Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Selective capture of carbon dioxide from hydrocarbons using a metal-organic framework


Efficient and sustainable methods for carbon dioxide capture are highly sought after. Mature technologies involve chemical reactions that absorb CO2, but they have many drawbacks. Energy-efficient alternatives may be realised by porous physisorbents with void spaces that are complementary in size and electrostatic potential to molecular CO2. Here, we present a robust, recyclable and inexpensive adsorbent termed MUF-16. This metal-organic framework captures CO2 with a high affinity in its one-dimensional channels, as determined by adsorption isotherms, X-ray crystallography and density-functional theory calculations. Its low affinity for other competing gases delivers high selectivity for the adsorption of CO2 over methane, acetylene, ethylene, ethane, propylene and propane. For equimolar mixtures of CO2/CH4 and CO2/C2H2, the selectivity is 6690 and 510, respectively. Breakthrough gas separations under dynamic conditions benefit from short time lags in the elution of the weakly-adsorbed component to deliver high-purity hydrocarbon products, including pure methane and acetylene.


Chemical separation processes consume vast quantities of energy1. Economical and practical pathways to alleviating this burden are required. This is especially relevant to the capture of CO2, which is a common impurity in crude gas streams. CO2 removal is integral to upgrading natural gas and biogas, for example, and to the purification of valuable hydrocarbons prior to polymerisation or chemical derivatization2. These processes are separations that rely on discrimination between CO2 and other gases. One established technology is to trap the CO2 by a chemical reaction with an absorbent. This typically involves chemisorption to an amine in aqueous solution3,4. Chemisorption incurs multiple drawbacks, however, including a high energy penalty during regeneration, amine losses due to degradation and evaporation, and the corrosion of hardware and pipelines5. Other conventional separation methods involve solvent extraction or cryogenic distillation, which are burdened with a high energy penalty and large amount of solvent waste.

The physisorption of CO2 in nanoporous materials is an attractive alternative6,7. Physisorption is governed by weak, noncovalent bonding interactions in pores that are structured on the molecular scale8. Ideally, they lower the energy requirements for regeneration since driving off the trapped CO2 simply involves breaking interactions that are inherently weak. Effective physisorbents combine rapid guest diffusion, recyclability and long-term stability with selectivity for CO2 over competing gases at relevant concentrations9. Thus, they may offer a sustainable solution to CO2 capture. In this context, metal-organic frameworks (MOFs) have risen to prominence10,11,12,13,14. MOF materials are built up from metal ions and organic ligands, and their pore shape, size and chemical environment can be systematically designed15,16. In turn, this allows interactions between framework hosts and molecular guests to be tailored. In the search of effective MOF physisorbents, simply searching for materials with ever-higher levels of CO2 uptake per se may not deliver adsorbents that are adept at gas separations since the adsorption of non-CO2 components may also increase. Instead, significant advances will emerge by suppressing the uptake of these competing gases17,18, developing scalable synthetic protocols, mitigating the impact of common impurities such as water vapour and oxygen, and developing low energy pathways to adsorbent recycling.

The removal of CO2 from hydrocarbons is an important process2. While natural gas and biogas are primarily composed of methane (at high pressure and low pressure, respectively), contamination by CO2 can prevent optimal heat release from gas combustion, and cause pipeline corrosion and dry ice formation19. MOFs, however, offer a means of reducing the CO2 concentration in the presence of dominant quantities of methane10,20,21. Acetylene (C2H2) is an essential feedstock for the industrial production of commodity materials22,23. When acetylene is generated, however, it typically coexists with CO2 impurities24. The separation of C2H2 and CO2 is challenging due to their similar physical properties (Supplementary Table 4). MOF physisorbents offer a potential solution but most show an affinity toward C2H2 rather than CO211. The selective adsorption of the CO2 component has seldom been reported despite its operational simplicity in process design and the promise of energy efficiency. Conversely, gas purification using hydrocarbon-selective MOFs requires additional stages if the eluent is contaminated by adsorbed CO2 during the desorption step25. Despite recent advances in MOF chemistry, challenges remain in producing framework adsorbents that combine good separation capabilities with wider performance characteristics such as scalability, recyclability and easy low-energy regeneration. MOF adsorbents that may be applied to methane purification and that preferentially adsorb CO2 from other hydrocarbons are in particular demand.

In this work, we present a MOF, termed MUF-16 (MUF = Massey University Framework) that exhibits inverse selectivity: the adsorption of carbon dioxide in preference to hydrocarbon guests. The carbon dioxide is efficiently sequestered by hydrogen bonding and a range of other favourable noncovalent interactions. This underpins high selectivities for the separation a range of gas mixtures that are relevant to natural gas and industrial feedstocks. Being economical to produce on scale, stable and recyclable, MUF-16 has many of the qualities of an attractive adsorbent.


