Section 1: Global and Regional Views of Carbon Management

Removing Carbon Dioxide from the Air to Stabilise the Climate

Richard C Dart on[1] [2] and Aidong Yang


The rate at which greenhouse gases accumulate in the atmosphere can be reduced by cutting emissions, but elevated levels of CO, will persist for centuries. Climate models show that peak CO,-induced warming depends mainly on cumulative emissions and not the emission pathway (Matthews, 2018). It is, therefore, expected to become necessary to remove CO, from the atmosphere in order to counter unacceptable climate change later in this century. Negative Emission Technologies (NETs) are the methods proposed for this. Nitrous oxide (N,0) and methane (CH4) mainly emitted by agriculture and industry are also significant contributors to global wanning, but this chapter focuses on carbon dioxide. NETs for CO, are often termed carbon dioxide removal (CDR). It is a more efficient use of energy to capture carbon from flue gas than from the air, but even if applied quickly on a vast scale, it is expected that Flue Gas Capture (FGC) on its own will not be sufficient to meet policy goals—for example, peak wanning of

1.5 °C above pre-industrial average temperature, the aspiration agreed at Paris in 2015 (Minx et al., 2018). The Intergovernmental Panel on Climate Change, which has reported the causes and effects of global wanning in great detail, is clear on the need to apply CDR:

“All pathways that limit global wanning to 1.5 °C with limited or no overshoot project the use of carbon dioxide removal (CDR) on the order of 100-1000 GtCO, over the 21st century” (EPCC, 2018).

It is difficult to predict the precise tuning and scale of CDR that might be needed, and many emissions scenarios har e been considered by the IPCC. Figure 1 shows a typical scenario to the year 2100, in which peak warming is limited to 2 °C, and CDR is introduced from 2030 to ramp up to the scale needed. The ambition here is to draw-down around 810 Gt of atmospheric CO, (range 440-1020 Gt) by the year 2100 and sequester it somewhere securely (UNEP, 2017). Greater emissions reductions could reduce this targeted draw-down (IPCC, 2018). Fuss et al. (2018) indicate a total NETs deployment across the 21st century equivalent to net draw-down of 150-1180 GtCO, as necessary to meet the 1.5 °C target.

Whilst it is relatively simple to incorporate NETs into climate models, whether they can be applied at the scale needed or within the time frame assumed remains an open question. The impacts on people and ecosystems need to be considered, for these will be very large industries. Some NETs rely on well-

Emissions scenarios to 2100 (Gt CO, equivalent) showing the potential need forNETs (UNEP, 2017 - Figure 7.2)

Figure 1. Emissions scenarios to 2100 (Gt CO, equivalent) showing the potential need forNETs (UNEP, 2017 - Figure 7.2).

known and proven technologies, but others still require a significant amount of research and development (Nemet et al., 2018). Moreover, recognising the objective of removing greenhouse gases from the air raises questions, such as how it will be paid for and managed. At the very least, NETs will compete with other desirable activities for resources and investment. Conjuring these industries into existence and then operating them for, say, a century, with the objective of managing the global climate, is an undertaking unlike any previously attempted.

It is important to distinguish between methods of capturing carbon dioxide from flue gas or other waste gases (Carbon Capture and Storage - CCS), and carbon dioxide removal from the atmosphere (CDR) followed by storage. Both CCS and CDR are responses to our failure to curb emissions of greenhouse gases sufficiently, but they have different characteristics:

  • a) CCS is a means of reducing emissions by using a process to capture carbon dioxide before it is released into the atmosphere, and storing the captured CO, securely. CCS is intended to reduce emissions particularly from large stationary sources of CO,—power plants burning fossil fuel, cement factories, chemical plants and the like. Capturing CO, from distributed sources like motor cars, aeroplanes and agriculture would be more difficult to do, and is not really envisaged. Deploying CCS will slow the global rate of emissions but will not affect the CO, already present in the atmosphere.
  • b) CDR takes carbon dioxide out of the atmosphere into secure storage. It does not distinguish between CO, arising from natural processes and from man-made emissions. Chemical processes for CDR need more energy than chemical processes for CCS, because the concentration of CO, in the air is very low. Deploying CDR to an increasing extent will first decrease the rate at which CO, is accumulating in the atmosphere; then, when the rate of CDR exceeds the rate of CO, emission (i.e., net negative CO, emissions), the atmospheric CO, concentration will start to fall. When the rate of CDR equals the rate of emission of CO, and all other greenhouse gases (in terms of СО,- equivalence), we will have net-zero emissions. In the scenario of Figure 1, this is projected to occur around 2090, after which we will have net negative greenhouse gas emissions.

