuals. Rosa hit some issues with this, apparently, with the soylent food bars (man I miss those things) to the point where they discontinued the bars and have yet to bring them back 2 years later.

For those unfamiliar the bars were causing some individuals to vomit and have diarrhea. It was believed, by the community (as far as I know Rosa never made a statement), that the algal flower was the culprit.

Rosa also discontinued their use of algal oil from TerraVia in their other lines shortly after, some suggested it was TerraVia actually cutting them off in retaliation for allowing people to suggest the flower was causing people to become ill but I can't imagine TerraVia would choose to 'fire' a rather large (and growing) customer. Rosa switched to a sunflower oil.

If I were this company I would immediately start testing their 'food' on a wide variety of people because I have a hunch they will find that a notable percentage of the population is going to have digestive issues.

Do you mean flour?

Haha, yeah. I didn't get much sleep last night, I have upstairs neighbors that apparently hate their heels and joints, or are Jötunn.

> This sounds like science fiction, but it is already approaching commercialisation.

Soviet union were feeding kindergarten kids with yeast porridge long before Soylent was hip...


the link doesn't state the kindergarten claim as far as I can tell.

Is this still done in both the West and Russia?

At least in mine, they did in 1993 or so. That porridge though... was known to give serious flatulence

This is there whitepaper:


The website prompts you to sign up to the mailing list to get a copy. Its very sparse on details unfortunately.

Here are more detailed papers that I found by way of the Guardian article about this company:

"Carbon emission avoidance and capture by producing in-reactor microbial biomass based food, feed and slow release fertilizer: Potentials and limitations"


"Microbial protein: future sustainable food supply route with low environmental footprint"


"Autotrophic nitrogen assimilation and carbon capture for microbial protein production by a novel enrichment of hydrogen-oxidizing bacteria"


I want this to be real, but this rings all the bells for yet another EU funded vaporware project, like the Semantic Web.

I hope I'm completely wrong.

Maybe this can be developed on a massive scale for CO2 capture. If people are hesistant to consume this protein, it could even be used to feed fish / animals.

Edit - could be a possible candidate for ycombinator nature fund.

Since this process is optimized for food production, I'd expect processes optimized for carbon capture to be more efficient at that.

I suspect the main environmental gains from this would be to offset beef production. Outside of reducing the ongoing emissions there, it opens the door to return pasture land to forest, tying up large amounts of carbon in trees.

Userful sidenote: The most efficient poultry producers today have about the same environmental footprint as tofu producers. If you don't feel called to eat vegetarian, simply switching beef for chicken can be almost as impactful. Just stay away from organic food, or all those gains are offset by the inefficiency of unscientific agriculture.

See the second chart here for the chicken/tofu showdown: https://www.theguardian.com/environment/2018/may/31/avoiding...

Unfortunately, if you care about animal suffering then Beef starts looking much better than Chicken does and the farms that tend to treat their chickens well also tend to be "organic". So there's certainly a place for this.

And in general most land used for cattle grazing doesn't receive enough rainfall to support forests; or farming without outside water inputs. If we were to eat cattle that were exclusively fed by grazing on semi-arid land then Beef would be pretty environmentally friendly apart from the methane emissions. The problem is that we don't have enough grazing land to do that so most beef if fed with farm-produced plant matter leading to huge inefficiencies.

>Maybe this can be developed on a massive scale for CO2 capture.

You'd still have to do something with it, either 'bury' it in the anoxic depths of the ocean and large bodies of fresh water, or bury it underground in either large pits or in caves/oil wells.

And you'd have to strip mine insane amounts of the necessary minerals that the algae needs, with regularity, to even approach sequestering a single gigaton a year, let alone the 40+ necessary to make a difference.

I'm all for figuring out carbon sequestration but I think YC's proposed paths for handling this in their recent call for startups looking into this are a bit silly. The flooding a significant portion of the Sahara for example will have extremely unpredictable weather implications which could be absolutel disastrous never mind you'd have to be mining minerals for use in the 'ponds' that is necessary for the life to grow in them which means you'd effectively need to fertilize a quarter plus of a continent with regularity every year for decades or even centuries. Even if you could fund that, it's not sustainable.

The real solution to carbon is immediate reduction of fossil fuel usage with a very short-term goal of nearly absolute cessation of greenhouse gas generation.

You then need to reforest 300 million acres of North America, restore countless miles of coastal kelp and seaweed, etc etc.

Even trying to replicate something like the Azolla event is unrealistic. If you take the 10 largest bodies of freshwater in the world and cover them entirely in Azolla, you are going to capture less than 10% of the carbon we are currently releasing each year and you won't even sequester all of that as to sequester it you have to sink the Azolla to an anoxic depth with every single generation of growth.

So, I doubt this algae would result in sequestering even a tenth of a percent of current carbon generation.