Synthesis and characterisation

Inspired by the superb properties of MOFs derived from straightforward and readily-available linkers26,27, our interest was captured by the MUF-16 series of materials. These frameworks are prepared by combining 5-aminoisophthalic acid (H2aip), an inexpensive, commercially-available linker, with cobalt(II), nickel(II), or manganese(II) salts in methanol (Fig. 1a). This delivers compounds with the general formula [M(Haip)2]28,29, referred to as MUF-16 (M = Co), MUF-16(Ni) and MUF-16(Mn), respectively. These easily-handled crystalline materials are high yielding on gram scales and tolerant to oxygen and water vapour. Their crystal structures were determined by single crystal X-ray diffraction (Supplementary Table 1). The three frameworks are isostructural, belonging to the I2/a space group. Individually, the metal ions adopt an octahedral geometry with four carboxylate and two amino donors arranged trans to one another. These ions are aligned into one-dimensional chains along a crystallographic axis supported on each side by μ2-bridging carboxylate groups (Fig. 1b). Adjacent chains are connected into two-dimensional sheets by Haip ligands that extend across the plane by coordinating to adjacent one-dimensional chains with both their amino and carboxylate donors (Fig. 1b). Only one of the two carboxyl groups of each Haip ligand coordinates to the metal. The other remains protonated and engages in hydrogen-bonding with a partner from an adjacent layer (Fig. 1c). These interactions link the layers into three-dimensional frameworks. The frameworks support one-dimensional channels of approximately 3.6 × 7.6 Å (accounting for the van der Waals surfaces of the atoms, Fig. 1d). In their as-synthesised form the pores contain occluded water, which can be easily removed by heating at 130 °C in vacuo.

Fig. 1: Synthesis and structure of MUF-16 materials.

a Synthetic routes to the MUF-16 family and optical micrographs of the reaction products. b Infinite secondary building units (iSBUs) in MUF-16 comprise one-dimensional cobalt(II) chains connected by μ2-bridging carboxylate groups of the Haip ligands (H2aip = 5-aminoisophthalic acid). The cobalt(II) ions are depicted as filled octahedra. c The iSBUs are linked into planar two-dimensional sheets by the Haip ligands and further connected into a three-dimensional framework by hydrogen bonding (depicted as dashed lines) between adjacent sheets. d MUF-16 features one-dimensional channels with approximate dimensions of 3.6 × 7.6 Å that propagate through the framework. The Connolly surface of the framework is shown in orange and defined with a probe of diameter 1.0 Å. Colour code: Co = magenta; O = red; C = grey, N = blue.

Thermogravimetric analysis demonstrated the thermal stability of the MUF-16 materials beyond 330 °C (Supplementary Fig. S2). Their purity was established by both elemental analysis and powder X-ray diffraction (Supplementary Fig. S5). The frameworks are chemically robust, being unaffected by soaking in water or exposure to humid air for prolonged periods, as confirmed by powder X-ray diffraction and gas adsorption analysis (vide infra and Supplementary Figs. S6S8, S13a).

As suggested by pore evident in their SCXRD structures, the MUF-16 frameworks are accessible to a range of incoming gases. Nitrogen adsorption isotherms measured at 77 K gave BET surface areas of 214, 205 and 204 m2/g for MUF-16, MUF-16(Mn), and MUF-16(Ni), respectively (Supplementary Figs. S19S21). Total pore volumes of 0.11 cm3/g were established for all three frameworks (Supplementary Table 3). These values are comparable with the geometric surface areas and pore volumes calculated from the crystallographic coordinates. The pore size distribution of MUF-16 also was calculated, which is consistent with the pore dimensions observed by SCXRD (Supplementary Fig. S12).

Gas adsorption measurements

CO2 isotherms were collected at 293 K and up to 1 bar (Fig. 2a and see Supplementary Fig. S11 for other temperatures). Both MUF-16 and MUF-16(Ni) take up 2.13 mmol/g (48 cm3/g) at 1 bar, and MUF-16(Mn) adsorbs 2.25 mmol/g (50.5 cm3/g). This equates to approximately 0.9 molecules of CO2 per metal site (Supplementary Table 5). CO2 uptake is only marginally higher at 273 K (Supplementary Fig. S11). The isosteric heat of adsorption (Qst) at zero-coverage was calculated to be 32 kJ/mol for MUF-16 and 37 kJ/mol for its Ni and Mn analogues (Fig. 2b). The Qst increases at higher loadings, which can be attributed to attractive intermolecular interactions when the CO2 loading levels are high, which enhance the framework-CO2 affinity. These interactions were experimentally verified by SCXRD (vide infra). The moderate Qst values, even at high CO2 loading30, are well below values observed for MOFs with open metal sites31. It follows that the energy required to regenerate the frameworks by CO2 desorption is likely to be low.

Fig. 2: CO2 adsorption on MUF-16 materials.

a Volumetric adsorption (filled circles) and desorption (open circles) isotherms of CO2 at 293 K and for MUF-16 (black), MUF-16(Mn) (red), and MUF-16(Ni) (blue). b Heats of adsorption (Qst) calculated for CO2 binding to MUF-16 (black), MUF-16(Mn) (red), and MUF-16(Ni) (blue) as a function of CO2 uptake. A high affinity for CO2 coupled to a moderate heat of adsorption promise an adsorbent that takes up significant quantities of gas yet is easily recycled. Source data are provided as a Source Data file.