Secure storage of CO,, meaning that it is removed from the atmosphere for a long time, is sometimes termed “sequestration” and CCS can also be taken to stand for Carbon Capture and Sequestration. Requirements for storage longevity, monitoring and verification are discussed in section 3.8. Another phrase sometimes encountered is Carbon Capture and Utilisation (CCU), describing the manufacture of products using biomass or captured CO,. The extent to which CCU can be regarded as either CCS or CDR depends on whether the product offers net long-term storage of CO,—this is discussed in sections

3.7 and 3.8.

In this chapter, we first describe, in section 2, the technologies that might be used to capture CO, from the atmosphere, and indicate the chemistry involved. In section 3, we review options for storing the captured CO,. Using abiotic sorbents to capture CO, is analysed in section 4, focussing in particular on the energy required and including an estimate of costs. Finally, we comment in section 5 on implications for setting policy.

Technologies for Capturing Carbon Dioxide from Air

Proposals to remove carbon dioxide from air har e been reviewed by McLaren (2012), McGlashan et al. (2012), Fuss et al. (2018), Royal Society and Royal Academy of Engineering (2018) and the National Academies of Sciences, Engineering, and Medicine (2019). Broadly speaking, there are two main types of capture—by photosynthesis, which causes plant growth, or by means of an abiotic sorbent.

Capture by photosynthesis

In photosynthesis, growing biomass absorbs carbon dioxide from the atmosphere. It reacts with water in the plant to form carbohydrate and oxygen in a reaction which can be represented as

Capture by photosynthesis is the basis of the coupled CDR system, BECCS (Bioenergy with Carbon Capture and Storage), which takes CO, from the air by growing crops that are used as fuel to provide useful heat or power. Carbon dioxide is then captured from the flue gas (a second capture of the CO,) and sequestered. Photosynthesis has the key advantage that it is powered by sunlight. However, growing plants specifically to capture carbon dioxide, whether in the sea or on land, would compete for space with fishing, fanning and other human activity and also with the need for preserving undisturbed ecosystems.

Capture by abiotic sorbent

Abiotic sorbents can be solid (adsorbents), or liquid (absorbents). In both cases, the sorbent has an affinity for carbon dioxide in the ail'. In Direct Air Capture (2.2.1), the sorbent is exposed to the air, then regenerated and re-used in a chemical process similar to industrial gas treating. The energy required to regenerate the sorbent is a major expense incurred in this route. Alternatively, in Enhanced Weathering and similar mineralisation schemes, air is contacted with rocks and minerals which react with CO, to form a stable product (2.2.2). The cost of regenerating the sorbent is avoided, though there will be other processing expenses.

2.2.1 Direct Air Capture (DAC)

The process parameters of DAC are rather different to most current industrial gas separations. Very large volumes of gas must be treated, potentially involving a lot of very large equipment: the pressure drop and efficiency of this equipment then become critical features of the design. The partial pressure of CO,, the contaminant to be captured, is relatively low (~ 0.0004 bar), making it unsuitable for purely physical absoiption processes—chemical reaction or chemisorption are needed. Fortunately, carbon dioxide is a fairly reactive chemical, and there are many proven industrial processes that can be used to remove it from various gases. Amongst these applications is flue gas CCS, usefully reviewed by the IPCC (Metz et al., 2005).

A well-known commercial solvent for carbon dioxide capmre is aqueous potassium carbonate. The reaction is

This reaction however is rather slow, and to reduce the size of the equipment needed, the absoiption is often speeded up by adding a homogeneous catalyst, such as arsenite or hypochlorite (Danckwerts and

Sharrna, 1966; Astarita et al., 1981). Fast-reacting promoters, such as piperazine, can also be added to potassium carbonate to enhance the absorption rate and reduce equipment size (Cullinane and Rochelle, 2004).

Another important class of chemical absorbents much used in oil. gas and chemicals processing are alkauolamines in aqueous solution. For example, with monoethanolamine (MEA), the main reaction chemistry is a carbamate formation

where R stands for the ethanolic group HOC,H4. The rate of reaction (3) is much gr eater than that of the uncatalysed reaction (2), which accelerates the absorption of CO, into the amine solvent.

In industrial processes for carbon dioxide removal, the sorbent is regenerated for use again by raising its temperature, which drives the equilibrium (e.g.. reactions (2) and (3)) back towards the left- hand side. Regeneration requires energy which is commonly provided by low pressure steam, at around 120-130 °C, and it produces carbon dioxide as an off-gas at a pressure and purity depending on the process design. If a liquid solvent is used, it is pumped around in a cycle between absorber and regenerator vessels. In this cycle, the hot regenerated solvent is cooled against the loaded solvent, and then cooled further against ambient air or water in a “trim cooler” prior to entering the absorber again. Solid adsorbents are seldom moved around, so it is common to have several beds of adsorbent operating in parallel. When one is sufficiently loaded with CO„ the inlet flow is switched to a fresh bed, whilst the loaded bed is regenerated and the CO, is driven off by raising the temperature and/or lowering the pressure. In this way, the adsorbent particles experience an adsorption/desorption cycle with pressure/ temperature swing, albeit without moving.

Research has been directed to developing tailor-made Direct Ah Capture technology. Conventional sorbents and existing proven process technology might be used, but novel sorbents, process line-ups and equipment can offer advantages for the specific requirements of ah capture (e.g., Shi et al., 2016; Keith et al., 2018). Following the capture process, provisions must be made for utilising or storing the product carbon dioxide. If it is intended to compress the CO, for underground storage, a rather high purity is required, usually greater than 96%.

2.2.2 Enhanced Weathering (EW) and other mineralisation schemes

In these schemes, the sorbents are rocks or minerals generally in particulate form. They are not regenerated, and the reaction product comprises the long-term storage of the drawn-down CO,.

For example, the “weathering” reaction of olivine, a silicate rock, with rainwater and CO„ can be represented as

and the weathering of carbonates, such as calcite, as

Once present in groundwater, the soluble products of these reactions can be stable for centuries, representing a net removal of carbon dioxide from the atmosphere. However, in a different environment, if the water flows into the sea, for example, further reactions will occur. These might typically be

These mineralisation reactions release carbon dioxide back to the atmosphere, but in the case of silicates, reactions (4) and (6) still cause a net draw-down of CO,. However, for carbonate rock, reaction (7) is the reverse of (5), and the net effect of both together is the transport of carbonate from one location to another (which may provide a temporary period of storage), with ultimately no net change in atmospheric carbon dioxide concentration. Engineered schemes to promote capture and storage through these reactions are known as Enhanced Weathering (EW).

Closely related to EW is the use of alkaline wastes from manufacturing or mining operations which similarly react with atmospheric carbon dioxide in the presence of water (Renforth, 2019). It is also possible to make a highly active particulate adsorbent by calcining carbonate minerals. For example, for calcium carbonate as present in limestone the chemistry is

Reaction (8) is the calcining reaction which, in order to produce sufficient partial pressure of carbon dioxide in the off-gas, needs to take place at temperatures above 800 °C. hi (9), the calcium oxide is “slaked” with water, a highly exothermic reaction. The resulting alkali can be exposed to atmosphere, and it will draw down CO, as in reaction (10). To maximise net positive removal, the carbon dioxide generated in (8) is captured and sequestered. The calcining operation can be engineered to produce rather pure CO„ which facilitates CCS in this case. The alkali can be used in Ocean Liming (section 3.6), or perhaps in flue gas treating at coastal locations (Rau, 2011).

  • [1] Department of Engineering Science, University of Oxford, Parks Road, Oxford OX1 3PJ, UK.Email: This email address is being protected from spam bots, you need Javascript enabled to view it
  • [2] Corresponding author: This email address is being protected from spam bots, you need Javascript enabled to view it
< Prev   CONTENTS   Source   Next >