The Carbonate Solution

Beyond the “brute force” approach that I wrote about last week, I want to cover three other ways to exploit the chemistries of common carbonate minerals for large scale carbon capture and storage. They are less direct, and each has its own issues, but all three are much more energy-efficient than brute force calcination of limestone. The latter takes 3.17 gigajoules (GJ) minimum per tonne of calcium carbonate, and more like 4.5 to 6 GJ per tonne in practice.  The alternatives have no corresponding theoretical minima that can be readily calculated, but practical needs should be at least a factor of 10 lower.

CCS for coastal power plants

In a CCS approach advocated by researchers at Stanford, UCSC, and LLNL, [q.v.] flue gas from a power plant is distributed and piped upward through a large bed of crushed limestone. The limestone is wetted by a flow of seawater trickling downward. CO₂ in the flue gas acidifies the flow, enabling it to dissolve calcium carbonate from the limestone surface. Upon dissolving, the carbonate ions (CO₃²⁻) take up hydrogen ions, becoming bicarbonate ions (HCO₃⁻). That neutralizes carbonic acid, enabling more CO₂ to dissolve. The dissolved CO₂ makes more carbonic acid, which in turn dissolves more CaCO₃. At the bottom of the bed, the water, now laden with calcium bicarbonate, is collected and sent back to the ocean.

That’s probably enough explanation for most readers, but the figure below gives a starting point for those interested in the physical chemistry. It’s a graph of relative concentrations of dissolved inorganic carbon (DIC) species (CO₂ / H₂CO₃, HCO₃⁻, and CO₃²⁻) in water, as a function of pH. The shaded blue bar shows the pH range of ocean waters, and the arrowhead shows how the range will shift in response to increasing CO₂ concentrations in the atmosphere. From the graph, we see that at a pH of 7,  the relative concentration of carbonate ions is about 1%; that of neutral CO₂ / carbonic acid is about 8%, while the remainder is bicarbonate.

Since the molar concentration of total DIC in seawater is only ~2.3 mmol / liter to begin with, the concentration of carbonate ions in seawater at pH 7 would be an extremely low 23 μmol / liter. With such a low concentration, the rate at which paired calcium and carbonate ions at the CaCO₃crystal surface escape into solution will be very much greater than the rate at which calcium and carbonate ions from solution will pair up on the surface; the solid CaCO₃ will dissolve.

The chief advantage of this approach to carbon mitigation is that it combines capture and storage. It is energy-efficient and doesn’t present a significant parasitic load on the power plant’s output. Also, the seawater used to irrigate the limestone bed and capture CO₂ can be the same water that was used for cooling. On the downside, the process consumes limestone at a high rate, and is practical only for coastal power plants located close to sizable limestone deposits.

To see that in quantitative terms, consider the overall reaction that this process implements. In the case of a coal-fired power plant, it’s:

C + O₂ + H₂O + CaCO₃ → Ca(HCO₃)₂ + energy

To a first approximation, coal can be taken as pure carbon, with a mass of 12 grams per mole. Limestone can be taken as pure calcium carbonate, with a mass of 100 grams per mole. That means that for every ton of coal burned, 8.3 tons of limestone are also consumed.

In practice, consumption would be closer to 6 tons of limestone, since coal is not pure carbon and not all of the CO₂ produced in burning it can be captured. But 6 tons of limestone per ton of coal is still a lot. It’s not that there’s any shortage of it in the world; limestone is hundreds of times more abundant than fossil carbon. But the logistics of mining, crushing, and transporting the required amounts are daunting. Consider: a large coal-fired power plant consumes one trainload of coal each day; this scheme would add six trainloads of limestone per day on top of that. Every four hours, a full trainload of crushed limestone would need to be added to the limestone bed to replace what was dissolved in the preceding four hours! That’s perhaps not impossible, but it will require a large limestone bed of at least several hectares..

For gas-fired power plants — which is most of what California has — the amount of CO₂ per megawatt-hour is only half as great; the amount of limestone consumed in capturing and storing it would likewise be half as great. If the plant were only used at an average duty cycle of 30% for backing wind and solar, it could get by with “only” one trainload of limestone per day. Given that this approach is one of the most efficient ways to capture and store CO₂ from gas or coal-fired power plants, the figure of one trainload of limestone per day to capture the carbon emissions from a single large gas-fired plant used merely to back wind and solar resources is sobering. It gives one a feel for the scale of the carbon emissions problem.

Ocean capture

The main problem with the above approach is its limited scope. As noted, it’s only applicable for coastal power plants within easy transport distance of limestone beds. There are more of those than one might think, but in the context of total anthropogenic carbon emissions, they’re not that significant. Completely capturing and storing their CO₂ would only reduce total fossil carbon emissions by perhaps 2%. Is there a way to do better? In principle, there is. However it requires — among other things– a different approach to dissolving CaCO₃.

Although CaCO₃, isn’t soluble in surface seawater at pH of 7.8 or higher, it becomes soluble at higher pressures and lower temperatures — conditions that can be found in the deep waters of the ocean. That suggests the possibility of simply dumping freighter loads of crushed chalk or limestone into deep ocean waters around the world. The dumped material would dissolve and natural carbonate to bicarbonate conversion would raise the pH in the deep waters. When those waters reached the surface, they would draw CO₂ from the atmosphere. Would that work?

Yes and no. With a carbon tax, the economics could possibly work. The ratio of crushed limestone to captured CO₂ is about 9:4. A carbon tax of $30 / tonne of CO₂ — toward the low side of what’s often mooted — would translate to $13.50 / tonne of limestone quarried, crushed, and shipped to a deep ocean area. That’s probably not out of reach for profitability for a large scale operation. But the timescale doesn’t work. The natural timescale for deep water alkalinity to migrate to surface waters is upward of 1000 years. No help over the next few centuries, let alone the next few decades.

