Dynamic Spacer Installation for Multirole Metal–Organic Frameworks

Apr 7, 2017 - Dynamic Spacer Installation for Multirole Metal–Organic Frameworks: A New Direction toward Multifunctional MOFs Achieving Ultrahigh Me...
5 downloads 12 Views 1MB Size
Download PDF
Communication pubs.acs.org/JACS

Dynamic Spacer Installation for Multirole Metal−Organic Frameworks: A New Direction toward Multifunctional MOFs Achieving Ultrahigh Methane Storage Working Capacity Cheng-Xia Chen,† Zhang-Wen Wei,† Ji-Jun Jiang, Shao-Ping Zheng, Hai-Ping Wang, Qian-Feng Qiu, Chen-Chen Cao, Dieter Fenske, and Cheng-Yong Su* MOE Laboratory of Bioinorganic and Synthetic Chemistry, Lehn Institute of Functional Materials, School of Chemistry, Sun Yat-Sen University, Guangzhou 510275, China S Supporting Information *

oped a sequential linker installation strategy to precisely insert additional linkers for MTV-MOFs11 and further engineered the pore environment with up to three different functional groups.12 These excellent works suggest the possibility of creating multirole MOFs, even though uninstallation of multifunctional units was not achieved. The key point to design swing-role or multirole MOFs is to achieve facile functional switching, for example, through simple installation and uninstallation of different functional units (Figure 1). In a chemistry word, a reversible and dynamic chemical process is needed for such functional switching. Thus, a swing- or multirole MOF must have two vital features: (1) versatile functionalization with different modules and (2) easy transformation between different functional versions. If a basic framework can be functionalized and optimized for different purposes and reused after a simple function-change process without de novo synthesis, it could be applied for different missions in a swing- or multirole way, thereby reaching the targets of cost reduction and high performance. This type of swing-role MOF strategy has not been reported to date, although MTV-MOFs functionalized via irreversible linker installation11,12 and reversible incorporation of organic linkers into MOFs13 have been separately reported. The combination of these two important features in a single MOF is unprecedented. A prototypical MOF that is competent for reversible functionalization and defunctionalization is LIFM-28, which was previously reported to demonstrate a postsynthetic variablespacer installation (PVSI) strategy to predictably implement kinetic installation/uninstallation of additional spacers with different lengths and functionality.14 Only half of the replaceable binding sites are utilized to manipulate the breathing and adsorbing behaviors. Herein, we further explore the functionalization potential of this proto-Zr-MOF as a parent framework to achieve facile in situ multifunctional switching, i.e., a swing- or multirole strategy for economical MOF applications. Taking advantage of all of the replaceable binding sites allows the generation of a series of fully functionalized MOFs based on proto-LIFM-28 in which two types of additional spacers (shorter vs longer, as shown in Figure 2) are accurately inserted into two types of pockets. These fully functionalized MOFs, LIFM-70−

ABSTRACT: A robust Zr-MOF (LIFM-28) containing replaceable coordination sites for additional spacer installation has been employed to demonstrate a swingor multirole strategy for multifunctional MOFs. Through reversible installation/uninstallation of two types of spacers with different lengths and variable functional groups, different tasks can be accomplished using the same parent MOF. An orthogonal optimizing method is applied with seven shorter (L1−7) and six longer (L8−13) spacers to tune the functionalities, achieving multipurpose switches among gas separation, catalysis, click reaction, luminescence, and particularly, ultrahigh methane storage working capacity at 5−80 bar and 298 K.

M

etal−organic frameworks (MOFs) have emerged as new porous materials with enormous application potentials in gas adsorption,1 chemical separation,2 detection,3 catalysis,4 drug delivery,5 etc. Along with the maturing of MOF syntheses, it is now possible to design functional MOFs for various purposes and demands.6 On the way toward industrial applications, one major obstacle is the high synthetic cost of MOFs. In addition to simply pursuing cheap MOFs, other design strategies to reduce the total industrial cost are awaiting exploration. In the military area, a swing-role fighter can perform different roles by changing devices or software on short notice, while a multirole fighter carrying all kinds of devices can perform various roles in different missions. Both design strategies are well-developed for cost reduction, in contrast to the single-role fighter solely for a specific mission. Likewise, cost-saving MOFs for variable applications may be achieved by developing swingor multirole MOFs. Multivariate MOFs (MTV-MOFs) with multiple functional groups also have been developed to meet multiple purposes. Yaghi and co-workers reported a series of MTV-MOFs containing up to eight distinct functionalities in one phase.7 Cohen and co-workers applied postsynthetic modifications to obtain multifunctional MOFs.8 By using linkers with specific geometry and different functional groups, Zhang et al.9 could program the MOF pores for complex functional behavior. Burnett and Choe10 developed a stepwise pillar insertion route to connect two-dimensional (2D) layers into multifunctional three-dimensional (3D) MOFs. Recently, Zhou group devel© 2017 American Chemical Society

