SEARCH

SEARCH BY CITATION

Keywords:

  • adsorption;
  • CO2 capture;
  • metal–organic frameworks;
  • microporous materials;
  • synthetic methods

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusion
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

In metal–organic framework (MOF) chemistry, polar functionalities greatly affect the gas adsorption properties. However, synthesis of MOFs with desired functionality is very challenging because many chemical functionalities cannot be achieved under the conditions for MOF assembly. Herein, a facile synthesis of new functionalized two-dimensional MOFs with preferential CO2 capture is presented, which uses two successive synthesis steps: 1) rational design and template-free synthesis of the parent MOF with designated pendant amino groups and 2) postsynthetic modification of the active amino groups with acetic acid and trimesoyl chloride functionalities. The only variation in structure arises from the functional groups of these materials. Experimental results demonstrate that the three 2D layered MOFs have remarkable thermal stability and moisture resistance, which are particularly advantageous for practical CO2 capture. Although their surface areas are moderate (270–340 m2 g−1), they still have excellent CO2 adsorption capacity (up to 2.9 mmol g−1 at 1 bar and 273 K) comparable to that of previously reported MOFs with much higher surface areas. Based on first-principles calculations, it is shown that the acidic carbonyl functionalities in addition to the amino groups are also favorable to bind CO2 molecules. The adsorption sites generated from polar functionalities are key factors leading to high CO2 uptake.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusion
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

Capture and separation of CO2 in an energy-efficient and economical manner have attracted tremendous attention because of the dual role of CO2 as a renewable carbon source and greenhouse gas.13 However, selective capture of CO2 from flue gas emissions still remains challenging.4 A series of technologies for CO2 capture including membrane separation, sorption by solvent, and adsorption onto solid have been widely explored.5 In particular, pressure-swing adsorption technology is an economical solution to reduce the high cost of CO2 capture from flue gases.6 To promote this technology, regenerable adsorbents with high adsorption capacity and rapid kinetics for CO2 capture and release need to be developed. In the past few decades, a series of CO2 adsorbents have been prepared,7, 8 including porous carbons,9 zeolites,10 amine-modified silicas,11 and new classes of porous materials.12, 13 Among them, metal–organic frameworks (MOFs) with high surface area, high thermal stability, and distinct porosity are emerging as one of the most promising materials.14 However, developing deliverable MOFs with high CO2 adsorption capacity and excellent moisture resistance is still ongoing.15

As we know, the capacity of gas adsorption in MOFs can be further enhanced by chemical functionalization16 of their pores with functional groups that can interact strongly with the gas molecules.17 Unfortunately, the synthesis of functionalized MOFs is rather difficult because many chemical functionalities cannot tolerate the harsh conditions for MOF assembly.1822 In this regard, developing new functionalized MOFs by a simple and efficient method is critically necessary and attractive. Inspired by the concept of “secondary seeded” synthesis of MOF membranes,23 we anticipate that MOFs with a variety of desired functionalities could be synthesized efficiently in a stepwise manner consisting of 1) rational design and synthesis of the parent MOF structure with designated active sites, which are able to interact with functionalities, and 2) postsynthetic modification of the active sites with desired functionalities.

Herein, we present the designed synthesis of three new functionalized MOFs with polar functionalities for CO2 capture by the combination of direct (template-free) synthesis and covalent modification. 5-Aminotetrazole, which has a strong tendency to form robust networks, was chosen as the organic linker.22 More importantly, its free amino groups are excellent for binding CO2 molecules and allow the as-synthesized frameworks to be decorated with a variety of functionalities.16 As shown in Scheme 1, the light yellow MOF crystal (1) was obtained by reaction of zinc nitrate hexahydrate and 5-aminotetrazole in N,N-dimethylformamide (DMF) solution with no structure-directing agent at 393 K for 72 h. Subsequently, compounds 2 and 3 were synthesized by the covalent modification of 1 with acetic acid (Ac) and trimesoyl chloride (TMC), respectively. The Ac and TMC were chosen as model reactants, which were expected to decorate 1 with polar functional groups that could improve the CO2 adsorption capacity.17

thumbnail image

Scheme 1. Synthesis of functionalized 2D MOFs used in this work.