Single-crystal X-ray diffraction was used to identify the CO2 binding sites in these frameworks32,33. MUF-16(Mn) was selected for this study since its darker colour streamlined crystal handling (the pale colour of the Co(II) and Ni(II) analogues make them difficult to see when loaded in a glass capillary). The results obtained for MUF-16(Mn) are directly applicable to MUF-16 and MUF-16(Ni) due to their identical structures and CO2 adsorption profiles (Fig. 2a and Supplementary Fig. S5). After transferring a MUF-16(Mn) single crystal into a capillary, it was activated in vacuo and the capillary flame-sealed. This allowed the guest-free structure of MUF-16(Mn) to be determined crystallographically (Supplementary Table 2). We then filled CO2 into the capillary to a pressure of 1.1 bar to determine the structure of the CO2-loaded framework. We noted only minor changes to the framework itself upon evacuation and filling with CO2. A clear picture of the affinity of MUF-16 for CO2 arises from the CO2-loaded SCXRD structure. First, the dimensions of the framework pores are well matched to the size of the CO2 molecules. This allows the guests to be enveloped by multiple non-covalent contacts (Fig. 3a). Second, these contacts are favourable since the electric quadrupole of the CO2 is complementary to the polarisation of the MUF-16 pore surface. For example, one of the electronegative oxygen atoms of each CO2 molecule engages in N-H···O and C-H···O hydrogen bonds with framework amino and phenyl groups at distances of 2.55, 2.81, and 2.87 Å. The electropositive carbon atom of each CO2 molecule engages in close-range contacts with the oxygen atoms of two non-coordinated carboxyl groups (2.87 and 3.04 Å). Two sites, which are related by crystallographic symmetry and share a common location for one of the oxygen atoms, are available to the CO2 guests. They are occupied with a 50/50 ratio and refinement of the CO2 occupancies gave 0.77 CO2 molecules per Mn centre, which agrees with the adsorption isotherm (Supplementary Table 5) allowing for uncertainties in the exact CO2 pressure in the X-ray capillaries. The CO2 guest molecules are aligned along the channels and tilted with respect to the pore axis (Fig. 3b). Attractive C···O intermolecular interactions between adjacent molecules are evident at a distance of 3.78 Å. This array of CO2 guests probably underlies the observed increase in Qst as a function of gas loading observed in the adsorption isotherms. A computational DFT model agrees with the SCXRD structure (Supplementary Fig. S60).

Fig. 3: CO2 capture by MUF-16.

a The adsorption sites of CO2 molecules in the pores of MUF-16(Mn), as determined by single-crystal X-ray diffraction. The CO2 is depicted in space-filling mode. Key intermolecular distances between MUF-16(Mn) and the adsorbed CO2 are shown with dashed orange lines. A second, symmetry-equivalent CO2 adsorption site exists. b Adsorbed CO2 molecules in MUF-16(Mn) highlighting the arrangement of adsorbed CO2 in the framework channels and potential attractive noncovalent interactions between adjacent guests. The CO2 molecules are shown in representative orientations in one of two symmetry-related crystallographic orientations. Colour code: manganese = lilac; nitrogen = blue; oxygen = red; carbon = grey; hydrogen = pale pink or white; pore Connolly surface = orange.

The strong adsorption of nitrous oxide, N2O, by MUF-16 corroborates this model of CO2 binding. The size and electrostatic distribution of N2O closely match those of CO2 (Supplementary Fig. S9). In parallel with CO2, N2O possesses atoms with partial negative charges at its termini that can bind to positively-charged regions of the pore surface, and vice-versa for its central nitrogen atom. MUF-16 adsorbs 1.91 mmol/g (43 cm3/g) of N2O at 1 bar and 293 K, which is only slightly less than the uptake of CO2.

The high uptake of CO2 by MUF-16 contrasts with its low affinity for hydrocarbons. Adsorption isotherms of CH4, C2H2, C2H4, C2H6, C3H6 and C3H8 were measured on MUF-16 at 293 K (Fig. 4a and Table 1). MUF-16 takes up just 1.20 cm3/g of CH4 at 1 bar and 293 K and 3.99 cm3/g of C2H2. The highest adsorption amount was 5.35 cm3/g observed for C3H6. Since only modest quantities of these gases are adsorbed, care was taken to ensure the accuracy of these measurements by using large sample quantities. The Qst values for the hydrocarbon gases are much lower than for CO2 (Supplementary Table 6). The water vapour adsorption isotherm of MUF-16 was measured at 298 K, showing the steady uptake of water until saturation is reached at around two molecules per Co centre (Supplementary Fig. S13b). The isotherm is fully reversible indicating that the adsorbed water is easily removed without perturbation of the framework.

Fig. 4: Gas uptake and calculated separation by MUF-16.

a Experimental CH4, C2H2, C2H4, C2H6, C3H6 and C3H8 adsorption (solid spheres) and desorption (open spheres) isotherms of MUF-16 measured at 293 K. b Predicted IAST selectivities, displayed with a log scale, of MUF-16 for various gas mixtures at 293 K. Source data are provided as a Source Data file.

Table 1 Summary of gas adsorption data and IAST-calculated selectivities for the MUF-16 family at 1 bar and 293 K.

Uptake ratios provide a useful indication of the preference of an adsorbent for certain gases over others. For MUF-16, the CO2/CH4 uptake ratio is 39.8 (293 K and 1 bar). This is comparable to [Cd2L(H2O)] (42.9)34 and exceeded by only one other reported material (SIFSIX-14-Cu-i, 85) (Supplementary Table 10)35. Typical physisorbents show a preference for unsaturated hydrocarbons over CO2, especially when bonding between the guest’s π electrons and open metal sites can occur25,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50. However, MUF-16 exhibits a uniform preference for CO2 over all C2 and C3 hydrocarbons at 293 K and 1 bar (Table 1). Here, the uptake ratios fall between 12 (acetylene), 15.6 (ethane) and 8.9 (propene). While the limited uptake of CH4 is a well-established function of its small size and low polarizability, the low affinity of MUF-16 for larger and more polar/polarizable hydrocarbon guests is notable. Inverted selectivity of this kind, that is, a preference for CO2 over small hydrocarbons, is a sought after yet seldom reported phenomenon25,51,52,53,54,55,56,57. With an uptake ratio of 12, MUF-16 surpasses previously reported materials that preferentially adsorb CO2 over C2H2, including SIFSIX-3-Ni (1.2 at 298 K and 0.1 bar)25, CD-MOF-2 (1.3 at 298 K and 1 bar)51, K2[Cr3O(OOCH)6(4‐ethylpyridine)3]2[α‐SiW12O40] (4.5 at 278 K and 1 bar)55, [Mn(bdc)(dpe)] (6.4 at 273 K and 1 bar)52 and [Tm2(OH-bdc)23- OH)2(H2O)2]58 (2.8 at 298 K and 1 bar) (Supplementary Table 11). The diminished affinity of MUF-16 for C2H2 results from the reversed quadrupole moment of this guest vis-à-vis CO2 (Supplementary Fig. S10). Since C2H2 is polarised oppositely to CO2 it is electrostatically repelled by the functional groups that line binding pockets in MUF-16. The upshot is inverse selectivity for CO2 over acetylene.

