Atomic Problems and Gigaton-Scale Thinking

Thinking at a meta-level about the carbon removal ecosystem

Thanks to Tito Jankowski for a long and fun conversation that led to this blog post.

Getting to gigaton scale carbon removal requires gigaton scale thinking. Carbon removal supply chains and business models will clearly look different if there are hundreds or thousands of projects removing carbon from the atmosphere in different ways, moving as much material as the entire global oil and gas industry, while also centering co-benefits and environmental justice. Reaching that point requires thinking backwards at the full carbon removal ecosystem level and finding ways to remove friction and increase collaboration.

Atomic problems are the specific challenging steps that across the carbon removal supply chain that will need to be executed hundreds if not thousands of times as carbon removal gets to gigaton scale, repeatedly by lots of startups or projects trying to remove carbon from the air. They may be hard problems without clear answers right now, and they may have been done successfully few times if any right now in January 2022, but the fact is that they need to be figured out and repeated at scale because of how essential they are to the basic operations of a large number of CDR companies.

For example, in a world at gigaton-scale with hundreds of companies scaling up biomass-based carbon removal processes, all of them will need to find a sustainable source of biomass feedstock for their process, perhaps at the scale of millions of tons. This may be a substantial challenge, but it’s not unique to one company or one process. Every company looking to do carbon removal via bio-oil sequestration, biochar production, BECCS, or biomass harvesting and storage needs to get good at sourcing lots and lots of waste biomass. 

Atomic problems exist for carbon removal startups both on the supply side and the demand side of their operations. By supply-side I mean everything that goes into the full lifecycle of removing each ton of CO2 from the air, from procurement of raw materials to the monitoring and verification of storage, and by demand-side I mean everything that goes into paying for it.

Examples of supply-side atomic problems include:

• Procuring a sustainable source of waste biomass feedstock for a biomass carbon removal process

• Procuring mine tailings for a carbon mineralization process, and implementing proper protocols for handling dangerous materials like asbestos

• Making contracts with energy companies to supply electricity or process heat for a carbon removal process

• Figuring out how to transport the CO2 captured from a direct air capture process

• Getting access to appropriately permitted sequestration wells 

• Getting access to a research boat for ocean-based carbon removal research

• Developing a techno-economic assessment for a carbon removal process in its early stages

Examples of demand-side atomic problems include:

• How to apply to Stripe’s procurement program

• Negotiating long-term offtake contracts with buyers

• How to get listed on carbon removal marketplaces or carbon offset registries like Verra or ACR

• How to work with Verra and other registries to develop a methodology that fits your process

• How to get funded by government CDR procurement programs like OpenAir’s CDRLA

I think of these more as complex operational challenges rather than problems requiring novel scientific research, although research may be needed to find the common solution that works for everybody at scale.

Solving Atomic Problems While Carbon Removal is Nascent

Getting to gigaton-scale carbon removal quickly requires reducing friction across all parts of the value chain. This means massively scaling solutions to atomic problems. At first, with carbon removal as a nascent industry, new companies are either starting from scratch on these problems or relying on the work and experience of related industries, like biofuels, mining, or oil and gas.  But over time, if every new company or carbon removal project needs to figure out these steps afresh for themselves and reinvent the wheel to find a solution, scaling carbon removal will be much slower than if we had shared solutions. 

Each carbon removal process may have vastly different details in implementation, and there may be plenty of steps unique to each company, but to create collective solutions for atomic problems is to solve for the large pieces at the intersection of each company’s operations. 

Tito Jankowski has seen that in the AirMiners community, people are sometimes hesitant to ask for help publicly when they run into atomic problems, either because they’re operating in stealth or because they still feel like beginners and don’t want to ask dumb questions. If the true goal is gigaton-scale carbon removal, we need to make talking about atomic problems publicly extremely common. We need up-to-date lists of best practices and new ideas, strategies being used by current companies to tackle atomic problems, stories about lessons learned, and contacts that may be able to help. To provide as much grease for the common friction points as possible.