Download figure to PowerPoint

Results and Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusion
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

Crystal structure of the as-prepared sample 1

The resulting compound 1 was structurally characterized and formulated by single-crystal X-ray diffraction (XRD) studies as [Zn2(CN5H2)3(H2O)3]6H2O. As shown in Figure 1, compound 1 crystallizes in the trigonal space group R3 with two zinc ions, one tetrazole ligand, and three water molecules included in an asymmetric unit to form an infinite two-dimensional (2D) layered framework. One of the zinc atoms is coordinated in a trigonal bipyramidal geometry with three nitrogen atoms from three different tetrazole ligands and two coordinated water molecules. Another zinc atom is tetrahedrally coordinated by three nitrogen atoms and one water molecule. Each organic ligand is linked with two zinc nodes and constructs 1D square channels along the b axis. The pendant amino groups are located around the cavities, thus providing strong binding sites for CO2 capture. The coordinated water molecules reside between the 2D layers as hydrogen-bonding donors and acceptors. The as-formed interlayer pores are available for accommodating guest molecules.

thumbnail image

Figure 1. The crystal structure of 1. a) 2D layer framework of 1 along the b-axis direction. b) The asymmetric unit of crystal 1; coordinated water molecules are omitted for clarity. c) The structure of each layer along the c axis. Color scheme: Zn, lilac; N, blue; C, gray; O, red; H, white.

Download figure to PowerPoint

Evolution of sample 1

Optical microscopy revealed that the as-synthesized single crystal 1 is well crystallized (Figure S1 in the Supporting Information). In addition, X-ray photoelectron spectroscopy (XPS; Figure S2) and simulated and powder XRD patterns of 1 (Figure 2) also confirmed that a pure phase of bulky sample was synthesized. The infrared (IR) analysis of 1 clearly showed double peaks between 3500 and 3100 cm−1, which were ascribed to the asymmetrical and symmetrical stretching vibration absorption of the amine groups, as shown in Figure 3.24 Moreover, the 1624 and 1338 cm−1 peaks correspond to the N[BOND]H bending vibration and the characteristic C[BOND]N stretching, respectively.25 Notably, the thermal stability of 1 is fairly high with a decomposition temperature above 593 K in air (Figure S3). There is approximately 20 % weight loss at 453 K, which is attributed to the removal of water or solvent molecules. As the coordinated water molecules reside between the 2D layers as hydrogen-bonding donors and acceptors, the loss of water after activation would cause a slight structural change to the frameworks as indicated by the powder XRD of activated and as-prepared samples (see Figure 2). Furthermore, the resulting crystal 1 displays an excellent water-resistance ability, as indicated by the unchanged powder XRD patterns after immersion in aqueous solution for 1, 7, and 14 days (see Figure S4). This is particularly advantageous for practical CO2 capture. Indeed, most of the previously developed MOFs are extremely moisture-sensitive, which limits their practical application because of an instability with respect to moisture, which results in the phase transformation and decomposition of the framework structure.26

thumbnail image

Figure 2. Powder XRD patterns of the simulated, as-synthesized, and activated sample 1.

Download figure to PowerPoint

thumbnail image

Figure 3. IR spectrum of the as-prepared sample 1.

Download figure to PowerPoint

To determine the pore structure and surface area of 1, the N2 adsorption isotherm was measured at 77.3 K, as shown in Figure S5. The completely reversible isotherms exhibited a Type I behavior. The isotherms showed a very sharp uptake at P/P0 from 10−5 to 10−1, which is a signature feature of microporous materials.27 We applied the Brunauer–Emmett–Teller (BET) model to the isotherm for P/P0 between 0.05 and 0.2 and obtained a surface area of 340.8 m2 g−1. The Langmuir surface area was calculated to be 433 m2 g−1. The Horvath–Kawazoe model was used to fit the adsorption isotherm to depict the pore structure of the sample, and the mean pore width was estimated to be 4.6 Å. Apparently, the high thermal and chemical stability, permanent microporosity, and pendant amino groups of 1 originate from the rigid MOF structure, which allows further decoration with a variety of functionalities.

Postsynthetic modification of sample 1

The first evidence for the successful postsynthetic modification reactions in Scheme 1 was obtained by IR spectroscopy. The IR analysis of the products 2 and 3 clearly showed peaks at around 1700 cm−1, which correspond to the C[DOUBLE BOND]O stretching vibration of the grafted carbonyl groups, and are displayed in Figure 4. In addition, the peaks between 900 and 650 cm−1 of 3 correspond to the characteristic CH units of the phenyl ring.

thumbnail image

Figure 4. IR spectral analysis of the as-prepared samples 13.