Isn't it going to be hard to use any biological process to sequester carbon for long periods? The cycling is fairly optimal for life (without smokestacks). I would think the attempts to pull it into more stable forms[1] have more potential (but you can't eat them).

1: https://www.technologyreview.com/s/540706/researcher-demonst...

>Isn't it going to be hard to use any biological process to sequester carbon for long periods?

Based on my research, in general, yes. However if you can store biological matter like algae or the Azolla fern in an anoxic environment, you significantly increase how long you can store it.

If you go read about the Azolla event, this is what is believed to have happened roughly 49 million years ago. Azolla is believed to have been growing in the warm waters of the arctic ocean (either by having freshwater being fed to the area and floating atop the salt water, or by diluting it enough to allow the Azolla to survive, I believe the Amazon does this where it empties into the Atlantic now) and as the generations would die off, they'd sink and reach depths that were anoxic and effectively prevented the bulk of the material to decay.

One avenue I found that shows promise for doing this via another method, would be making biochar in massive quantities, Pacific Pyrolysis is a company that interests me here https://pacificpyrolysis.com/about.html. Some of this you would sell as fertilizer, tilling it into the soil, which would only lock it up relatively short term. Some of it you would store in depleted mines that you can effectively cap, basically doing reverse-coal mining.

Other options exist but nothing that will scale the way we need it, an operation in Iceland shows us that you can at least have a negative-emissions power plant by injecting CO2 into basalt where it converts to carbonates, becoming rock.

Dr. Klaus Lackner at the Center for Negative Carbon Emissions also has some promising research with a polymer. You basically make this 'plastic tree' which garbs CO2 out of the air, it gets trapped in the polymer and you then 'wash' it out with water and capture it in the washing process, of course you still have to do something with that captured gas. Here you could sell some for industrial applications and take the rest and pump it into depleted natural gas and oil operations and then cap it in a sort of reverse mining-process again.

I am far from an expert but I've spent more than 100 hours looking into options this summer and fall and I do not see a currently available solution, that even after several generations of refinement, are workable.

I think we need to get lucky and figure out fusion, and fast, I also think we need a group like Oklo Inc to get their reactor up (it should operate on waste from current fission reactors) which would scale more rapidly. These only really solve coal and natural gas though, there's over a billion passenger vehicles on the planet right now that mostly use petroleum.

I know YC is pretty damn optimistic, I believe Altman said something a few weeks ago at the wired event along the lines of 'we will figure out fusion because we have to' but man, I'm considerably more pessimistic. I think the next 10-50 years are going to be interesting and quite tense.

> The real solution to carbon is immediate reduction of fossil fuel usage with a very short-term goal of nearly absolute cessation of greenhouse gas generation.

The problem is that while this is accurate given the current state of science/technology, it's a complete political non-starter. There's always incentive for someone to be a bad actor, moreso if everyone else is behaving themselves. Unfortunately, very few nations are going to be willing to commit to massive greenhouse gas reduction while any other nation is profiting from it.

I think the comment from Altman (we will figure out fusion because we have to) has the subtext: 'or we're all dead.' It seems unlikely that killing fossil fuels and other greenhouse gas polluters will get killed by political fiat. That leaves killing them economically, which means creating a more efficient energy source.

> it's a complete political non-starter.

Oh, I unfortunately agree. We can't even get everyone to acknowledge global warming or even to stop using CFCs ( https://www.newsweek.com/mystery-source-ozone-depleting-subs... ) so yeah it's highly unrealistic to get people to abandon fossil fuels even if a fleet of alien ships showed up in the sky this afternoon and said "Dammit! Stop it! We will give you enough fusion reactors and have them set up by the end of the week!"

I think we need to see the overall efficiency first.

As I understand it this process requires that you generate electricity then use that electricity to produce proteins.

I suspect that good old plants may be more efficient.

This technology may be more useful in space and/or in circumstances where lack of growing space is the overwhelming factor.

Also, since the production is food the overall cycle is carbon neutral.

Interesting idea - biogeneration of "food" for space travel first, expanding to terrestrial once the price is down. But desperately short on detail.

What protein do they produce? Not all proteins are equally healthy as far as I know.

Proteins are broken down into constituent amino acids during digestion anyway, so as long as it has a similar (complete) amino profile as, say, albumin or chicken, it should in theory be fine.

That's the question - whether or not it has a good amino profile (and reasonable low insulinemic index).

Also gluten is a composite of proteins too but it's actually harmful not only to people with celiac disease as it triggers production of zonulin.

Can vegetarians on Earth benefit from this sustainable protein as well?

It would be great to have options beyond soy, vital wheat gluten, and the usual bean/grain/yeast medleys!

The single-cell protein sounds a lot like Mycoprotein (brand name: Quorn) which is normally grown/fed with glucose syrup as well as CO2. Sounds like they've come up with a novel variation or process that doesn't need the glucose?


I believe that fungi are considered closer to animals than to plants.

I'll let vegans ponder that one...