Separations using MUF-16

Building on the preferential affinity indicated by the uptake ratios, we quantified the selectivity of MUF-16 by Ideal Adsorbed Solution Theory (IAST) calculations59. At 293 K and 1 bar, the IAST selectivity of MUF-16 for CO2 over CH4 (50/50 mixture) is 6690 (Fig. 4b). MUF-16 is thus the best physisorbent known for this separation that does not operate by molecular sieving (Fig. 5 and Supplementary Table 10). For equimolar mixtures of CO2 and C2H2, C2H4, C2H6, C3H6 or C3H8 the selectivity of MUF-16 is also high (Table 1). With a selectivity of 510, MUF-16 is elevated well beyond other materials for the capture of CO2 from CO2/C2H2 (50/50) mixtures (Fig. 5 and Supplementary Table 11). As recognised in the literature for related systems17,18,60, these high selectivities emerge by suppressing the uptake of the hydrocarbon gases while maintaining proficient CO2 capture.

Fig. 5: Separation performance of MUF-16 compared to top-performing materials.

IAST selectivity of MUF-16 in comparison to a selection of physisorbents for CO2/CH4 (50/50) and CO2/C2H2 (50/50) mixtures at ambient temperature and 1 bar (see Supplementary Table 11 for details). For clarity, the y axis is broken in two parts with different scales.

While the pore characteristics of MUF-16 clearly favour the uptake of CO2 over other gases, its affinity could potentially rely on molecular sieving if the larger adsorbates are excluded from the framework on the basis of their size. This was ruled out by measuring hydrocarbon adsorption isotherms at 195 K, which showed that MUF-16 can adsorb CH4, C2H2 and C2H6 (Supplementary Fig. S15). Guest molecules of this size can freely enter the pore network of MUF-16 at this low temperature. However, since uptake is low at ambient temperatures interactions of these gases with the framework must be weak. Further, the kinetics of adsorption of several guest molecules were measured (Supplementary Fig. S16). All gases display a similar kinetic profile and reach their equilibrium uptake in well under one minute. Therefore, thermodynamic—rather than kinetic—effects have the most decisive impact on the differential affinity of these gases for MUF-16. We also considered whether a structural change of the framework might underly the gas selectivity, as observed for related systems52. However, XRD measurements show that the framework structure is largely conserved around room temperature in vacuo, in air and under CO2 (Supplementary Fig. S6). The CO2 adsorption isotherms at elevated temperatures show no sign of flexibility or gate opening (Supplementary Fig. S11), nor does the CH4 isotherm at high pressure (Figure S15). In the specific case of C2H4 at 195 K, there is evidence of modest gate opening, which will be fully evaluated in future work (Figure S15).

Invigorated by these results, we then investigated the feasibility of CO2/hydrocarbon separations under dynamic conditions. Experimental breakthrough curves were measured for various gas mixtures at 293 K and 1.1 bar: CO2/C2H6 (50/50), CO2/C2H4 (50/50), CO2/C2H2 (50/50 and 5/95) and CO2/CH4 (50/50 and 15/85) (Fig. 6a, b; Supplementary Figs. S44 and S51). Figure 6a, b shows the dimensionless concentration of CO2 and the hydrocarbons (measured independently) exiting an adsorbent bed packed with MUF-16 (0.9 gram) as a function of time.

Fig. 6: Gas separation by MUF-16.

a Experimental breakthrough curves for 50/50 mixtures of CO2 and the three C2 hydrocarbons (measured independently) at 293 K and 1.1 bar in an adsorption column packed with MUF-16. b Experimental breakthrough curves for 50/50 mixtures of CO2 and CH4 at 293 K and 1.1 bar in an adsorption column packed with MUF-16. c Twelve separation cycles for a CO2/C2H2 mixture (50/50 mixture). Each separation process was carried out at 293 K and 1.1 bar. MUF-16 was regenerated between cycles by placing it under vacuum at ambient temperature for 20–25 min. d Experimental desorption profile of MUF-16 following the separation of CO2 and C2H2 upon heating under a helium flow of 5 mlN/min at 1.1 bar. No adsorbates were removed upon further heating at 130 °C indicating that they had been fully expelled at lower temperatures. e Experimental breakthrough curves for a 15/80/4/1 CO2/CH4/C2H6/C3H8 mixture at 1.1 bar and 293 K in an adsorption column packed with MUF-16. f CO2 adsorption isotherms (293 K) of as-synthesized MUF-16 after four consecutive adsorption-desorption cycles, after exposing it to air with ~80% humidity for 12 months, and after immersion in water for 48 hours. Source data are provided as a Source Data file.