Received: February 7, 2017 Published: April 7, 2017 6034

DOI: 10.1021/jacs.7b01320 J. Am. Chem. Soc. 2017, 139, 6034−6037

Communication

Journal of the American Chemical Society

Figure 1. (a) Synthesis of the parent LIFM-28 for multirole MOFs showing replaceable coordination sites and pockets. (b) Transformation to functional MOFs LIFM-70−86 via spacer installation. (c) Framework topology interconversion. (d) Four types of channels in the parent MOF. (e) Straight and cage-window channels in functionalized MOFs.

which is simply accomplished via uninstallation of the inserted spacers. Similar to LIFM-29−33 described in a PVSI strategy,14 LIFM-70−86 can be converted back to LIFM-28 when soaked in water for various times (Figure 1a,b). This reversible process is representatively testified by crystal transformations between LIFM-28 and LIFM-77, -82, -83, and -86 and has been characterized by powder X-ray diffraction (PXRD) and 1H NMR spectroscopy (Figures 2, S3−S6, and S63−S66). The experiments started from a batch of LIFM-28 crystals, which were transformed into LIFM-77 first, then turned back into LIFM-28. The second round was the interconversion between LIFM-28 and -83, and this was followed by reversible transformations to LIFM-82 and -86. PXRD monitoring of these crystal transformations confirmed perfect and repeatable structural interconversions, as further verified by the CO2 adsorption measurements, which gave the same adsorption capacity for the transformed and as-prepared LIFM-79 (Figure S89c,d). These spacer installation/uninstallation experiments prove the parent LIFM-28 is robust enough to carry on successive and repeating functionalization switches, tunable between different functional versions, and compatible with varied functional units. Hence, proto-LIFM-28 is distinguished with two key features of a swing- or multirole MOF that is able to fulfill different tasks via repeated loading/release of diverse functional spacers. The phase purity of the MOFs was demonstrated by PXRD (Figures S7−S24). 1H NMR spectral examination of digested LIFM-70−86 compared with LIFM-28 verified nearly quantitative installation and uninstallation of one short and one long spacer per Zr6 cluster in LIFM-70−86 (Figures S46−S62 and Table S3), in agreement with the single-crystal results. To evaluate the stability and robustness for practical applications, thermogravimetric analysis and variable-temperature PXRD of LIFM-70−86 were performed (Figures S27−S45), revealing that most of functionalized MOFs retain their framework integrity and crystallinity above 400 °C. Nitrogen adsorption at 77 K was measured for LIFM-70−86 to evaluate their permanent porosity after activation (Figures S67−S85), revealing typical type-I isotherms of micropore nature. The Brunauer−Emmett−Teller (BET) surface areas range from 1079 to 1899 m2 g−1, the total pore volumes from 0.41 to 0.79 cm3 g−1, and the density functional theory (DFT)-

