Since the launch of our Carbon Removal Fund, we’ve seen lots of exciting approaches to, well, carbon removal. We’ve been aggressively funding the very best and couldn’t be prouder of the portfolio of badass companies we have assembled so far. From Verdox to Charm to Heirloom to Running Tide, we are working with the teams that carbon fears.
Yet, there are still some carbon removal methods that we think could use some more love. They probably won’t all work, and most of them require a lot more fundamental research to understand potential risks and tradeoffs. But the stakes are so damn high and the opportunities are so big, it’s worth really digging in.
To create a little structure, we split them into two broad buckets. The first includes physical and geochemical approaches to carbon removal, leveraging the slow carbon cycle in Earth’s crust with rocks, acids, and bases. The second bucket is biological and contains living forms of carbon removal, leveraging the speed of photosynthesis and mechanisms such as enzymatic activity to facilitate long-term carbon storage.
These aren’t our only areas of interest, but we think they are big opportunities for great founders to have gigaton-scale impact. If you’re working on any of this stuff, or something else in CDR that’s even more impressive, please hit us up!
(Note: The rest of this post is going to be pretty geeky. But, in fairness, that’s who we are. We will not apologize for the climate nerdery that follows.)
Physical and geochemical pathways: rocks, acids, and bases
The long-term carbon cycle is dominated by the earth’s crust and oceans: rocks dissolve, they mineralize CO2 from the air, and the ocean balances with the atmosphere. Solutions in this category don’t feel like nature, and may seem boring or counterintuitive, but have tremendous potential for atmospheric-scale carbon drawdown. (For a wider overview on geochemical CDR pathways, here’s a new comprehensive review paper from The Climate Map.)
Zero-emission metal oxide production
Ready for this? Less than 1% of the carbon atoms on earth are in plants, animals, bacteria, and all living things. 99+% of earth’s carbon is in rocks in the crust (e.g. limestone) and rocks dissolved in the ocean (e.g. bicarbonate in solution) as part of the slow carbon cycle. These stores are an ideal, durable, and thermodynamically favorable sink for atmospheric CO2. To capture and store carbon this way (generally referred to as mineralization), you need a Ca or Mg cation to bind the CO2 in the presence of water, forming a calcium or magnesium carbonate.
This is a double-edged sword, though, because the earth system is wildly good at carbonating these cations naturally over geologic time: you’ll find huge amounts of limestone (calcium carbonate) but almost no lime (calcium oxide) just hanging out on the earth’s surface. Anyway, we think that magnesium and calcium (hydr)oxides with no upstream emissions are way more valuable than people seem to realize as a commodity material for carbon removal, and anything that can produce them sustainably at large quantities cheaply is really interesting. All Ca and Mg originally come from the crust or ocean, so these would probably look somewhat like mining or brine processing companies. There’s a huge variety of techniques and processes that could work too: from more traditional mining operations to hydrometallurgy to new electrochemical approaches.
While hard rock silicates (basalt has a lot of Magnesium!) are an obvious choice, keep in mind that subsurface aquifer brines contain large amounts of Mg (though often in chloride or sulfide, not oxide) form, as do many forms of mine tailings.
Carbonation and monitoring of zero-carbon alkalinity
Those metal oxides described above are only useful if we have simple, cheap, and easy to monitor and verify ways to carbonate it (especially at atmospheric CO2 concentrations). Any of these solutions will require robust monitoring, reporting and verification (MRV) to both ensure the carbonation as well as the permanence (if your carbonate is exposed to acid, the CO2 could release right back into the air!). These monitoring and baselining methods will look different across different approaches and chemistries, and their costs will matter.
On land: There are already promising startups carbonating metal oxides for carbon removal. However, these approaches tend to rely on a looping process where the metal oxide functions as a CO2 sorbent material and then is recycled. However, in a world of abundant zero-carbon metal oxides, you could avoid the high-energy cycling process altogether and just treat the metal oxide as a once-through material flow, trading off mass for energy.
In other words, the energy associated with extracting metal oxides to carbonate with CO2 needs to be lower than the energy required to loop the metal oxide material plus transport and store CO2. In this frame, instead of focusing on the speed of carbonation and energetics of looping, you’d try to solve the mass transfer problem with much more flexibility on carbonation rate. On some land, creating a giant pile of carbonate might be tenable! But we’re curious what a scalable and reasonable approach to single-pass carbonation would look like. Perhaps something like injecting the carbonated material back into the subsurface in a slurry?