Complete separation was realised by MUF-16, whereby the hydrocarbons broke through from the column at an early stage because of their low affinity for the framework. Conversely, the signal of CO2 was not detected for at least 10 minutes due to its adsorption by MUF-16. The dynamic adsorption capacity for CO2 fell in the range 1.2–1.5 mmol/g which is nearly identical to the equilibrium capacity at the relevant partial pressures of CO2 (Supplementary Table 7). Significant volumes of pure hydrocarbons can be obtained in this way. Productivity calculations showed 1 kg of MUF-16 produces 27 L of the hydrocarbons from an equimolar mixture with CO2 at 293 K and 1 bar. The ability of MUF-16 to selectively adsorb CO2 is an important advantage of this MOF as pure hydrocarbons can be produced directly in a single adsorption stage. In literature reports to date, the capture of CO2 over C2 hydrocarbons has so far largely been restricted to cryogenic temperatures and/or static conditions52,53,54,55,57,61. With respect to CO2/C2H2 mixtures at ambient temperatures, we are aware of only a few reported materials, CD-MOF-151, CD-MOF-251 SIFSIX-3-Ni25, and [Tm2(OH-bdc)23- OH)2(H2O)2]58 for which this inverse trapping of CO2 has been verified by experimental breakthrough measurements. Since these MOFs adsorb C2H2 (in addition to CO2) strongly at moderate pressures, their uptake ratios are modest. They are limited to very low partial pressures of CO2 and suffer from low productivity.

Subsequent tests revealed that MUF-16 maintains its CO2 uptake and the complete removal of CO2 over at least 12 separation cycles (Fig. 6c). MUF-16 was regenerated between cycles by placing it under vacuum or by purging with an inert gas (Fig. 6d). Virtually all of the adsorbed acetylene and around half of the CO2 can be removed from the bed by purging at room temperature. The remainder can be fully desorbed at 80 °C.

To investigate separations involving trace CO2, we simulated breakthrough curves of feed gases with low CO2 partial pressures. First, a mass transfer coefficient was empirically determined based on measured breakthrough results to produce a match between simulated and experimental breakthrough curves26,62. With this realistic mass transfer coefficient in hand, we predicted breakthrough curves using feeds containing 0.1% CO2 in C2H2 (Supplementary Fig. S57). These calculations revealed that MUF-16 can eliminate trace quantities of CO2, as often required in industrial processes.

We then turned our attention to the separation of more complex gas mixtures. MUF-16 captures the CO2 from CO2/CH4/C2H6/C3H8 (15/80/4/1) feed mixtures at 1.1 bar. Here, we observed CH4, C2H6 and C3H8 to break through quickly with steep elution profiles (Fig. 6e). Crucially, the larger C2H6 and C3H8 components do not diminish the CO2 capture capabilities of MUF-16. This is an important observation for the removal of CO2 from natural gas, where mixed-gas separations involving these hydrocarbons are often required yet the pool of competent materials is limited19,63. To further probe the applicability of MUF-16 to natural gas sweeting, we conducted breakthrough measurements at a higher pressure of 9 bar. CO2 was cleanly removed from the gas stream (Supplementary Figs. S45 and S46). Breakthrough simulations at pressures relevant to natural gas processing (50 bar) lead to the prediction that MUF-16 can capture CO2 from natural gas (Supplementary Fig. S50). Water vapour is a component of crude natural gas streams and it can affect gas adsorption by physisorbents64,65. To test the moisture resistance of MUF-16, we measured its CO2 adsorption properties after exposure to air and immersion in water (Fig. 6f). The framework retains its CO2 adsorption capacity following these mistreatments. More detailed analysis, including the impact of water vapour on gas separation and the resistance of MUF-16 to other common natural gas impurities such as H2S, is an important next step.

In summary, the pores in MUF-16 are complementary to CO2 in size and electrostatic potential. This allows H-bonding and other noncovalent interactions to trap the guest CO2. Other guests, specifically methane and the C2 hydrocarbons, do not bind efficiently. This arises from the reversed polarity of these guests with respect to CO2 and results in a strong preference for CO2 over methane and inverted selectivity for CO2 over C2 and C3 hydrocarbon guests. MUF-16 shows exceptional performance for CO2/CH4 and CO2/C2H2 separations across a range of CO2/hydrocarbon compositions and pressures. These observations are relevant to the practical challenges of purifying natural gas and industrial feedstocks. MUF-16 has the potential to be produced economically on large scales and its chemical stability and recyclability meet the demands of a long-lived physisorbent. Given these characteristics, MUF-16 is a promising physisorbent for the capture of CO2.

Data availability

Source data are provided with this paper. Crystallographic data and files of MUF-16 as synthesized, under vacuum and loaded with CO2 have been deposited (CCDC 1948901 - 1948905). Additional graphics, TG curves, PXRD diffractograms, multiple cycle adsorption isotherms, dual site Langmuir isotherm model fitting, isosteric heat of adsorption calculations, BET surface area calculations, IAST calculations of adsorption selectivities, breakthrough curves simulations and models used and column breakthrough test setup with procedures and measurements, and the DFT results are available as Supplementary Information.

Further data that support the findings of this study are available from the corresponding author upon reasonable request. Source data are provided with this paper.


  1. 1.

    Sholl, D. S. & Lively, R. P. Seven chemical separations: to change the world: purifying mixtures without using heat would lower global energy use, emissions and pollution–and open up new routes to resources. Nature 532, 435–438. (2016).

    ADS  PubMed  Article  Google Scholar 

  2. 2.

    Ravanchi, M. T. & Sahebdelfar, S. Carbon dioxide capture and utilization in petrochemical industry: potentials and challenges. Appl. Petrochemical Res. 4, 63–77 (2014).

