top of page

How crazy is this? Integrating DAC with power plant carbon capture

Updated: Aug 8, 2023

There is a fascinating new preprint out from MIT and the carbon capture startup 8 Rivers, "Optimization of a combined power plant CO2 capture and direct air capture concept for flexible power plant operation". To understand why this paper is important - and why it is incredibly technically challenging - you first need to know four things about the future, renewably-powered grid:

  1. Solar and wind will be the cheapest sources of energy on the grid, but they are intermittent, and tend to disappear unexpectedly for days or even weeks at a time. Even with cross-geographic interconnects and batteries for short term storage, renewables can be relied on for at best 80-90% of a utility's total load.

  2. For the remaining capacity, a dispatchable load is needed, and natural gas will provide, by far, the cheapest such energy for the foreseeable future. But burning gas results in CO2 emissions.

  3. Natural gas plants that operate part time to balance out the variability of renewables are poor candidates for point source carbon capture, because the capex for capture is expensive, and interest payments don't idle when the plant does. As the utilization of a natural gas plant goes down, its suitability for carbon capture goes down as well.

  4. Direct air capture (DAC), where carbon dioxide is removed directly from the air, uses similar infrastructure to the equipment used for point source carbon capture from a natural gas plant. It's not entirely the same, but it's pretty close.

The paper from MIT and 8 Rivers presents a detailed proposal for a carbon capture/DAC hybrid system that is located at a dispatchable natural gas power plant. When the sun is shining and the wind is blowing, the utility will produce clean, cheap energy, which will power the DAC system. When the sun sets and the air stills, the emissions from the natural gas power plant will be scrubbed of CO2. The carbon capture system uses a process called "calcium looping", which isn't mature at the scale of this application, but it's overall fairly well understood. The DAC startup Heirloom is pursing this approach as well. Occidental Petroleum, a company valued at almost $60 billion as I write this, highlights their variation of calcium looping for DAC on the landing page of their website (again, as I write this). The industry has fairly high confidence in the overall technology.

The cleverness of this particular MIT and 8 Rivers study is in the details, not the basic chemistry. The team built the chemical engineering models and did the math to conclude that a calcium looping capture system can be easily "decoupled" from the power plant, meaning that it does not need to be integrated with the thermal systems of the natural gas turbine. This is useful, because most convention carbon capture systems require tight integration to get competitive economics, and it is as a result difficult to append capture to already-existing plants. For the calcium looping process, the natural gas plant and capture plants will share electricity and physical footprint, but little else. It helps to site the plant near a limestone quarry, because that is where the calcium is sourced from. Overall, compromises are few and far between, which makes engineering much easier, and the solution impressively scalable.

Still, this is super-complicated stuff! A high level schematic of the process is shown below, where the power plant is the little blue box, and the carbon capture is contained in the purple, green, and orange boxes. The paper contains three pages of small fonts just to list the parameters that had to be defined for the model. At this stage in human history the world is pretty good at chemical engineering, but it will still take a decade or more before this approach can be perfected.

Still, the MIT/8 Rivers solution meets the minimum goal of offering value to society: The authors estimate the design can pay for itself if carbon credits are valued at about $150/ton, which less less than the best estimate of the "social cost of carbon" (SCC), the damage done by each ton of emissions. The EPA estimates the SCC to be $190/ton, and while the exact figure is somewhat controversial, the speculation here is that this proposed solution will come out cheaper.

Truth be told, we don't know yet what the actual cost will be. There are a lot of details in the paper that could prove wrong, such as assumptions that capital equipment loans can be paid back over 30 years, or that interest on those loans is just 7.25% (a half point less than 30 year mortgages today)! New technologies, even based on well-understood technology like calcium looping, are riskier than these assumptions allow for. Still, this paper provides a good reason for society to look further. If we are serious about limiting our CO2 footprint, we should give proposals like this one a good, long look.

2 views0 comments

Recent Posts

See All

Frontier spreads its bets on CDR

One of the things I find intellectually appealing about Carbon Dioxide Removal (CDR) is that it's amenable to myriad different approaches. It's not obvious that any one of them is right, but it's not

The world tells Big Oil to stay out of offset markets

A few months ago the European oil giant Shell quietly decided to abandon its $100M/yr plans to invest in high quality carbon offsets. Bloomberg covered the story in a fascinating article: Six months a


bottom of page