In the ocean: The oceanic pH buffer has enough capacity to store the entire atmosphere’s worth of CO2 with careful alkalinity addition, but doing so has yet to be attempted at meaningful scale.
In a world of abundant zero-C metal oxides, we imagine the metal oxide production company may need a partner to safely deploy and monitor their alkalinity dispersal into the ocean for carbon removal. This job would entail permitting and regulatory management of outfalls, ocean alkalinity sensing, monitoring, material dispersal, and all of the associated MRV to create a high-quality carbon removal credit.
Mass-transfer-driven ocean alkalinity enhancement
Alongside the point above, there may be potential to do ocean alkalinity enhancement with “non-reactive” alkalinity – eg by adding material that isn’t a metal oxide and therefore not directly carbonating, but still serves as an ocean pH buffering agent enabling more atmospheric carbon to flux into the sea and be safely stored as bicarbonate. Ideas in this area would take advantage of existing ocean cycles and imbalances of alkalinity, but the full space of them definitely hasn’t been mapped yet. For inspiration, you can take a look at these two papers:
- Mitigating the atmospheric CO2 increase and ocean acidification by adding limestone powder to upwelling regions
- CO2 capture by pumping surface acidity to the deep ocean
Durable storage of a dilute CO2 stream
Basic thermodynamics tells us that stepping up CO2 concentration from atmospheric (0.04%) to something more like 10-20% is much easier than stepping up from atmospheric to 95%, as is typically required for geologic injection. So that begs the question – can we find a way to store a 10-20% concentration CO2 stream permanently that’s cheaper than bumping the concentration up much higher? There’s plenty of room in geologic pore space, but the lower the purity, the higher the compression energy requirements or the longer it takes to inject an equivalent volume of higher purity CO2 stream. So what’s the cutoff on energy saved by keeping to lower concentrations of CO2 but compromising on compression energy and injectivity? Are there other interesting options here? It could dramatically decrease the all-in cost of DAC if someone were to figure this out!
Low-volume injection sites
Like most engineered systems, these are spec’d based on economics and precedent. Injection wells and boreholes are typically made of a certain size and when below a certain depth must have a minimum diameter size. And for CO2 injection these have typically been spec’d for an injection rate of CO2 of ~ 1MtCO2/year at minimum. However, this value is not a physical minimum for injection into such a wellhead, but rather has been set by the economics of the site which have historically dictated that when amortized CO2 storage on land should cost 10$/tCO2 vs. offshore ~ 30-40$/tCO2.
Flipping the economics of storage vs. capture
In the context of carbon capture/direct air capture and storage, the storage piece has very often been an afterthought that should cost within a certain bound. However, the two prompts above suggest a paradigm shift in the way we should think of CO2 storage economics. Imagine if, instead of capture as the main focus, storage were instead the driver of cost. This could be the case with a dilute stream of CO2 (cheaper to separate) or higher cost of storage because of injecting low volumes. Then, how would resources/effort be directed?
Low-energy approaches to rearrange CO2 into other stable products
Transforming CO2 gas into stable, easy-to-store, non-gaseous, products would be helpful in the quest to store CO2 permanently. Examples could include decomposing CO2 into solid C and O2 gas, or allowing it to react with a readily available and inexpensive coreactant like water to release liquid or solid hydrocarbons that could be sequestered. For most transformations of CO2, an energy penalty will be incurred, so processes that achieve near the thermodynamic minimum energy use are especially of interest. Researchers developing thermodatalytic, electrocatalytic, and plasma-based approaches are all making strides in this area. We’d be excited to see more attempts in this space.
Biological pathways: living things for carbon removal
Growing plants is an extremely fast mechanism for taking carbon out of the air, but storing it durably gets tricky. Not impossible, but tricky. It’s very difficult to make biological forms of carbon durable. However, there are definitely ways to make the C captured by plants stay out of the atmosphere almost indefinitely (gestures wildly at the entirety of fossil fuel deposits, gestures wildly at mummies, gestures wildly at perfectly preserved shipwrecks in the Black Sea). So, here are some ways we can essentially speed up mummification, and more.
Cheap, stable, and easy-to-monitor preservation of existing biomass
Part of the reason storing biomass for hundreds to thousands of years is tough is the absolutely massive parameter space of decomposition (to CO2 and, even worse, to CH4). For a given feedstock of biomass, there’s all sorts of things you could do to attempt to preserve it: make it really dry, really wet, really cold, really hot, really anoxic/anaerobic, really pressurized, really alkaline, really acidic, really salty, really formaldehyde-y, and/or any combination of these things and beyond.
This one is tricky, though, because you need to be extremely sure that whatever preservation you attempt won’t accidentally lead to CH4 production – under reasonable GWP assumptions, you only need to leak a few percentage points of methane to end up in a worse situation than if you had just put all the biomass in a pile and lit it on fire.
Additionally, we’ll need to be careful that we don’t accidentally preserve too many important nutrients out of the carbon cycle along with the C that we’re aiming for. One of the downsides of biology is that, almost always, C uptake comes in a ratio with limiting nutrients, like iron and phosphorous. For example, there’s much less phosphorous in tree trunks (lignin) than leafy (cellulosic) biomass, but if we effectively remove lignin from cycling, that may still not be enough phosphorus buffer in many ecosystems (monitoring this will be hard), and we’ll run into nutrient limitations much more quickly with things like waste leaf matter or algae.
So, yeah… please be careful here… but hit us up if you have any bright ideas.
Growing biomass that doesn’t decompose
Adjacent to the idea above, we want to take advantage of photosynthesis’s ability to rapidly draw down carbon but we also want that carbon to behave as similarly to carbon in the slow cycle as possible. That could mean growing new forms of engineered biomass that decompose so slowly as to be functionally stable.
Some naturally produced biopolymers such as suberin, lignin, and sporopollenin seem to exhibit this property, and there are others yet unknown. We could also explore mechanisms to grow recalcitrant biomass based on other principles, such as modified biostructures that don’t decompose because they’re brand new and unfamiliar to all organisms that could possibly decompose them. For example, there was a period in the late Carboniferous where dramatic carbon sequestration may have resulted in the evolution of lignin an epoch ahead of fungus or other decomposers able to process it. Could we recreate these conditions today? Novel chiral lignin, anyone?
Biologically accelerated silicate weathering
Lichens, mosses, and fungi already contribute to natural rock weathering cycles through both chemical and physical means, so it’s not totally out there to assume that we could achieve orders-of-magnitude speedups in the natural rate of silicate weathering using biology. We could manipulate mineralization kinetics (and therefore C uptake) with things like chelating agents, oxalates, silicases and other enzymes, or acidifying organisms. We could also try to speed up weathering physically by increasing the rate at which surface area is exposed to the air (meaning: make big rocks tiny). Crushing, grinding, and ultrasound all work, but what if we used a bioengineered moss, lichen, or lithotroph that can degrade the rocks more quickly, or with less energy required? (Bonus points if the organism also creates a more acidic environment around it, speeding up the process further).
Biology-inspired CO2 concentration and mineralization
CO2 transport, concentration, and mineralization mechanisms exist throughout living systems from microbes to humans to plants (and even ants!). And we’re getting ever-better at optimizing enzymes through computational biology… so what if we combine biological and abiological technologies in interesting ways to make novel CO2 capture and concentration systems? As an example, it seems feasible that we could engineer enzyme-enhanced membranes for carbon filtration and concentration. For inspiration you can look to components like carbonic anhydrase in your lungs (which facilitates CO2 exchange) or hemoglobin in your blood (which selectively transports 23% of CO2 in the body as carbaminohemoglobin). There’s also a lot to leverage in biomineralization – we’d love to see more clever concepts using enzymes or organisms that could drive direct mineralization in engineered systems.
Don’t be strangers
Alright, you fired up? Our team is. While I’m thrilled to be focused on this work, the ideas above include contributions from across our squad, in particular our partners Dr. Clea Kolster on the chemical engineering and subsurface, Kristin Ellis on biology, and Dr. Christina Chang on physics. Our team literally lives and breathes carbon (see what I did there?) and we have an extensive network of experts that continually challenge and advance our thinking.
So please – get in touch. Bonus points if you’re working on something that immediately makes us feel silly for not including it here. There’s never been a better time to build a carbon removal company!