Download figure to PowerPoint

To further interrogate the modification process, all three samples were also analyzed by solid-state 13C NMR spectroscopy, displayed in Figure 5. Compared to the 13C NMR spectrum of compound 1, new signals were clearly observed in the spectrum of the products 2 and 3. The signals at around 160 and 21 ppm of 2 are caused by the presence of the acetamide functionality.28 The spectrum of 3 clearly exhibits signals at around 160 ppm, which is ascribed to the presence of the modified functionality. However, the signals between 110 and 140 ppm of the phenyl ring are not clearly observed. This probably suggests that a small amount of TMC reacted with the amino groups because of its large molecular size. Note that the crystalline structure of 1 was well retained in the modified products 2 and 3 after covalent modification (Figure S6). Furthermore, 2 and 3 showed slightly better thermal stability than 1, as shown in Figure 6. Based on the residual weights of the three samples, the reaction rates of amino groups with Ac (compound 2) and TMC (compound 3) were estimated to be 67 and 7 %, respectively. Thus, the weak peaks at around 1700 cm−1 (IR) and few signals of the phenyl ring (13C NMR) of 3 can be rationalized by the small amount of TMC that reacted with amino groups. On the other hand, the smaller residual weights of samples 2 and 3 also confirmed that the functional groups were grafted into the frameworks. However, the BET surface areas of 2 and 3 are decreased to 270.3 and 313.7 m2 g−1 (Figure S7), respectively. The surface area of compound 2 drops more significantly because Ac is smaller than TMC so that more Ac can be incorporated into the pores, thereby reacting with amino groups of 1.

thumbnail image

Figure 5. Solid-state 13C NMR spectra of samples 13. Circles and asterisks represent signals of 5-aminotetrazole and DMF, respectively. Squares and triangles represent signals of the modified functionality.

Download figure to PowerPoint

thumbnail image

Figure 6. Thermogravimetric analysis of activated samples of 13 in an air atmosphere.

Download figure to PowerPoint

Gas adsorption properties of as-prepared samples 1–3

Finally, we studied CO2, N2, and CH4 adsorption on samples 13. As shown in Figure 7, all three samples display higher adsorption uptakes for CO2 than N2 and CH4 at 273 K. This is attributed to the much higher quadrupole moment of CO2 than those of N2 and CH4.29 After all, the governing driving force for binding CO2 on these frameworks is the electrostatic interaction. To ascertain the impact of polar functionality, we calculated the electrostatic potential (ESP) of the organic linker of 1 and modified the organic linkers of 2 and 3 based on first-principles methods. Indeed, it has been shown that the topology of ESP is capable of connecting electronic structure and electrostatic reactivity.30 As displayed in Figure 8, the electron-withdrawing Ac and TMC functionalities alter the polarity of the parent frameworks31 and reduce the basicity of the amino groups. Nevertheless, both functionalities are able to interact with CO2 molecules, as evidenced by the unequally distributed ESP near the carbonyl groups. In other words, samples 2 and 3 provide extra adsorption sites to bind CO2 in addition to the amino groups. Therefore, the CO2 capture ability of 2 and 3 is still excellent even though they have significantly lower surface areas than 1 and many other MOFs. Indeed, the CO2 adsorption capacities of 13 did not scale with their BET surface areas. The measured CO2 uptakes of 2 and 3 were up to 2.8 and 2.9 mmol g−1 at 1 bar and 273 K, which are 15 and 20 % higher than that of 1, respectively. In contrast, the CH4 and N2 adsorption capacities of 2 and 3 are only slightly improved, which implies that the functionalization also improves the selectivity of CO2 over these two species.

thumbnail image

Figure 7. Gas uptakes for samples 1 (green), 2 (blue), and 3 (red) at 273 K. Squares, CO2; circles, CH4; triangles, N2.

Download figure to PowerPoint

thumbnail image

Figure 8. 2D contours of ESP of a) organic linker of 1 and b) modified organic linker of 2 and c) 3. The red and yellow solid lines represent negative (isovalue from −0.001 to −0.08 electrons bohr−3) and positive (isovalue from 0.08 to 0.001 electrons bohr−3) areas, respectively. In (c), the left and right parts display the 2D ESP contours projected on the surface of tetrazole and the benzene ring, respectively.

Download figure to PowerPoint

At high temperatures, the CO2 uptakes of the three samples drop inevitably, as shown in Figure S8. Nevertheless, 3 still displays a fairly good CO2 adsorption capacity of 1.9 mmol g−1 at 1 bar and 298 K (Figure 9). Compared to some previously reported MOFs,17, 32, 33 the as-prepared functionalized MOFs display preferable CO2 adsorption capacity, albeit they possess lower surface areas (Table S1 in the Supporting Information). Thus, adsorption sites generated from polar functionalities are key factors leading to high CO2 uptake.

thumbnail image

Figure 9. CO2 uptakes for sample 3 at different temperatures.

Download figure to PowerPoint

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusion
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

This study demonstrates that three new functionalized MOFs have been originally designed and synthesized in a tandem manner. The as-prepared 2D layered MOFs display high thermal stability and moisture resistance. Although their surface areas are moderate, they still have good CO2 adsorption capacity comparable to that of previously reported MOFs with much higher surface areas. Results show CO2 uptakes at 1 bar and 273 K ranging from 2.4 mmol g−1 in 1 to 2.9 mmol g−1 in 3. Moreover, 3 displays a fairly good CO2 adsorption capacity of 1.9 mmol g−1 at 1 bar and 298 K. Based on first-principles calculations, we show that the acidic carbonyl functionality in addition to the amino groups is also favorable to bind CO2 molecules. The adsorption sites generated from the polar functionality are key factors leading to high CO2 uptake.

Experimental Section

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusion
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

Template-free synthesis of crystal 1

A well-crystallized zeolitic tetrazolate framework (1) was achieved by reaction of Zn(NO3)26H2O (0.895 g, 0.2 M) and 5-aminotetrazole (0.51 g, 0.2 M) in an N,N-dimethylformamide (DMF, 30 mL) solution with no structure-directing agent at 393 K for 72 h. The light yellow crystals were clearly observed in the Teflon vessel. The as-prepared products were recovered by filtration, washed with DMF at room temperature two or three times every other day, and then the products were further washed with alcohol at room temperature two or three times to remove the solvent DMF. Finally, the as-synthesized products were dried at 80 °C for 12 h. The products were activated under vacuum at 180 °C for 12 h.

Covalent modification of 1

Compounds 2 and 3 were obtained by the covalent modification of 1 using acetic acid (Ac) and trimesoyl chloride (TMC), respectively. Compound 2 was achieved by reaction of Ac with 1 at 160 °C for 2 h, and then it was sufficiently washed with deionized water several times. Compound 3 was achieved by reaction of TMC with 1 in hexane solution under stirring for 60 min. The products were isolated by filtration, sufficiently washed with hexane solution several times, and then washed with deionized water. Before adsorption tests, both compounds 2 and 3 were activated under vacuum at 180 °C for 12 h.

Single-crystal X-ray crystallography

Single-crystal XRD analysis of 1 was performed on a Bruker SMART APEX CCD diffractometer with graphite-monochromated Mo (λ=0.71073 Å) radiation by using the SMART and SAINT programs. The structures were solved by direct methods and refined on F2 by full-matrix least-squares methods with SHELXTL version 5.1. CCDC-881046 http://www.ccdc.cam.ac.uk/cgi-bin/catreq.cgi contains the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif.

Characterization

For morphology determination, crystal 1 was examined by optical microscopy (MM50). Powder XRD patterns of the samples were recorded on a D8-Advance Bruker AXS diffractometer with Cu (λ=1.5418 Å) radiation at room temperature. Nitrogen adsorption/desorption isotherms were measured on ASAP 2020 M apparatus at 77.3 K, and then the BET surface area was calculated over the range of relative pressures between 0.05 and 0.20. The chemical composition and bonding states were measured by XPS by using a Kratos AXIS ULTRADLD instrument with a monochromic Al X-ray source (=1486.6 eV). IR spectra were recorded on KBr/NMOF pellets in a Thermo model Nicolet 6700 spectrometer. Room-temperature 1H[RIGHTWARDS ARROW]13C cross polarization/magic angle spinning NMR experiments were performed with a double-tuned 4.0 mm probe on a Bruker Avance III 400 spectrometer in a magnetic field strength of 9.4 T at Larmor frequencies of 400.13 MHz (1H) and 100.66 MHz (13C). Thermal stabilities of the samples were measured with a system provided by Mettler Toledo (model TGA/DSC1) in air at a heating rate of 5 °C min−1 up to 850 °C.

Computational methods

The ESP was calculated by Gaussian software. The 6-311G(d) basis set was employed for the C, H, O, and N atoms. Negative ESP corresponds to the attraction of protons by the concentrated electron density in the molecules (from lone pairs, pi bonds). Positive ESP corresponds to the repulsion of protons by the atomic nuclei in regions where low electron density exists and the nuclear charge is incompletely shielded.

Gas adsorption measurements

The adsorption isotherms of the probe gas CO2 (purity 99.999), N2 (purity 99.999), and CH4 (purity 99.99) were measured by the volumetric technique with an apparatus from SETARAM France (PCTpro-E&E). Before each measurement, the sample was evacuated at 180 °C for 12 h.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusion
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

We acknowledge financial support from the National Science Foundation of China (Grant No. 51072204, 51172249), Technology Innovative Research Program of Ningbo Municipality (Grant No. 2009B21005), Ningbo Science Foundation (Grant No. 2012A610167, 2012A610100), and the Science Technology Department of Zhejiang Province (Grant No. 2012C21108).

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results and Discussion
  5. Conclusion
  6. Experimental Section
  7. Acknowledgements
  8. Supporting Information

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
cplu_201200270_sm_miscellaneous_information.pdf340Kmiscellaneous_information

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.