    Article  CAS  Google Scholar 

  3. 3.

    Rochelle, G. T. Amine scrubbing for CO2 capture. Science 325, 1652–1654 (2009).

    ADS  CAS  PubMed  Article  Google Scholar 

  4. 4.

    Yu, C.-H., Huang, C.-H. & Tan, C.-S. A review of CO2 capture by absorption and adsorption. Aerosol Air Qual. Res. 12, 745–769 (2012).

    CAS  Article  Google Scholar 

  5. 5.

    Kohl, A. & Nielsen, R. Gas Purification. 5th ed. (Gulf Publishing Company, Houston, 1997).

  6. 6.

    Oschatz, M. & Antonietti, M. A search for selectivity to enable CO2 capture with porous adsorbents. Energ. Environ. Sci. 11, 57–70 (2018).

    CAS  Article  Google Scholar 

  7. 7.

    Sreenivasulu, B., Sreedhar, I., Suresh, P. & Raghavan, K. V. Development trends in porous adsorbents for carbon capture. Environ. Sci. Technol. 49, 12641–12661 (2015).

    ADS  CAS  PubMed  Article  Google Scholar 

  8. 8.

    Yang, R. T. Gas separation by adsorption processes. (Butterworth-Heinemann, 2013).

  9. 9.

    Lu, A.-H. & Dai, S. Porous materials for carbon dioxide capture. (Springer, 2014).

  10. 10.

    Lin, R.-B., Xiang, S., Xing, H., Zhou, W. & Chen, B. Exploration of porous metal–organic frameworks for gas separation and purification. Coord. Chem. Rev. 378, 87–103 (2019).

    CAS  Article  Google Scholar 

  11. 11.

    Li, H. et al. Porous metal-organic frameworks for gas storage and separation: status and challenges. EnergyChem 1, 100006 (2019).

    Article  Google Scholar 

  12. 12.

    Ding, M., Flaig, R. W., Jiang, H.-L. & Yaghi, O. M. Carbon capture and conversion using metal–organic frameworks and MOF-based materials. Chem. Soc. Rev. 48, 2783–2828 (2019).

    CAS  PubMed  Article  Google Scholar 

  13. 13.

    Mukherjee, S., Kumar, A. & Zaworotko, M. J. 2 - Metal-organic framework based carbon capture and purification technologies for clean environment. In Metal-Organic Frameworks (MOFs) for Environmental Applications (Ed. Ghosh, S. K.) 5–61 (Elsevier, 2019).

  14. 14.

    Qazvini, O. T. & Telfer, S. G. A robust metal–organic framework for post-combustion carbon dioxide capture. J. Mater. Chem. A 8, 12028–12034 (2020).

    CAS  Article  Google Scholar 

  15. 15.

    Qazvini, O. T., Macreadie, L. K. & Telfer, S. G. Effect of ligand functionalization on the separation of small hydrocarbons and CO2 by a series of MUF-15 analogues. Chem. Mater. 32, 6744–6752 (2020).

    CAS  Article  Google Scholar 

  16. 16.

    Patil, K. M., Telfer, S. G., Moratti, S. C., Qazvini, O. T. & Hanton, L. R. Non-interpenetrated Cu-based MOF constructed from a rediscovered tetrahedral ligand. CrystEngComm 19, 7236–7243 (2017).

    CAS  Article  Google Scholar 

  17. 17.

    Balashankar, V. S., Rajagopalan, A. K., Pauw, R. D., Avila, A. M. & Rajendran, A. Analysis of a batch adsorber analogue for rapid screening of adsorbents for postcombustion CO2 capture. Ind. Eng. Chem. Res. 58, 3314–3328 (2019).

    Article  CAS  Google Scholar 

  18. 18.

    Rajagopalan, A. K., Avila, A. M. & Rajendran, A. Do adsorbent screening metrics predict process performance? A process optimisation based study for post-combustion capture of CO2. Int. J. Greenh. Gas. Con. 46, 76–85 (2016).

    CAS  Article  Google Scholar 

  19. 19.

    Rufford, T. E. et al. The removal of CO2 and N2 from natural gas: a review of conventional and emerging process technologies. J. Petrol. Sci. Eng. 94, 123–154 (2012).

    Article  CAS  Google Scholar 

  20. 20.

    Madden, D. G. et al. Highly selective CO2 removal for one-step liquefied natural gas processing by physisorbents. Chem. Commun. 55, 3219–3222 (2019).

    CAS  Article  Google Scholar 

  21. 21.

    Belmabkhout, Y. et al. Natural gas upgrading using a fluorinated MOF with tuned H2S and CO2 adsorption selectivity. Nat. Energy 3, 1059–1066 (2018).

    ADS  CAS  Article  Google Scholar 

  22. 22.

    Matar, S. & Hatch, L. F. Chemistry of Petrochemical Processes. (Gulf Professional Publishing, 2001).

  23. 23.

    Hort, E. V. & Taylor, P. Acetylene‐Derived Chemicals. Kirk‐Othmer Encyclopedia of Chemical Technology (2000).

  24. 24.

    Pässler, P. et al. Acetylene. Ullmann’s Encycl. Ind. Chem. 1, 177–227 (2008).

    Google Scholar 

  25. 25.

    Chen, K.-J. et al. Benchmark C2H2/CO2 and CO2/C2H2 separation by two closely related hybrid ultramicroporous materials. Chem 1, 753–765 (2016).

    CAS  Article  Google Scholar 

  26. 26.

    Qazvini, O. T. et al. Ethane-trapping metal–organic framework with a high capacity for ethylene purification. J. Am. Chem. Soc. 141, 5014–5020 (2019).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  27. 27.