86 with isomorphous structures but variant functional units, can be obtained via one-step postsynthetic modification and restored to the parent LIFM-28. Such reversible and dynamic processes of spacer installation and uninstallation guarantee easy multifunctional switching based on LIFM-28 and endow the prototypical framework with various functionalities for targeted tasks, e.g., gas separation, catalysis, click reaction, luminescence, and extraordinary methane storage via introduction of different functional units (Figure 2; vide infra). The proto-LIFM-28 contains an 8-connected Zr6 cluster that has four pairs of replaceable H2O terminals as site A along the c axis and site B along the a/b axis. Two sites A form a space named pocket A that is suitable for insertion of shorter spacers of biphenyldicarboxylate (BPDC) derivatives (L1−7), while two sites B build a space named pocket B adequate for insertion of longer spacers derived from terphenyldicarboxylate (TPDC) (L8−13). All of the spacers carry distinctive functional units for orthogonal optimization of desired properties (see the Supporting Information for synthetic details). Postsynthetic modification of proto-LIFM-28 can be readily accomplished by soaking pristine LIFM-28 crystals in a DMF solution containing both short and long spacers at 85 °C for 40 h, offering the functionalized versions LIFM-70−86 (Figure 2). Single-crystal analyses of all of the functionalized MOFs (Table S1) confirmed that the short and long spacers are pinpointed in pockets A and B, respectively (Figure S1), bringing in functional units including amine, trifluoromethyl, fluorine, methyl, phenolic hydroxyl, 2,2-bipyridine, Pd-coordinated 2,2-bipyridine, and azide groups. The spacer installation changes the framework topology from 8-connected bcu to 12connected bcu-x (Figure 1) and consolidates the flexibility of proto-LIFM-28. Before installation, LIFM-28 has four types of straight channels (channels A and B along the a/b axis and channels C and D along the c axis Figure 1). After installation, all of the A- and B-type channels are blocked. The biggest channel C is compartmentalized into two types of tetrahedra (cages A and B), leading to narrowed pores (∼5 Å) ascribed as cagewindow channels decorated with −CF3 and functional groups (Figures 1 and S2). Meanwhile, channel D (∼9 Å) is preserved with its surface modified by functional units (Figure S2). The pivotal feature of proto-LIFM-28 as a parent MOF for multirole switching arises from its defunctionalization dynamics, 6035

DOI: 10.1021/jacs.7b01320 J. Am. Chem. Soc. 2017, 139, 6034−6037

Communication

Journal of the American Chemical Society

and -79 with every unit cell containing six and four amine groups, respectively. The CO2, CH4, and N2 adsorption analyses show an evident enhancement of CO2 capture upon amine functionalization, resulting in higher CO2 adsorption capacity and isosteric heat (Qst) with increasing amine group density (Figures S87 and S98). The adsorption selectivities for CO2/N2 and CO2/CH4 binary mixtures were calculated using ideal adsorption solution theory (IAST),19 revealing exceptional CO2/CH4 and CO2/N2 separation capabilities of 9.9 and 34.7, respectively, in LIFM-77 (Figure 3a and Table S5), representing

Figure 3. (a, b) IAST-calculated CO2/CH4 (50:50), CO2/N2 (15:85), and R22/N2 (10:90) selectivities at 273 K. (c, d) Excess and total CH4 adsorption isotherms at 298 K.

2- and 6-fold increases compared with LIFM-28. Along this way, the optimal separation of chlorodifluoromethane (R22), an ozone-depleting and potent greenhouse species, was examined by introducing methylated or fluorinated spacers, giving LIFM82 and -86. The results showed a rather positive effect on R22 adsorption through functionalization with both methyl and trifluoromethyl/fluorine units (Figures S90 and S100). The initial IAST R22/N2 selectivities were increased by a factor of 2.6 with methylated LIFM-82 and 5.6 with fluorinated LIFM-86 compared with LIFM-28, reaching values of 110.2 and 232.9, respectively (Figure 3a and Table S6). These results clearly show that installing spacers with amine units successfully turns proto-LIFM-28 into a functional version for CO2 capture, while installing spacers with fluorinated units effectively changes it to another version for R22 separation. Meanwhile, methane storage is an important MOF application, and methyl groups have been proved to benefit CH4 adsorption,18 so LIFM-28 was functionalized with different amounts of −CH3 groups to form LIFM-82 and -83 (Figure S86). Each unit cell of LIFM-82 and -83 has six and four methyl groups, respectively. Moreover, the installed spacers occupy open metal sites (OMSs) of proto-LIFM-28, which is helpful for the working capacity since less CH4 is adsorbed in the lowpressure region.16 As expected, the two MOFs show ultrahigh CH4 volumetric uptakes and working capacities (Tables 1 and S7). At 80 bar, the total CH4 uptakes are 271 cm3(STP) cm−3 for LIFM-82 and 265 cm3(STP) cm−3 for LIFM-83 (Figures 3c,d and S92−S93), very close to the record of the best MOFs.15 Most strikingly, their volumetric working capacities at 5−80 bar and 298 K reach 218 and 213 cm3(STP) cm−3, respectively. These values outperform the benchmark HKUST-1 under the

Figure 2. Two types of spacers installed into LIFM-28 by combination with functional units A and B in LIFM-76−86 and multifunctional switching among targeted tasks via a swing-role strategy.

calculated pore sizes from 10.9 to 13.2 Å (Table S4). In general, the former two pore values are enlarged after spacer installation, except for those of bulky units; however, the comparable pore sizes of LIFM-70−86 and LIFM-28 by DFT evaluation may only reflect the cage diameter because of compartmentalization of the biggest channels C with reduced window openings. Since two types of spacers can be predictably and precisely installed to introduce variable functionalities and tune the pore attributes, orthogonal optimization is readily applicable for the multirole MOF to achieve optimal desired properties, as depicted in Figure 2. As carbon capture, methane storage, and gas separation are main directions of MOF applications, we first demonstrate the swing-role strategy by optimizing the sorption properties toward different targets (Figures S86−S104). Spacers L3, L4, and L10 bearing amine groups can be installed into LIFM28 to improve CO2 adsorption because of their well-known strong interactions with CO2 molecules;18 this gives LIFM-77 6036

DOI: 10.1021/jacs.7b01320 J. Am. Chem. Soc. 2017, 139, 6034−6037

Communication

Journal of the American Chemical Society

Table 1. Summary of MOF Densities, Pore Volumes, BET Surface Areas, Isosteric Heats, Total Methane Uptakes, and Working Capacities (Desorption at 5 bar) at 35 and 80 bar and 298 K working capacity at 5−80 (5−35) bar

total uptake at 80 (35) bar

a

material

Dc (g cm−3)

Vp (cm3 g−1)a

BET surface area (m2 g−1)

cm3 cm−3

LIFM-28np LIFM-82 LIFM-83 MOF-905b HKUST-1 PCN-14 UTSA-76ac

1.043 0.922 0.917 0.537 0.883 0.829 0.699

0.44 0.71 0.72 1.34 0.78 0.83 1.09

940 1624 1715 3490 1850 2000 2820

223 (126) 271 (196) 265 (192) 228 (145) 272 (227) 250 (202) 257a (211)

g g−1 0.152 0.210 0.206 0.296 0.220 0.215 0.263

(0.086) (0.152) (0.150) (0.188) (0.184) (0.174) (0.216)

cm3 cm−3 183 (86) 218 (143) 213 (140) 203 (120) 200 (150) 178 (125) 197 (151)

g g−1 0.127 0.169 0.166 0.264 0.162 0.153 0.201

(0.059) (0.111) (0.109) (0.156) (0.122) (0.108) (0.154)

Qst (kJ/mol)

ref

− 17.5 16.5 11.7 17.0 17.6 15.4

this work this work this work 15 16 16 17

Calculated from N2 uptake at P/P0 = 0.97. bBased on pycnometer density data. cMeasured at 298 K and 65 bar.

ORCID

same conditions and are comparable with the record holder MOF-905 (which was strictly evaluated by pycnometer data considering defect sites)15 and further verified by wellrepeatable adsorption measurements (Figure S94). Besides methyl group optimization and OMS exclusion, the reduced pore aperture and cage diameter contribute to the outstanding working capacity,15,18 as spacer installation leads to compartmentalized cages. For more general multirole functional switches (Figure 2), LIFM-85 is designed to carry in an azide group, which opens a way for click reaction, as successfully demonstrated by the transformation to LIFM-87 via cyclization with propargylamine (Figures S105−S108). Luminescence of proto-LIFM-28 is able to be turned on by inserting polyaromatic TPDC analogues (Figure S109), which is obviously responsive and tunable by different functional substituents. Furthermore, functionalization with catalytic-unit-carrying spacers like L7 to switch to LIFM-81type MOFs may offer a platform for catalysis studies. These investigations toward multifunctional MOFs are underway. In summary, we have described a swing- or multirole strategy for the facile creation of MTV-MOFs based on dynamic postsynthesis with a robust Zr-MOF as the parent framework. Through simple and pinpoint installation/uninstallation of functionalized spacers, proto-LIFM-28 can be reversibly transformed to different functional versions for various applications. By virtue of orthogonal optimization, the amine-functionalized LIFM-77 shows excellent CO2 sorption and selectivity, the fluorinated LIFM-86 displays potential for fluorocarbon sequestration and separation, and the methylated versions LIFM-82 and -83 show exceptional methane storage and working capacities at 5−80 bar and 298 K. Multifunctional switches to catalysis, organic reaction, and luminescence functionality are also promising. Therefore, the swing- or multirole strategy may introduce a new avenue toward the synthesis of MTV-MOFs for multiple purposes via procedures involving easy manipulation and reduced total cost.

■

Ji-Jun Jiang: 0000-0001-9483-6033 Cheng-Yong Su: 0000-0003-3604-7858 Author Contributions †

C.-X.C. and Z.-W.W. contributed equally.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the NNSFC (21573291, 21603278), the STP Project of Guangzhou (201504010031), and the NSF of Guangdong Province (S2013030013474).

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b01320. Methods and additional data (PDF) Crystallographic data in CIF format (ZIP)

■

REFERENCES

(1) He, Y.; Zhou, W.; Qian, G.; Chen, B. Chem. Soc. Rev. 2014, 43, 5657. (2) Van de Voorde, B.; Bueken, B.; Denayer, J.; De Vos, D. Chem. Soc. Rev. 2014, 43, 5766. (3) Hu, Z.; Deibert, B. J.; Li, J. Chem. Soc. Rev. 2014, 43, 5815. (4) Zhang, T.; Lin, W. Chem. Soc. Rev. 2014, 43, 5982. (5) Zhou, H. C.; Kitagawa, S. Chem. Soc. Rev. 2014, 43, 5415. (6) Gao, W.-Y.; Chrzanowski, M.; Ma, S. Chem. Soc. Rev. 2014, 43, 5841. (7) Deng, H.; Doonan, C. J.; Furukawa, H.; Ferreira, R. B.; Towne, J.; Knobler, C. B.; Wang, B.; Yaghi, O. M. Science 2010, 327, 846. (8) Garibay, S. J.; Wang, Z.; Tanabe, K. K.; Cohen, S. M. Inorg. Chem. 2009, 48, 7341. (9) Zhang, Y.-B.; Furukawa, H.; Ko, N.; Nie, W.; Park, H. J.; Okajima, S.; Cordova, K. E.; Deng, H.; Kim, J.; Yaghi, O. M. J. Am. Chem. Soc. 2015, 137, 2641. (10) Burnett, B. J.; Choe, W. CrystEngComm 2012, 14, 6129. (11) Yuan, S.; Lu, W.; Chen, Y.-P.; Zhang, Q.; Liu, T.-F.; Feng, D.; Wang, X.; Qin, J.; Zhou, H.-C. J. Am. Chem. Soc. 2015, 137, 3177. (12) Yuan, S.; Chen, Y.-P.; Qin, J.-S.; Lu, W.; Zou, L.; Zhang, Q.; Wang, X.; Sun, X.; Zhou, H.-C. J. Am. Chem. Soc. 2016, 138, 8912. (13) Park, H. J.; Cheon, Y. E.; Suh, M. P. Chem. - Eur. J. 2010, 16, 11662. (14) Chen, C.-X.; Wei, Z.; Jiang, J.-J.; Fan, Y.-Z.; Zheng, S.-P.; Cao, C.C.; Li, Y.-H.; Fenske, D.; Su, C.-Y. Angew. Chem., Int. Ed. 2016, 55, 9932. (15) Jiang, J.; Furukawa, H.; Zhang, Y. B.; Yaghi, O. M. J. Am. Chem. Soc. 2016, 138, 10244. (16) Peng, Y.; Krungleviciute, V.; Eryazici, I.; Hupp, J. T.; Farha, O. K.; Yildirim, T. J. Am. Chem. Soc. 2013, 135, 11887. (17) Li, B.; Wen, H.-M.; Wang, H.; Wu, H.; Yildirim, T.; Zhou, W.; Chen, B. Energy Environ. Sci. 2015, 8, 2504. (18) Sumida, K.; Rogow, D. L.; Mason, J. A.; McDonald, T. M.; Bloch, E. D.; Herm, Z. R.; Bae, T.-H.; Long, J. R. Chem. Rev. 2012, 112, 724. (19) Myers, A. L.; Prausnitz, J. M. AIChE J. 1965, 11, 121.

AUTHOR INFORMATION

Corresponding Author

*[email protected] 6037

DOI: 10.1021/jacs.7b01320 J. Am. Chem. Soc. 2017, 139, 6034−6037