Some of this is coming into existence already. Often this type of ecosystem-level thinking occurs at nonprofits or academic institutions that think about the space as a whole more so than any particular project or process. Two examples I can think of are:

• On the supply side, Grant Faber put together detailed templates and structures at the Global CO2 Initiative at the University of Michigan to help very early-stage carbon removal companies put together techno-economic assessments and life-cycle analyses. 

• On the demand side, CarbonPlan (a non-profit) created the standard application that Stripe has companies fill out for its procurement process. The application is intended to be open source and usable by anyone looking to create a process for buying high quality carbon removals, hence greasing the wheels for CDR demand.

I’m not sure what the right format is for creating this type of industry-wide knowledge sharing for atomic problems. Time-intensively, someone working at the CDR-ecosystem level could interview as many startups that were sourcing biomass feedstock as possible and turn their collective advice and stories into guides, and do this at regular intervals to keep it updated. Group brainstorming sessions between industry players, old and new, could help push the ball forward on collective blockers. 

Atomic Problems Becoming Startups

Eventually, as the carbon removal industry grows, you can imagine a world in which many of these atomic problems get solved by standalone startups that spin off to get really good at just that one part of the carbon removal supply chain.

Continuing the biomass feedstock example, you could imagine a startup that got really good specifically at collecting large amounts of waste biomass from lots of different sources and transporting them in a zero or very low emissions manner either to regional hubs or to biomass carbon removal projects themselves. The CDR startups could then spend more time focused on getting really good what they do with that feedstock, having this atomic problem solved for them. And the biomass collection startup could optimize the hell out of their operation of that one step, making it more efficient and driving down costs with bulk transactions and economies of scale. 

At gigaton scale, if there’s a transition to also using purpose-grown biomass for CDR, the feedstock collection startup could spend a lot of time thinking about how to do this in a way that minimizes its land footprint, maximizes the value it gets from marginal land, and preserves biodiversity and centers environmental justice concerns. A company that operates across all parts of the carbon removal value chain will have a harder time giving very dedicated focus like this to each step. It’s already very hard to scale-up a specific step in a process to the scale of billions of tons. It becomes way harder to do so across an entire complex supply chain all at once. Tito Jankowski has a great video about this called Your carbon removal startup doesn’t need to do everything. 

As another example, you could imagine a company that put all its time into figuring out how to quickly permit and buy up access to as many Class VI sequestration wells as they could, and then offered CO2 sequestration-as-a-service: give us your CO2 and we’ll get it in the ground and make sure it stays there. And lots of DAC, BECCS, and mineralization companies producing streams of CO2 could have this atomic problem solved for them by this company.

Companies solving one atomic problem in carbon removal are springing up on the demand side. Patch exists to connect CDR supply to buyers via an API.  But for the most part these companies solving just one atomic problem in carbon removal don’t exist yet. I don’t know of many examples of organizations like this that exist yet for durable carbon removal, but in the soil carbon sequestration world, companies measuring the amount of carbon sequestered in soil like Yard Stick come to mind.

Maybe the supply-side for durable carbon removal isn’t built up enough yet. A company that exists to monitor and verify seaweed sequestration in the deep ocean doesn’t make sense unless there are growing companies getting funded to sink seaweed that can afford to pay the verifiers. A company that buys up all the sequestration wells only makes sense if there’s a lot of CO2 coming their way to realize the advantages of doing their sequestration-as-a-service process at scale, and that presupposes the existence of a large carbon removal market.

Corporate carbon removal purchases like Stripe’s program, often with the explicit goal of helping nascent carbon removal tech get to scale, are so far only making contracts and signing up as customers for projects that actually do the carbon removal step, but I think it would be helpful for them to explicitly fund companies working on a narrower part of the value chain too, even if they themselves aren’t doing the removal. 

Even before these startups selling standalone picks and shovels for carbon removal emerge, I think it makes sense to anticipate these atomic problems and start facilitating initial conversations between carbon removal startups that are looking for initial solutions themselves right now, and then preserving and sharing the knowledge and ideas they’ve acquired. If we can abstract away and grease the wheels for the common and repeatable parts of carbon removal processes, then we allow companies to double down on scaling up the novel parts of their processes.

Do you have other ideas for atomic problems in carbon removal? What am I missing? How should the carbon removal industry as a whole go about addressing atomic problems?