There’s a possible way around the timescale problem that would be worth trying.  That’s to float a large diameter vertical pipeline in the ocean to pump cold water from the depths where the chalk and limestone dissolve up to the surface. As it happens, that’s also exactly what’s needed for an Ocean Thermal Energy Conversion (OTEC) power plant. So the system could be designed as a renewable energy resource while simultaneously countering ocean acidification, removing CO₂ from the atmosphere, and lowering global temperatures.

I won’t go into details here, but I believe there’s a way to engineer a deep cold water pipe that would have very low capital and operational costs relative to capacity. The design I’m thinking of, if workable, would go a long way toward making OTEC installations economically feasible. An interesting feature, in the context of this discussion, is that a good part of the energy needed to lift the denser cold water to near the surface could be supplied by a bucket conveyor running beside the cold water pipe. It would carry crushed chalk and limestone to depths at which it would dissolve, generating power in the process.

It’s unlikely that the arrangement could generate any net power, because after dissolving, most of material lowered by the conveyor would be carried back up the cold water pipe. But it would at least reduce the amount of power diverted from the OTEC generators to drive the flow up the cold water pipe. And it might possibly do more than that: a small amount of heat is released by the dissolution of solid carbonate minerals and conversion to bicarbonates. I believe it’s only enough to warm the ascending water flow by a couple of degrees, but that’s enough to lift the water through the lowest two kilometers of the pipe before pumping energy would need to be supplied.

I need to run more calculations to quantify the parameters for this type of extended OTEC system. But I haven’t figured out a good way to quantify what could be the system’s most important climate effect: the lowering of surface water temperatures in the down-current vicinity of the platform. If it’s able to reduce the absolute humidity of the air blowing over the cooler surface water, the reduced greenhouse effect from that air mass could be important for mitigating global warming. It could conceivably be more important than the captured CO₂. At this point, I simply don’t know.

One thing that is easy to calculate, however, is the total amount of chalk and limestone that would ultimately need to be consumed annually, if this approach were used to capture and store the entire 40 annual gigatonnes of anthropogenic CO₂ emissions. It comes to 91 gigatonnes. For context, the Wikipedia article on bulk carriers cites 2005 data for the total tonnage of all bulk cargo shipments worldwide as 1.7 gigatonnes. So we’re looking at more than a 50-fold increase over that number, just for hauling crushed chalk and limestone from quarry sites to extended OTEC stations.

Accelerated weathering in soils

The final carbonate chemistry approach to CCS that I want to cover is a bit different. It uses the soils in warm tropical regions rather than the oceans for air capture and storage of CO₂. Instead of dissolving limestone, it uses the natural alkalinity of olivine and other ultramafic silicates found in basalt rocks to mineralize and permanently sequester CO₂.

Olivines have the generic formula (Mg²⁺, Fe²⁺)₂SiO₄. The notation (Mg²⁺, Fe²⁺) indicates a mix of magnesium and iron atoms in the +2 oxidation state. In the case where the mix is 100% magnesium (Mg₂SiO₄) the weathering reaction can be represented simply as:

Mg₂SiO₂ + 4CO₂ + 2H₂O → 2Mg²⁺ + 4HCO₃⁻ + H₄SiO₄

The products on the right hand side are soluble. In weathering by naturally acidified rainfall, they’re removed by the runoff water as magnesium bicarbonate and silicic acid.  The latter, when it reaches the ocean, fertilizes the growth of siliceous phytoplankton know as diatoms.

Accelerated weathering has received substantial academic and some commercial attention. One of its most active proponents is Dr. Olaf Schuiling of the University of Utrecht, Netherlands. In addition to his own research and published papers, he started the Olivine Foundation, which maintains the SmartStones website. I won’t attempt to summarize a very large topic that is well-covered elsewhere (e.g., here), but there’s one particular approach that I find interesting. It’s one that has been dismissed as nice in theory, but not practical. It’s the spreading of serpentine sand or crushed ultramafic minerals over soils where they will rapidly weather.

Forest soils in warm, wet climates are ideal places for accelerated weathering to take place. Due to decomposition of leaf litter and other dead organic matter, the CO₂ concentration in such soils can be quite high. In fact, it can approach the concentration found in flue gases from a power plant. That means that the CO₂-induced acidity of liquid water in these soils is many times higher than it is in raindrops. That makes for a high rate of chemical weathering for any rock particles that happen to be present.

If olivine sand or similar crushed minerals could be spread over the forest floors, it would build up and fertilize the soil while soaking up much of the CO₂ that would otherwise be released by decay of plant matter. But there is no existing way to do that efficiently and economically. The sheer amount of material and the vast areas over which it would need to be distributed before it could make a real difference have relegated the approach to the “neat idea, but..” category.

I may have a solution for that. Again, it’s too much to write about here, but I may have more to say about it after I’ve done some more homework.

Conclusions

What I’ve tried to show in this series of postings is in part just the sheer magnitude of the CCS challenge. There are a range of technically feasible approaches, but even the best of them involve a degree of capital investment and operations that are staggering, if the approach is implemented at the scale needed to offset all emissions.

It makes sense to do what we reasonably can with resources available; a carbon tax would make many of these approaches profitable. But it will come as no surprise to opponents of CCS that the main thrust of world mitigation efforts still needs to be reducing our use of fossil carbon in the first place.

I don’t personally care much whether the reduction is achieved through conservation and energy efficiency, renewables, or nuclear power. Common sense would dictate focusing our resources on what will give us the most mitigation “bang” for the buck. But political feasibility does factor in, and in any case there won’t be any one silver bullet. The problem of decarbonization and the urgency of moving on it are such that an “all of the above” approach will be necessary.

Energy Collective