News & Views | Published:

Materials chemistry

Cooperative carbon capture

Nature volume 519, pages 294295 (19 March 2015) | Download Citation

Enzymes bind carbon dioxide from the atmosphere in a highly precise way, whereas synthetic materials just passively adsorb it. Or do they? A study of compounds called metal–organic frameworks now challenges this picture. See Article p.303

The combustion of fossil fuels to produce heat and electricity creates more than 13 gigatonnes (13 billion tonnes) of the greenhouse gas carbon dioxide each year1, thus contributing to climate change. Eliminating these carbon emissions is currently unfeasible because of the large amounts of energy required to use even the best carbon-capture technologies. On page 303 of this issue, McDonald et al.2 describe a mechanism for 'cooperative', enzyme-like CO2 adsorption in a porous solid that could move us closer to practical carbon capture.

On the face of it, capturing CO2 should be easy: it is an acidic gas that reacts readily with bases. However, immobilizing CO2 is only part of the story; the gas must then be released from the capture medium for sequestration (storage) without the consumption of too much energy. This is where current capture technologies run into problems. For example, solutions of basic amine compounds in water absorb CO2 effectively3, but subsequent heating to release the gas is energetically costly — partly because the amine–CO2 complex that forms is highly stable, but also because heating water requires a lot of energy.

In principle, physical adsorption of CO2 in porous solids offers lower-energy penalties than amine solutions4, but has not yet been realized in a practical, scalable way for carbon capture. One problem is that the physical adsorption of CO2 by many porous solids is weak, so that not enough gas can be taken up within a workably small volume of material. This is effectively the reverse problem of that of amine solutions: these solids do not adsorb CO2 strongly enough.

Some porous solids do adsorb CO2 strongly, but also readily adsorb water, which can outcompete CO2 under 'wet' conditions5. There is therefore a great need for porous solids that adsorb CO2 and then desorb it within small pressure or temperature changes, to minimize energy losses, and that do this without competition from water vapour — which will be produced in any process that involves fossil-fuel combustion.

Crystalline metal–organic frameworks (MOFs) are constructed from metal-containing nodes connected by organic linkers (ligands)6. MOFs are attractive candidates for CO2 capture and separation4 because they can be porous, and because both the metals and the organic ligands can be varied in a modular way. But despite their broad structural diversity, preparing porous MOFs that have practical potential for CO2 capture has proved difficult. In particular, even if issues of chemical stability7 are overcome, it is hard to make MOFs that adsorb and release a large volume of CO2 over a narrow pressure or temperature range.

McDonald et al. have made a major conceptual advance by identifying a series of compounds that they call “phase change” adsorbents: that is, MOFs that adsorb CO2 by chemical insertion of the gas into the metal–ligand bonds of the framework. Crucially, the strength of the metal–ligand bonds, and hence the ease of insertion and removal of the CO2, can be tuned systematically by varying the metal to create a series of frameworks. Using this modular chemical control, the authors have produced materials for which the CO2-adsorption isotherm — a graph of the amount of CO2 absorbed by a material versus pressure at fixed temperature — approaches an ideal 'binary' plot. In other words, the gas is adsorbed and then desorbed over an extremely narrow pressure window (see Fig. 5 of the paper2).

Such step-like isotherms have previously been observed in MOFs. For example, some MOFs have surprisingly flexible structures8, which can lead to the sharp uptake of gas at a particular pressure. But the change of volume observed in the solid when McDonald and co-workers' MOFs adsorb CO2 is too small for this to be the operative mechanism. To explain the observed isotherms, insertion of CO2 into the metal–ligand bonds of the authors' MOFs must happen in a concerted way at a specific CO2 pressure that varies with the strength of the metal-ligand bond. But how?

The authors probed this question using a combination of spectroscopy, X-ray diffraction and computation, and it seems that the answer lies in cooperativity. In effect, the first CO2 insertion creates a carbamate group that destabilizes the neighbouring metal–ligand site in the MOF, facilitating the insertion of another CO2 molecule at that position. This creates a chain reaction that propagates to form a one-dimensional line of carbamates in the framework (see Fig. 4 of the paper2). This mechanism somewhat parallels the 'self-accelerating' adsorption of carbon monoxide in a soft porous crystal that was reported last year9.

Remarkably, the mechanism and the CO2-capture performance persist in the presence of water, and the adsorption–desorption cycle can operate at elevated temperatures (between 100 and 150 °C, for example). The ability to function at high temperatures is pivotal for many applications, because the temperature of the flue gas from fuel combustion is typically higher than ambient temperatures. The mechanism also raises issues that might affect industrial-scale applications of the MOFs. For example, rapid step-like CO2 adsorption might be accompanied by a sudden release of heat. This would need to be dissipated quickly so that the material does not warm spontaneously, thus desorbing the CO2 prematurely.

McDonald and colleagues' study takes full advantage of the crystalline nature of MOFs. Porous materials with extremely high surface areas can be made from robust amorphous polymer networks10, but it is hard to imagine the observed cooperative mechanism working efficiently in a disordered material. Indeed, the mechanism is a function of the precise relative placement of the chemical groups in the framework, and of the strength of the metal–ligand bonds.

This mechanistic feature brings to mind biological systems, such as enzymes, whose functions are also determined by the relative placement of organic species around specific metal centres. As the authors note, their magnesium-containing MOF shows strong structural similarities to the active site of the ribulose-1,5-bisphospate carboxylase/oxygenase enzyme (Rubisco), which is responsible for biological CO2 fixation (Fig. 1). This may not be a coincidence: the superior CO2-capture capacity of the magnesium MOF at low CO2 pressures might shed light on why magnesium ions are found in Rubisco, which also functions at low CO2 concentrations. The authors' work therefore suggests that future developments in carbon capture might be bioinspired.

Figure 1: Similarities between Rubisco and a metal–organic framework (MOF).
Figure 1

a, The Rubisco enzyme is responsible for the biological fixation of CO2. Its active site contains a magnesium (Mg) ion that binds to five ligands (red) through oxygen atoms and to a reactive amine ligand (blue); CO2 (green) is adsorbed when the gas inserts between the amine and the magnesium ion. The amino acids that donate three of the ligands are designated Lys 201, Glu 204 and Asp 203. Broken lines are metal coordination bonds. b, McDonald et al.2 report a magnesium-containing MOF in which magnesium ions are bound within a similar chemical environment. This similarity might explain why Rubisco and the MOF can successfully fix (in the case of Rubisco) or adsorb (in the case of the MOF) gaseous CO2 at low CO2 concentrations. Ph, Phenyl group.



  1. 1.

    International Energy Agency. CO2 Emissions From Fuel Combustion: Highights (IEA, 2013); see

  2. 2.

    et al. Nature 519, 303–308 (2015).

  3. 3.

    Science 325, 1652–1654 (2009).

  4. 4.

    et al. Chem. Rev. 112, 724–781 (2012).

  5. 5.

    et al. J. Am. Chem. Soc. 134, 10741–10744 (2012).

  6. 6.

    , , & Science 341, 1230444 (2013).

  7. 7.

    & Angew. Chem. Int. Edn 53, 7471–7474 (2014).

  8. 8.

    et al. J. Am. Chem. Soc. 124, 13519–13526 (2002).

  9. 9.

    et al. Science 343, 167–170 (2014).

  10. 10.

    et al. Angew. Chem. Int. Edn 48, 9457–9460 (2009).

Download references

Author information


  1. Andrew I. Cooper is in the Department of Chemistry, University of Liverpool, Liverpool L69 3BX, UK.

    • Andrew I. Cooper


  1. Search for Andrew I. Cooper in:

Corresponding author

Correspondence to Andrew I. Cooper.

About this article

Publication history



Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

Newsletter Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing