Power plants that capture CO₂ today use an older technology whereby flue gases are bubbled through organic amines in water, where the carbon dioxide binds to amines. The liquid is then heated to 120-150 degrees Celsius (250-300 degrees Fahrenheit) to release the gas, after which the liquids are reused. The entire process is expensive: it consumes about 30% of the power generated, while sequestering underground costs an additional though small fraction of that.

The removal of CO₂ from low-pressure flue gas mixtures is currently effected by aqueous amine solutions that are highly selective for acid gases. As a result of the large energy penalty for de-sorbing CO₂ from such liquids, solid adsorbents with appreciably lower heat capacities are frequently proposed as promising alternatives. In particular, as a result of their high surface areas and tunable pore chemistry, the separation capabilities of certain metal-organic frameworks have been shown to meet or exceed those achievable by zeolite or carbon adsorbents.

Recently, the attachment of alkyldiamines to coordinatively unsaturated metal sites lining the pores of selected metal-organic frameworks has been demonstrated as a simple methodology for increasing low-pressure CO₂ adsorption selectivity and capacity. Most notably, functionalization of Mg₂(dobpdc) (dobpdc^4- = 4,4'-dioxidobiphenyl-3,3'-dicarboxylate), an expanded variant of the well-studied metal-organic framework Mg₂(dobdc) (dobdc^4- = 2,5-dioxidobenzene-1,4-dicarboxylate), with N,N'-dimethylethylenediamine (mmen) generated an adsorbent with exceptional CO₂ capacity under flue gas conditions and unusual, unexplained step-shaped adsorption isotherms.

Here we elucidate the unprecedented mechanism giving rise to these step-shaped isotherms and demonstrate that replacing Mg²⁺with other divalentmetal ions enables the position of the CO₂ adsorption step to be manipulated in accord with the metal-amine bond strength. As we will show, the resulting mmen-M₂(dobpdc) (M = Mg, Mn, Fe, Co, Zn) compounds, here designated ‘phase-change’ adsorbents, can have highly desirable characteristics that make them superior to other solid or liquid sorbents for the efficient capture of CO₂.

—McDonald et al.

The diamine-appended metal-organic framework before and after binding of carbon dioxide. The view is a cross section through one of the pores of the MOF, showing diamine molecules (containing blue nitrogen atoms) attached to metal (manganese) atoms (green). Carbon dioxide molecules (grey carbon atoms with two red oxygen atoms) bind through a cooperative mechanism akin to a chain reaction along the pore surfaces. Some H atoms (white) are omitted for clarity. (Graphic by Thomas McDonald, Jarad Mason, Jeffrey Long/UC Berkeley) Click to enlarge.

The new diamine-appended MOFs can capture carbon dioxide at various temperatures, depending on how the diamines are synthesized, and releases the CO₂ at only 50 °C above the temperature at which CO₂ binds, instead of the increase of 80-110 °C required for aqueous liquid amines. Because MOFs are solid, the process also saves the energy costs of heating the water in which amines are dissolved.

MOFs are composites of metals—in this case, magnesium or manganese—with organic compounds that, together, form a porous structure with microscopic, parallel channels. Several years ago, senior author Jeffrey Long, a UC Berkeley professor of chemistry and faculty senior scientist at Lawrence Berkeley National Laboratory, and his lab colleagues developed a way to attach amines to the metals in a MOF to produce pores of sufficient diameter to allow CO₂ to penetrate rapidly into the material.

They found that MOFs with attached diamines are very different from other carbon-capture materials, in that the CO₂ seems to load into the material very quickly at a specific temperature and pressure, then come out quickly when the temperature is raised by 50 °C. In the new paper, UC Berkeley graduate students Thomas McDonald and Jarad Mason, together with other co-workers, describe how this works.

This material is unique in that it binds CO₂ in a cooperative mechanism. When the first CO₂ starts to adsorb at a very specific pressure, all of a sudden it facilitates more CO₂ adsorption, and the MOF rapidly saturates. That is really a different property from any other CO₂ adsorbent based on amines. Then, if you raise the temperature by applying heat, at some temperature all the CO₂ will come flooding off.

—Jeffrey Long

Long’s team found that the diamines bind to the metal atoms of the MOF and then react with CO₂ to form metal-bound ammonium carbamate species that completely line the interior channels of the MOF. At a sufficiently high pressure, one CO₂ molecule binding to an amine helps other CO₂ molecules bind next door, catalyzing a chain reaction as CO₂ polymerizes with diamine like a zipper running down the channel. Increasing the temperature by 50 °Celsius makes the reaction reverse just as quickly.

The pressure at which CO₂ binds to the amines can be adjusted by changing the metal in the MOF. Long has already shown that some diamine-appended MOFs can bind CO₂ at room temperature and CO₂ levels as low as 300 parts per million. The current atmospheric concentration of CO₂ is now 400 parts per million (ppm).

Last summer, Long co-founded a startup, Mosaic Materials, to use the new technology to reduce the cost of chemical separations, with plans in the works for a pilot study of CO₂ separation from power plant emissions. This would involve creating columns containing millimeter-size pellets made by compressing a crystalline powder of MOFs.

The researchers are also hoping to develop something that might be tested in a submarine. That would pave the way for eventual scale-up to capturing CO₂ from natural gas plants, which produce emissions containing about 5% CO₂, to the higher concentrations of coal-fired power plants.

We got lucky. We were just trying to find a simple way to attach these amines to our MOF surface, because they are one of the best compounds for selectively binding CO₂ in the presence of water, which can be a problem in flue gas. And it just happens we got the right length in the amine to make these one-dimensional chains that bind CO₂ in a cooperative manner.

—

Long suggested as well that the findings may have relevance for the fixation of CO₂ by plants, owing to striking structural similarities between the magnesium-based MOF and the naturally occurring CO₂-fixing photosynthetic enzyme RuBisCO.

Long received assistance from colleagues at Zhejiang University in Hangzhou, China; the University of Turin in Italy; the University of Minnesota in Minneapolis; the Université Grenoble Alpes and the Centre National de la Recherche Scientifique in France; the Norwegian University of Science and Technology in Trondheim, Norway; and the École Polytechnique Fédérale de Lausanne in Switzerland.

The work is supported by grants from ARPA-E and the US Department of Energy-funded Center for Gas Separations Relevant to Clean Energy Technologies, an Energy Frontier Research Center operated jointly by UC Berkeley and LBNL.

Resources

• Thomas M. McDonald, Jarad A. Mason, Xueqian Kong, Eric D. Bloch, David Gygi, Alessandro Dani, Valentina Crocellà, Filippo Giordanino, Samuel O. Odoh, Walter S. Drisdell, Bess Vlaisavljevich, Allison L. Dzubak, Roberta Poloni, Sondre K. Schnell, Nora Planas, Kyuho Lee, Tod Pascal, Liwen F. Wan, David Prendergast, Jeffrey B. Neaton, Berend Smit, Jeffrey B. Kortright, Laura Gagliardi, Silvia Bordiga, Jeffrey A. Reimer & Jeffrey R. Long (2015) “Cooperative insertion of CO₂ in diamine-appended metal-organic frameworks” Nature doi: 10.1038/nature14327