    Qazvini, O. T., Babarao, R. & Telfer, S. G. Multipurpose metal–organic framework for the adsorption of acetylene: ethylene purification and carbon dioxide removal. Chem. Mater. 31, 4919–4926 (2019).

    CAS  Article  Google Scholar 

  28. 28.

    Tang, E. et al. Two cobalt(II) 5-aminoisophthalate complexes and their stable supramolecular microporous frameworks. Inorg. Chem. 45, 6276–6281 (2006).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  29. 29.

    Tian, C.-B. et al. Four new MnII inorganic–organic hybrid frameworks with diverse inorganic magnetic chain’s sequences: syntheses, structures, magnetic, NLO, and dielectric properties. Inorg. Chem. 54, 2560–2571 (2015).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  30. 30.

    Zhai, Q.-G., Bu, X., Zhao, X., Li, D.-S. & Feng, P. Pore space partition in metal–organic frameworks. Acc. Chem. Res. 50, 407–417 (2017).

    CAS  PubMed  Article  Google Scholar 

  31. 31.

    Sumida, K. et al. Carbon dioxide capture in metal–organic frameworks. Chem. Rev. 112, 724–781 (2012).

    CAS  PubMed  Article  Google Scholar 

  32. 32.

    Wong-Ng, W. In situ diffraction studies of selected metal–organic framework materials for guest capture. In Materials and Processes for CO2 Capture, Conversion, and Sequestration (Eds. Li, L., Wong-Ng, W., Huang, K. & Cook, L. P.) (2018).

  33. 33.

    Easun, T. L., Moreau, F., Yan, Y., Yang, S. & Schröder, M. Structural and dynamic studies of substrate binding in porous metal–organic frameworks. Chem. Soc. Rev. 46, 239–274 (2017).

    CAS  PubMed  Article  Google Scholar 

  34. 34.

    Hou, L. et al. A rod packing microporous metal–organic framework: unprecedented ukv topology, high sorption selectivity and affinity for CO2. Chem. Commun. 47, 5464–5466. (2011).

    CAS  Article  Google Scholar 

  35. 35.

    Jiang, M. et al. Controlling pore shape and size of interpenetrated anion-pillared ultramicroporous materials enables molecular sieving of CO2 combined with ultrahigh uptake capacity. ACS Appl. Mater. Interfaces 10, 16628–16635 (2018).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  36. 36.

    Moreau, F. et al. Unravelling exceptional acetylene and carbon dioxide adsorption within a tetra-amide functionalized metal-organic framework. Nat. Commun. 8, 14085 (2017).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  37. 37.

    Xiang, S., Zhou, W., Gallegos, J. M., Liu, Y. & Chen, B. Exceptionally high acetylene uptake in a microporous metal−organic framework with open metal sites. J. Am. Chem. Soc. 131, 12415–12419 (2009).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

  38. 38.

    Li, P. et al. A rod-packing microporous hydrogen-bonded organic framework for highly selective separation of C2H2/CO2 at room temperature. Angew. Chem., Int. Ed. 54, 574–577 (2015).

    CAS  Google Scholar 

  39. 39.

    Lee, J. et al. Separation of acetylene from carbon dioxide and ethylene by a water-stable microporous metal–organic framework with aligned imidazolium groups inside the channels. Angew. Chem. Int. Ed. 57, 7869–7873 (2018).

    CAS  Article  Google Scholar 

  40. 40.

    Luo, F. et al. UTSA-74: A MOF-74 isomer with two accessible binding sites per metal center for highly selective gas separation. J. Am. Chem. Soc. 138, 5678–5684 (2016).

    CAS  PubMed  Article  Google Scholar 

  41. 41.

    Zhang, J.-P. & Chen, X.-M. Optimized acetylene/carbon dioxide sorption in a dynamic porous crystal. J. Am. Chem. Soc. 131, 5516–5521 (2009).

    CAS  PubMed  Article  Google Scholar 

  42. 42.

    Zhang, L. et al. Efficient separation of C2H2 from C2H2/CO2 mixtures in an acid–base resistant metal–organic framework. Chem. Commun. 54, 4846–4849 (2018).

    CAS  Article  Google Scholar 

  43. 43.

    Peng, Y.-L. et al. Robust ultramicroporous metal–organic frameworks with benchmark affinity for acetylene. Angew. Chem., Int. Ed. 57, 10971–10975 (2018).

    CAS  Article  Google Scholar 

  44. 44.

    Scott, H. S. et al. Highly selective separation of C2H2 from CO2 by a new dichromate-based hybrid ultramicroporous material. ACS Appl. Mater. Interfaces 9, 33395–33400 (2017).

    CAS  PubMed  Article  Google Scholar 

  45. 45.

    Matsuda, R. et al. Highly controlled acetylene accommodation in a metal–organic microporous material. Nature 436, 238 (2005).

    ADS  CAS  PubMed  Article  Google Scholar 

  46. 46.

    Duan, X. et al. A novel metal-organic framework for high storage and separation of acetylene at room temperature. J. Solid State Chem. 241, 152–156 (2016).

    ADS  CAS  Article  Google Scholar 

  47. 47.

    Duan, X. et al. A new metal–organic framework with potential for adsorptive separation of methane from carbon dioxide, acetylene, ethylene, and ethane established by simulated breakthrough experiments. J. Mat. Chem. A 2, 2628–2633 (2014).

    CAS  Article  Google Scholar 

  48. 48.

    Gao, J. et al. Mixed metal–organic framework with multiple binding sites for efficient C2H2/CO2 separation. Angew. Chem. Int. Ed. 59, 4396–4400 (2020).

    CAS  Article  Google Scholar 

  49. 49.

    Ye, Y. et al. Pore space partition within a metal–organic framework for highly efficient C2H2/CO2 separation. J. Am. Chem. Soc. 141, 4130–4136 (2019).

    CAS  PubMed  Article  Google Scholar 

  50. 50.

    Fan, W. et al. Optimizing multivariate metal–organic frameworks for efficient C2H2/CO2 separation. J. Am. Chem. Soc. 142, 8728–8737 (2020).

    PubMed  Article  CAS  PubMed Central  Google Scholar 

  51. 51.

    Li, L. et al. Inverse adsorption separation of CO2/C2H2 mixture in cyclodextrin-based metal–organic frameworks. ACS Appl. Mater. Interfaces 11, 2543–2550 (2019).

    CAS  PubMed  Article  Google Scholar 

  52. 52.

    Foo, M. L. et al. An adsorbate discriminatory gate effect in a flexible porous coordination polymer for selective adsorption of CO2 over C2H2. J. Am. Chem. Soc. 138, 3022–3030 (2016).

    CAS  PubMed  Article  Google Scholar 

  53. 53.

    Duan, J. et al. A family of rare Earth porous coordination polymers with different flexibility for CO2/C2H4 and CO2/C2H6 separation. Inorg. Chem. 52, 8244–8249 (2013).

    CAS  PubMed  Article  Google Scholar 

  54. 54.

    Yang, W. et al. Selective CO2 uptake and inverse CO2/C2H2 selectivity in a dynamic bifunctional metal–organic framework. Chem. Sci. 3, 2993–2999 (2012).

    CAS  Article  Google Scholar 

  55. 55.

    Eguchi, R., Uchida, S. & Mizuno, N. Inverse and high CO2/C2H2 sorption selectivity in flexible organic–inorganic ionic crystals. Angew. Chem., Int. Ed. 51, 1635–1639 (2012).

    CAS  Article  Google Scholar 

  56. 56.

    Horike, S. et al. Dense coordination network capable of selective CO2 capture from C1 and C2 hydrocarbons. J. Am. Chem. Soc. 134, 9852–9855 (2012).

    CAS  PubMed  Article  Google Scholar 

  57. 57.

    Noro, S.-i. et al. Selective gas adsorption in one-dimensional, flexible CuII coordination polymers with polar units. Chem. Mater. 21, 3346–3355 (2009).

    CAS  Article  Google Scholar 

  58. 58.

    Ma, D. et al. Inverse and highly selective separation of CO2/C2H2 on a thulium–organic framework. J. Mater. Chem. A 8, 11933–11937 (2020).

    CAS  Article  Google Scholar 

  59. 59.

    Myers, A. & Prausnitz, J. M. Thermodynamics of mixed‐gas adsorption. AIChE J. 11, 121–127 (1965).

    CAS  Article  Google Scholar 

  60. 60.

    Subraveti, S. G. et al. Cycle design and optimization of pressure swing adsorption cycles for pre-combustion CO2 capture. Appl. Energy 254, 113624 (2019).

    CAS  Article  Google Scholar 

  61. 61.

    Yanai, N. et al. Gas detection by structural variations of fluorescent guest molecules in a flexible porous coordination polymer. Nat. Mater. 10, 787 (2011).

    ADS  CAS  PubMed  Article  PubMed Central  Google Scholar 

  62. 62.

    Qazvini, O. T. & Fatemi, S. Modeling and simulation pressure–temperature swing adsorption process to remove mercaptan from humid natural gas; a commercial case study. Sep. Purif. Technol. 139, 88–103 (2015).

    CAS  Article  Google Scholar 

  63. 63.

    Krishna, R. Metrics for evaluation and screening of metal–organic frameworks for applications in mixture separations. ACS Omega 5, 16987–17004 (2020).

    CAS  PubMed  PubMed Central  Article  Google Scholar 

  64. 64.

    Mukherjee, S. et al. Trace CO2 capture by an ultramicroporous physisorbent with low water affinity. Sci. Adv. 5, eaax9171 (2019).

    ADS  CAS  PubMed  PubMed Central  Article  Google Scholar 

  65. 65.

    Masala, A. et al. CO2 capture in dry and wet conditions in UTSA-16 metal–organic framework. ACS Appl. Mater. Interfaces 9, 455–463 (2017).

    CAS  PubMed  Article  PubMed Central  Google Scholar 

Download references


We would like to thank Victoria-Jayne Reid for technical assistance, Seok June (Subo) Lee and Adil Alkas for useful discussions regarding X-ray crystallography, and Steve Denby for expert engineering support. We gratefully acknowledge the MacDiarmid Institute and RSNZ Marsden Fund (contract 14-MAU-024) for financial support. R.B. acknowledges the National Computing Infrastructure (NCI) for providing the supercomputing facility.

Author information




The manuscript was written through the contributions of O.T.Q. and S.G.T. who designed and performed the experiments, analysed the results and jointly wrote the paper. R.B. performed the DFT calculations. A patent on MUF-16 has been lodged (WO 2020/130856 A1).

Corresponding author

Correspondence to Shane G. Telfer.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Peer review information Nature Communications thanks Mario Wriedt and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

Source data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Qazvini, O.T., Babarao, R. & Telfer, S.G. Selective capture of carbon dioxide from hydrocarbons using a metal-organic framework. Nat Commun 12, 197 (2021).

Download citation

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.


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

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