I'll leave that here:


"This congruence among multiple lines of evidence strongly suggests, in contrast to traditional and current classification, that animals and fungi are sister groups while plants constitute an independent evolutionary lineage."

Regardless of their evolutionary lineage, the important ethical consideration is whether one's food has a central nervous system. Most animals have those whereas fungi and plants do not.

It seems to me that the "ethical considerations" are highly personal and vary from people to people.

In any case, I note that tongue-in-cheek comments are not appreciated. Sad.

You can't just retreat to ethics are personal when you just implied that a group of people were irrational for being willing to eat fungus. If you didn't intend to imply that then then downvotes show you should work on your communication skills.

Or perhaps you should work on your interpersonal and maturity skills... I didn't imply anything, you lost it over nothing trying to see things where there was obviously none. I was just tongue-in-cheek by putting some scientific facts out there, which may in fact trigger a real reflection of what not eating any animal product might mean to someone (and your point was just a generalisation that does not represent all views).

It's like Reddit here. Too bad. It was a good resource.

The poster that you'r replying to didn't "[lose] it". They replied to the contents of your comments and addressed your points.

"Tongue-in-cheek" comments are more popular on reddit.

There's a suggestion in a piece on this in yesterday's Guardian they hope it can become a bulk filler in foods. Still light on details as it goes off into "what ifs".

Unlike Quorn and other microbials it doesn't need a carbohydrate input so should be able to reduce agriculture needs markedly if it works as hoped.


Normal mushrooms quite protein rich as well, 2.5 g protein for every 4.3 g of carbs: https://ndb.nal.usda.gov/ndb/foods/show/301907?format=Full

(Not to mention nuts, peas, lentils, etc)

They also tend to have all the protein varieties you need in your diet.

Speculating that they plan to use hydrogen-reducing bacteria: This will require substantial post-processing since bacteria have too much RNA and DNA to safely be a major dietary component for humans.

Since the website mentions a Mars play, it's relevant that this drawback has been known in the space habitation literature since at least the 1970s.

What is the dietary danger of excessive nucleic acids? What happens if you eat too much of them? Could it be alleviated by cooking the food before eating it?

The purine bases are metabolized to uric acid which, in excess, can cause gout and kidney stones. Cooking does not modify them in a relevant way.

I think this is kind of similar to what some friends have done with Ecoduna.com. The challenge with Solar Foods must be to produce enough food from a spaceship sized unit.

Not having a nearby sun wild be a challenge too, friending on the mission. Presumably they're looking to Mars first though.

"It's a single-celled protein combined with synthetic aminos, vitamins, and minerals. Everything the body needs."

Finally, something we can eat when the robots take over

What's the areal efficiency of this process? The primary input to the growth reactor is hydrogen gas. The primary input to producing renewable hydrogen is electricity. A solar farm in a good location, like Desert Sunlight, achieves an annualized power density of about 10 megawatts per km^2 (0.1 megawatts (100 kW) per hectare).

It takes about 50 kWh of electricity to produce a kilogram of hydrogen via electrolysis:


According to this paper by some of the people behind Solar Foods, "Carbon emission avoidance and capture by producing in-reactor microbial biomass based food, feed and slow release fertilizer: Potentials and limitations" ( https://www.researchgate.net/publication/326571432_Carbon_em... ), it takes about 560 kg of hydrogen to produce 1000 kg of dry microbial based biomass with a protein content of 70% [1].

At 100 kW/ha, a solar farm in a good location can produce

(100 / 50) * 24 * 365 = 17,520 kg of H2 per hectare, per year.

That in turn can produce

17520 / 0.56 = 31,286 kg of dry microbial biomass per year, containing

31286 * 0.7 = 21,900 kg of protein per hectare per year.

According to some (admittedly quite dated, circa-1972) data collected on this page, the crop with the best areal productivity of edible protein is soybeans at 400 kg/hectare/year.


According to table 2 in the article, the essential amino acid profile of this bacterial protein is equal or superior to soy in all respects. The areal protein productivity of a solar farm coupled to microbial reactors may be more than 50 times that of growing conventional crops. If the bacterial protein is used as animal feed for animals with a good feed conversion ratio (e.g. farmed salmon), it even looks like you could get more animal protein per hectare this way than a vegetarian diet can achieve with conventional farming. And the water requirements are reduced even more drastically than the area requirements. And the electricity production can take place on non-arable land. I would be interested to see more modern areal productivity figures for soy; presumably there has been some additional intensification since the early 1970s, though not 50x improvement.

[1] Hydrogen consumption is not stated directly, but they say that hydrogen at $3/kg makes up 60% of the $2800 cost to produce a dry tonne of bacterial biomass. (0.6 * 2800) / 3 = 560 kg of hydrogen.

and hydrogen feed rate and hence production can be temporarily throttled on the time scales of the microbe feeding patterns so that it can absorb fluctuations in grid energy consumption!

i'm pretty sure solar protein is called plants.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact