Versatile Assembly of Metal-Coordinated Calix[4]resorcinarene

May 23, 2017 - We propose a design strategy for assembly of metal-coordinated calix[4]resorcinarene cavitands and cages by tuning of the ancillary lin...
0 downloads 0 Views 8MB Size
Article pubs.acs.org/JACS

Versatile Assembly of Metal-Coordinated Calix[4]resorcinarene Cavitands and Cages through Ancillary Linker Tuning Wen-Yuan Pei,†,⊥ Guohai Xu,‡,§,∥,⊥ Jin Yang,*,† Hui Wu,§ Banglin Chen,∥ Wei Zhou,*,§ and Jian-Fang Ma*,† †

Key Lab for Polyoxometalate Science, Department of Chemistry, Northeast Normal University, Changchun 130024, China Key Laboratory of Jiangxi University for Functional Materials Chemistry, School of Chemistry and Chemical Engineering, Gannan Normal University, Ganzhou, Jiangxi 341000, China § NIST Center for Neutron Research, National Institute of Standards and Technology, Gaithersburg, Maryland 20899-6102, United States ∥ Department of Chemistry, University of Texas at San Antonio, One UTSA Circle, San Antonio, Texas 78249-0698, United States ‡

S Supporting Information *

ABSTRACT: We propose a design strategy for assembly of metal-coordinated calix[4]resorcinarene cavitands and cages by tuning of the ancillary linkers. Assembly of newly functionalized cavitand with angular isophthalic acid analogs affords three intriguing metal-coordinated cavitands with deep cavities, 1a−1c. Further, by mediating appropriate spacers between two isophthalic acids, two bowl-shaped cavitands are successfully joined together to produce three elegant coordination cages with tunable sizes and shapes, 2a−2c. The cavitand and cage crystals possess considerable amount of accessible porosities, as clearly established by gas adsorption measurements. Remarkably, 1a−1c also exhibit high structural flexibilities, reversibly transforming between the open-pore and the narrow-pore structures, upon removal and sorption of guest molecules, as evidenced by diffraction and gas adsorption measurements. By combining experimental studies with density functional theory (DFT) calculations, we thoroughly elucidated the mechanism of the structural transformations in response to external stimuli in this new class of flexible porous solids.



INTRODUCTION Coordination-driven self-assembly based on metal−ligand bonds has evolved as an efficient way to achieve discrete metal−organic molecular architectures.1−5 In this facet, metalcoordinated cavitands and cages with well-defined internal voids have attracted a great deal of attention because of their appealing structures and potential applications in gas sorption, molecular recognition, and catalysis.6 In general, rational assembly of highly directional ligands and appropriate metal ions can produce elaborate metal-coordinated cavitands and cages with various sizes, shapes, and cavities.7 In this context, calixarenes, as a special family of macrocyclic host compounds, are excellent ligand candidates,8 which have been extensively utilized in supramolecular chemistry.9 As an important subset of calix[4]arenes, calix[4]resorcinarenes, featuring a bowl-shaped aromatic cavity, have been recognized to be a greatly versatile class of molecules.10 The upper rims of the calix[4]resorcinarenes can be well modified with various functional groups making them intriguing cavitands.11 The utilization of the decorated cavitands in combination with metal ions has resulted in a range of metalcoordinated cavitands to date.12 Nevertheless, metal-coordinated cavitands with highly deep cavities are exceedingly rare. © 2017 American Chemical Society

Importantly, the functionalized calix[4]resorcinarene cavitands with bowl-shaped structures are also ideal scaffolds for coordination cage assembly. Some typical dimeric calix[4]resorcinarene cages have been constructed through metalmediated assembly, in which two calix[4]resorcinarene-based cavitands are directly joined together by metal ions.7b,12 To our knowledge, however, coordination cage of two metalcoordinated calix[4]resorcinarenes being held together by ancillary poly(carboxylic acid) linkers remains almost unknown, probably owing to the synthetic challenge.6−8 Here, we put forth a design strategy for assembly of metalcoordinated calix[4]resorcinarene cavitands and tunable cages through tuning of the ancillary linkers. On the basis of this idea, two new bowl-shaped calix[4]resorcinarene cavitands (MeTPC4R and Pen-TPC4R) were designed by introducing four chelating 2-(4H-pyrazol-3-yl)pyridine moieties at the upper rim (Schemes 1 and S1 of the Supporting Information, SI). Assembly of the cavitands with variable isophthalic acid analogs and metal ions affords a series of intriguing isostructural metalcoordinated cavitands with highly deep cavities, [Zn4(MeReceived: March 31, 2017 Published: May 23, 2017 7648

DOI: 10.1021/jacs.7b03169 J. Am. Chem. Soc. 2017, 139, 7648−7656

Article

Journal of the American Chemical Society Scheme 1. Structures of Me-TPC4R and Pen-TPC4R Cavitands Used in This Work (Me-TPC4R: R = Methyl; PenTPC4R: R = Pentyl)

was found that the angular isophthalic acid analogs support the formation of the metal-coordinated cavitands during the assembly of 1a−1c. When we varied the substituted groups of the isophthalic acid analogs, isostructural metal-coordinated cavitands were obtained. For the tetracarboxylic acid analogs, appropriate length, geometry, and flexibility of the spacer between two isophthalic acids are important factors that influence the cage assembly. Crystal Structures of 1a−1c. Single-crystal XRD reveals that 1a−1c exhibit isoreticular structures with variable isophthalic acid analogs (Scheme 2; Figures 1 and 2), and crystallize in the same tetragonal space group P4/n (Table S1). Hence, the structure of 1a will be described here in detail as a representative. Scheme 2. Angular Isophthalic Acid Analogs Used in This Work

TPC4R)(L1)4]·6DMF·2H2O (1a), [Zn4(Me-TPC4R)(L2)4]· 3DMF·H2O (1b) and [Zn4(Me-TPC4R)(L3)4]·5DMF (1c) (H2L1 = 5-aminoisophthalic acid, H2L2 = 5-hydroxyisophthalic acid, H2L3 = 5-(pyridin-4-yl)isophthalic acid and DMF = N,N′dimethylformamide). Inspired and motivated by the successful preparation of the metal-coordinated cavitands, we envision that tetracarboxylic acid analogs with appropriate spacers between two isophthalic acids may link two cavitands into tunable cages in the presence of metal ions. To our great delight, a family of nanoscale calix[4]resorcinarene-based cages with tunable sizes, shapes, and cavities, [Zn 8 (MeTPC4R)2(L4)4]·5DMF·20H2O (2a), [Zn8(PenTPC4R)2(L5)4]·7DMF (2b), and [Zn8(Pen-TPC4R)2(L6)4]· 6DMF (2c), were successfully prepared by using this strategy (H4L4 = 5,5′-methylene-bis(oxy)diisophthalic acid, H4L5 = 5,5′(1,3-phenylenebis(methylene))bis(oxy)diisophthalic acid and H 4 L 6 = 5,5′-(1,4-phenylenebis(methylene))bis(oxy)diisophthalic acid). 2a−2c represent a new class of tunable cages where two metal-coordinated calix[4]resorcinarene cavitands are joined together by systematically tuning ancillary poly(carboxylic acid)s. Supramolecular assemblies of these metal-coordinated calix[4]resorcinarene cavitands and cages are fairly stable and highly crystalline. After desolvation, they exhibit permanent gasaccessible porosities. This was clearly established by gas adsorption measurements. Remarkably, 1a−1c also exhibit high structural flexibilities, reversibly transforming between the open pore (op) and the activated narrow pore (np) form, upon removal and reintroduction of guest molecules. This behavior is similar to those found in flexible metal−organic framework (MOFs) or soft porous crystals (SPCs), featuring reversible structural transformations triggered by external stimuli.13,14 Therefore, 1a−1c represent a nice new addition to the family of solid porous stimuli-responsive materials.4a,15 Importantly, by combining experimental studies with density functional theory (DFT) calculations, we were also able to thoroughly elucidate the mechanism of the structural transformations in response to external stimuli in this new class of flexible porous solids.

Figure 1. Side view of the structure of the metal-coordinated cavitand [Zn4(Me-TPC4R)(L1)4] in 1a. Zn: pink; C: green; N: blue; O: red. The large yellow sphere represents the inner cavity of the cavitand.

In 1a, each L1 linker bridges two neighboring Zn(II) ions around the bowl rim to form a macrocyclic [Zn4(L1)4] unit, in which each Zn(II) atom is further chelated by the 2-(4Hpyrazol-3-yl)pyridine group at the upper rim, thus leading to an appealing metal-coordinated cavitand (Figures 1 and 2). The Zn(II)···Zn(II) distance across the calix[4]resorcinarene bowl is approximately 13.3 Å (Figure S1a). It is of interest to note that the cavity deepness of 1a is about 10.7 Å from the lower rim to the upper rim (Figure S1b), and the diameter of the bowl-shaped cavitand between the lower rim and the upper rim varies from ca. 5.21 Å to ca. 16.20 Å. Adjacent cavitands are packed against each other through π−π interactions between the 2-(4H-pyrazol-3-yl)pyridine groups [the distances of centroid-to-centroid and centroid-to-plane (based on the pyridine groups) are 3.63 and 3.58 Å, respectively] to yield fascinating supramolecular layers, in which the upper rims and the lower rims of the bowls are alternately arranged. Furthermore, adjacent supramolecular layers are stacked along



RESULTS AND DISCUSSION Self-Assembly of 1a−1c and 2a−2c. All metal-coordinated cavitands and cages were assembled by introducing isophthalic acid or tetracarboxylic acid analogs into the solvothermal reaction of Zn(NO3)2·6H2O and Me-TPC4R or Pen-TPC4R systems. During the syntheses of 1a−1c and 2a− 2c, careful control of the reaction temperature and the solvent ratio can favor the formation of large single crystals (see SI). It 7649

DOI: 10.1021/jacs.7b03169 J. Am. Chem. Soc. 2017, 139, 7648−7656

Article

Journal of the American Chemical Society

Figure 2. Structures of the metal-coordinated cavitands [Zn4(Me-TPC4R)(L1)4] in 1a, [Zn4(Me-TPC4R)(L2)4] in 1b, and [Zn4(Me-TPC4R)(L3)4] in 1c.

the c axis, through van der Waals interaction, to furnish an intriguing 3D porous architecture, as shown in Figure 3. The total solvent-accessible volume is ca. 4279.6 Å3 (approximately 52.1% of the unit-cell volume), calculated by the PLATON program.16

volume) is a little bigger than that of 1a. Since the pores of 1c were partially occupied by the pyridyl groups of H2L3, the total solvent-accessible volume for 1c is ca. 4212.0 Å3 (approximately 48.3% of the unit-cell volume), which is the smallest one of the three compounds. It needs to be pointed out that although a range of metal-coordinated cavitands have been reported thus far,7b 1a−1c, featuring the modular structures and highly deep cavitand cavities, represent an unusual example of metal-coordinated calix[4]resorcinarene cavitands involving variable isophthalic acid analogs. Crystal Structures of 2a−2c. Encouraged by the successful assembly of the metal-coordinated cavitand structures above, we conceived that two bowl-shaped calix[4]resorcinarene cavitands may be assembled into a coordination cage by introducing tetracarboxylic acid analogs with appropriate spacers between two isophthalic acids. A family of coordination cages, 2a−2c, featuring tunable sizes, shapes, and cavities, were successfully achieved by adjusting the spacers of the tetracarboxylic acid linkers (Schemes 3 and S2). Single-crystal XRD reveals that 2a crystallizes in the orthorhombic system with space group Pccn, 2b in the tetragonal system with space group I4/m, and 2c in the triclinic system with space group P-1 (Table S2). The structure Scheme 3. Tetracarboxylic Acid Analogs Used in This Work

Figure 3. Crystal structure of 1a. The 3D porous supramolecular architecture is stabilized by π−π interactions and van der Waals interactions. The large yellow spheres represent the inner cavities of the cavitands, and the green cylinders represent the intersupermolecule 1-D pores along the crystallography c axis.

By varying the isophthalic acid analogs, isostructural 1b and 1c were also achieved (Figures 2, S2, and S3), demonstrating the generality of this assembly strategy. Noticeably, 1c has a much higher cavity deepness of ca. 12.5 Å (Figure S4) than those of 1a and 1b. In 1c, the pyridyl groups are free and point toward the interior of the channel, which partially occupied the packing pores (Figure S3). The total solvent-accessible volumes for 1b (ca. 4351.6 Å3, approximately 53.2% of the unit-cell 7650

DOI: 10.1021/jacs.7b03169 J. Am. Chem. Soc. 2017, 139, 7648−7656

Article

Journal of the American Chemical Society

Figure 4. Structures of [Zn4(Me-TPC4R)(L4)4] cage in 2a, [Zn8(Pen-TPC4R)2(L5)4] cage in 2b and [Zn8(Pen-TPC4R)2(L6)4] cage in 2c. The methyl groups in 2a and the pentyl groups in 2b and 2c are omitted for clarity.

of the cage consists of two metal-coordinated calix[4]resorcinarene cavitands and four semiflexible tetracarboxylic acid linkers, featured as a fascinating nanometer-sized coordination cage. As shown in Figure 4, each of the four 2(4H-pyrazol-3-yl)pyridine groups at the upper rim chelates one Zn(II) ion to produce a metal-coordinated [Zn4-TPC4R]8+ cavitand unit. Further, two [Zn4-TPC4R]8+ units are joined together by four semiflexible tetracarboxylic acids to give a remarkable [Zn8(TPC4R)2(Ln)4] nanocage (n = 4−6). As expected, the modular cavitand units [Zn4(TPC4R)(Ln)4]4+ (n = 1−3) in 1a−1c were well-maintained in cages 2a−2c, further demonstrating the versatile applicability of our assembly strategy. It should be noted that 2a and 2b exhibit spindle-shaped cavities, while 2c has a ellipsoid-shaped one, as illustrated in Figure 4. This is ascribed to the different structural flexibility and geometry of the tetracarboxylic acid linkers. The longitudinal diameters of the cages are tuned by the variable tetracarboxylic acid spacers, which well correspond to the increasing length of the tetracarboxylic acids in the order of H4L4 < H4L5 < H4L6. The longitudinal dimensions of the internal cavity are approximately 2.3, 2.8, and 3.0 nm for 2a− 2c, respectively, identified as the distance between the two opposite centroids of the two cavitand lower rims (Figure S5). These are longer than most of the cages directly joined together by two metal-coordinated calix[4]resorcinarenes reported in the literature.12 The cages are stacked together through van der Waals interactions, to form 3D crystals. As illustrated in Figures 5 and S6, in addition to the large internal spherical cavities, there are also obvious porosities presented between the cages of 2a. The pore spaces are presumably occupied by disordered solvent molecules. PLATON calculations indicate the presence of approximately 43.7, 62.5, and 43.1% void space in 2a−2c, respectively (Figures S7 and S8).16

Figure 5. Crystal structure of 2a. The large yellow spindle-shaped spheres represent the inner cavities of the cages, and the green cylinder represents the intersupermolecule 1-D pores along the crystallography c axis.

High Structural Flexibility of 1a−1c. After describing the crystal structures of the as-synthesized metal-coordinated cavitands and cages, we next discuss the corresponding desolvated phases. Guest molecules in the as-synthesized crystals can be removed by acetone-exchange and activation under vacuum at elevated temperature (see SI). Upon desolvation, the PXRD patterns of 1a−1c changed (Figures S9−S12), suggesting a structural transformation. Interestingly, upon resolvation (immersing the activated samples in DMF), 7651

DOI: 10.1021/jacs.7b03169 J. Am. Chem. Soc. 2017, 139, 7648−7656

Article

Journal of the American Chemical Society the PXRD patterns can almost be restored (Figures S9−S11), implying that the process is reversible. These observations prompted us to investigate the structural flexibility in the metal-coordinated cavitand crystals in detail.17 Here, we focus on 1a as the representative in our discussion, since the three compounds are isostructural and are expected to behave similarly. Direct single-crystal XRD measurement on activated 1a sample was difficult. Thus, we relied on PXRD to elucidate the structure of the desolvated phase. The PXRD pattern was indexed to space group P4/n, which is the same as that of the as-synthesized crystal. Therefore, the tetragonal crystal symmetry of 1a remains unchanged before and after solvent removal. The a/b axis and cell volume V, however, significantly contracts from 25.82 and 8165.1 Å3 to 21.39 and 5503.8 Å3, respectively, indicating that the structure of 1a is indeed highly flexible. The coefficients of the contraction for a/b and V are 17.2% and 32.6%, respectively. To elucidate the mechanism of the structural flexibility of 1a, we performed a detailed computational first-principles investigation, based on dispersion-corrected density functional theory (DFT-D) (see SI for details). Variable-cell relaxation on the crystal structure of 1a was performed to simulate the lattice contraction process upon activation. Cell dimensions were changed smoothly on the fly, starting from those of the assynthesized crystal, shifting toward those of the desolvated crystal. An animation consisting of snapshots of the structural relaxation shows clearly how the structure of 1a evolves upon lattice contraction (see the movie in SI). Essentially, the cavitand supermolecules undergo large degree of rotation and distortion, leading to narrowed intersupermolecule voids, a denser packing, and a large crystal lattice shrinkage (particularly, within the a-b plane). The calculated pore volume reduction as a function of the lattice constant a decrease was quantitatively shown in Figure 6.

Figure 7. Rietveld refinement of the experimental powder X-ray diffraction pattern of activated 1a. The calculated pattern (line) well corresponds to the experimental data (circles) as evidenced by the difference pattern (line below observed and calculated patterns). Vertical bars show the calculated positions of Bragg peaks. Goodness of fit data: Rwp = 0.0868, Rp = 0.0635.

Comparing the experimental structures of the as-synthesized phase and the desolvated phase, one can clearly see how the supermolecular packing dramatically changes upon lattice contraction. As shown in Figure 8(a,b), 1a transforms from the op type to the np type upon guest removal, with a significant void fraction decrease from 52.1% to 12.7%, as calculated by PLANTON.16 The intersupermolecule 1-D pore is almost completely diminished in the np structure. While the internal cavity of each individual metalcoordinated cavitand of 1a is largely remained, careful structural comparison shows that the subunits of the cavitand also undergo substantial distortion upon desolvation. The 2-(4Hpyrazol-3-yl)pyridine moiety was twisted around the −CH2− group, in which the angle between the Cphenyl and Npyrazol atoms decreases from 103.1° to 95.8°, as illustrated in Figure 8(c,d). Additionally, the L1 linker rotated around the Zn(II)···Zn(II) axis in response to solvent removal. The dihedral angle between the Zn4 plane and the phenyl ring of the L1 linker greatly decreases from 50.8o to 28.7o (Figure S13), resulting in an increase of the N···N distance across the cavitand from 16.2 to18.3 Å. Noticeably, these findings represent an exceedingly rare example where the structural transformation mechanism for the discrete metal−organic architectures is fully elucidated.4a,15b It also shows that theoretical calculations combining diffraction measurements is a feasible strategy to uncover the mechanism of reversible structural transition in porous crystals. Gas Adsorption Induced Pore Opening in 1a and 1b. The reversible structural transformation of the metal-coordinated cavitands upon desolvation/solvation prompted us to investigate their gas sorption properties, and how the structures respond to adsorption/desorption of gas molecules. Measurements were carried out on both activated 1a and 1b (1c is omitted due to its lower porosity). We first tested N2 adsorption measurement at 77 K, and found that the N2 diffusion kinetics is very slow, most likely due to the narrow pore openings of the np phases. Therefore, we focused on CO2 adsorption.

Figure 6. Calculated pore volume reduction of 1a with decreasing lattice constant a upon desolvation.

Theoretical investigation not only revealed the mechanism of the structural flexibility of 1a, but also provided us a reasonable starting structural model to refine the PXRD data of the activated 1a. Rietveld refinement of the PXRD data of activated 1a was carried out. A reasonable goodness of fit was achieved (see Figure 7), and an experimental crystal structure was obtained. 7652

DOI: 10.1021/jacs.7b03169 J. Am. Chem. Soc. 2017, 139, 7648−7656

Article

Journal of the American Chemical Society

Figure 8. Packing structures of 1a before (a) and after (b) desolvation. The structure transformation upon desolvation and resolvation is reversible. Parts (c) and (d) show in more details the structural change of individual metal-coordinated cavitand unit.

from the dense intersupermolecule packing structure to an op structure. The CO2 adsorption increases sharply on the isotherm curve. To certain extend, this gas adsorption induced pore-opening process is analogous to the sample resolvation process. Of course, the gas adsorption induced op structure would not be as “open” as the fully solvated structure, since the gas-cavitand interaction is much weaker than the solventcavitand interaction. Indeed, the saturated CO2 uptake at 196 K in 1a is ∼174 cc/g, corresponding to a pore volume of ∼0.33 cc/g, significantly lower than the theoretical “maximal” pore volume of ∼0.56 cc/g of the as-synthesized 1a. To further evaluate the CO2 binding nature, we performed DFT-D calculations on activated 1a. We found that the initial CO2 adsorption mainly takes place inside the cavity of the metal-coordinated cavitand. Two strong adsorption sites are identified, as schematically shown in Figures 10 and S14. Site I is located near the upper rim of the cavitand, while site II is at the center of the lower rim of the cavitand. The calculated static CO2 binding energies are 31.6 and 27.7 kJ/mol, for the two sites, respectively. These are consistent with the experimental isosteric heat of adsorption Qst data (initial Qst ≈ 34.5 kJ/mol, see Figure S15). Full occupation of the two sites inside the

As illustrated in Figure 9, CO2 adsorption in activated 1a and 1b exhibits interesting two steps, with hysteresis in desorption. The presence of distinct adsorption steps in activated 1a and 1b suggests that CO2 molecules are able to induce a structural transformation between the contracted and the expanded lattice. In the first step, the CO2 molecules get adsorbed into the activated structure, and reach the first plateau in the adsorption isotherm upon pore saturation. On the basis of the CO2-196 K data of 1a, the first plateau (CO2 uptake ∼80 cc/g) corresponds to a pore volume of ∼0.15 cc/g, close to the calculated geometric pore volume of ∼0.12 cc/g of activated 1a. This suggests that the first step of CO2 adsorption does not involve significant lattice expansion (or pore opening), as expected. With CO2 pressure increased to a critical point (often called “gate pressure”), at ∼0.6 bar/196 K for 1a, the dense intersupermolecule packing of the cavitands gets pushed open to form an op structure. The CO2 adsorption proceeds into the second step on the isotherm curve. The CO2 adsorption of 1b at 196k is similar to that of 1a, and the first plateau corresponds to CO2 uptake of ca. 95 cc/g (Figure 9b). There is no significant lattice expansion (or pore opening). When the CO2 pressure increased to 0.33 bar (gate pressure), 1b was pushed 7653

DOI: 10.1021/jacs.7b03169 J. Am. Chem. Soc. 2017, 139, 7648−7656

Article

Journal of the American Chemical Society

where majority of the adsorption takes place in the channel pore. For comparison purpose, CH4 adsorption in 1a/1b was also investigated. Interestingly, from the isotherm data at 298 and 273 K, up to 65 bar, we do not observe a 2-step behavior equivalent to that of the CO2 adsorption (Figure S16). The CH4 uptakes of 1a and 1b are modest, ∼41 and 61 cm3(STP)/ g at 298 K and 65 bar, respectively. These are mainly from CH4 adsorption taking place inside the metal-coordinated cavitand, similar to the first-step of CO2 adsorption. Clearly, the interaction between CH4 and the cavitand is too weak to induce a pore opening under the measurement condition, as evidenced by the relatively low Qst of CH4 adsorption (Figure S17) compared to that of CO2. Permanent Porosity and Gas Adsorption in 2a. To evaluate the potential of metal-coordinated cage crystals for sorption related applications, we performed detailed gas adsorption measurements, using 2a as a representative. Unlike the three highly flexible metal-coordinated cavitand crystals, the structure of 2a is relatively rigid, as shown by the PXRD patterns before and after desolvation (Figure S18). N2 adsorption isotherm at 77 K indicates that 2a has a permanent porosity with a fully reversible type-I behavior (Figure S19). The maximum N2 uptake is ∼320 cm3·g−1, corresponding to a total pore volume of 0.499 cm3·g−1. The experimental BET surface area is ∼1158 m2·g−1 (Langmuir surface area ∼1302 m2·g−1). The pore size distribution derived from the N2 adsorption data shows that the pore sizes are mainly in the range of ∼6.5−14 Å (Figure S20a), consistent with the geometric pore size analysis based on the crystal structure (Figure S20b). We next examined the high-pressure CO 2 and CH 4 adsorption capacities of 2a. The isotherm data are shown in Figure 11. The CO2 adsorption of 2a can reach a maximum uptake of 257 cm3(STP)/g at 40 bar and 298 K. CH4 uptakes at 298 K are ∼123 cm3(STP)/g and 157 cm3(STP)/g, at 35 and 65 bar, respectively, showing the potential of 2a as a promising supramolecular porous material for CH4 storage.18 It is instructive to compare the adsorption energetics of CO2/CH4 in 2a with those in 1a/1b. According to the experimental data (Figure S21), the initial Qst values of CO2/CH4 in 2a are approximately the same as those found in 1a/1b. This indicates that the initial strong adsorption in the metal-coordinated cage crystal takes place at locations similar to those found in the metal-coordinated cavitand crystals, i.e., sites located inside the cavitand.

Figure 9. CO2 total adsorption isotherms for 1a (a) and 1b (b), measured at various temperatures.



CONCLUSIONS In summary, we have demonstrated a design strategy for assembly of metal-coordinated cavitands and tunable cages by precise tuning of ancillary linkers. This strategy opens up a versatile route to construct novel metal-coordinated cavitands and cages. Remarkably, 1a−1c represent a new class of flexible discrete metal−organic architectures, featuring a reversible structural transition between the op and np phases in response to guest molecules. Importantly, theoretical calculations combined with diffraction measurements were successfully employed to model the pore contraction/opening process, and to elucidate the structural transformation mechanism. Our metal-coordinated cage crystals (2a as a prototype) exhibit permanent porosities and relatively high gas uptakes, and thus may find promising applications for gas storage. Work is currently under way in our lab to identify further potential

Figure 10. DFT-D-calculated CO2 adsorption sites inside the cavitand of activated 1a. C and O of CO2 are shown in black and yellow, respectively, for clarity.

cavitand corresponds to 10 CO2 per unit cell, i.e., an uptake of ∼51 cm3(STP)/g. This largely accounts for the first step of the CO2 adsorption in activated 1a. In contrast, adsorption outside the cavitand at low pressure is of much smaller amount due to the narrow intersupermolecule pore space. At high pressure, a widening of the intercavitand channel pore is triggered, leading to the second step of CO2 uptake in the adsorption isotherm, 7654

DOI: 10.1021/jacs.7b03169 J. Am. Chem. Soc. 2017, 139, 7648−7656

Journal of the American Chemical Society



Article

AUTHOR INFORMATION

Corresponding Authors

*[email protected] *[email protected] *[email protected] ORCID

Hui Wu: 0000-0003-0296-5204 Banglin Chen: 0000-0001-8707-8115 Wei Zhou: 0000-0002-5461-3617 Jian-Fang Ma: 0000-0002-4059-8348 Author Contributions ⊥

These authors contributed equally.

Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This work was supported by the National Natural Science Foundation of China (Grant No. 21471029 and 21301026), and the state scholarship fund (Grant 201408360111) from China Scholarship Council (CSC).



(1) (a) Saha, M. L.; Yan, X.; Stang, P. J. Acc. Chem. Res. 2016, 49, 2527. (b) Cook, T. R.; Stang, P. J. Chem. Rev. 2015, 115, 7001. (c) Cook, T. R.; Zheng, Y.-R.; Stang, P. J. Chem. Rev. 2013, 113, 734. (d) Seidel, S. R.; Stang, P. J. Acc. Chem. Res. 2002, 35, 972. (e) Chakrabarty, R.; Mukherjee, P. S.; Stang, P. J. Chem. Rev. 2011, 111, 6810. (2) (a) Tranchemontagne, D. J.; Ni, Z.; O’Keeffe, M.; Yaghi, O. M. Angew. Chem., Int. Ed. 2008, 47, 5136. (b) Lu, Z.; Knobler, C. B.; Furukawa, H.; Wang, B.; Liu, G.; Yaghi, O. M. J. Am. Chem. Soc. 2009, 131, 12532. (c) Furukawa, H.; Kim, J.; Ockwig, N. W.; O’Keeffe, M.; Yaghi, O. M. J. Am. Chem. Soc. 2008, 130, 11650. (d) Wang, H.; Xu, J.; Zhang, D.-S.; Chen, Q.; Wen, R.-M.; Bu, X.-H.; Chang, Z. Angew. Chem., Int. Ed. 2015, 54, 5966. (3) (a) Ballester, P.; Fujita, M.; Rebek, J., Jr. Chem. Soc. Rev. 2015, 44, 392. (b) Yoshizawa, M.; Klosterman, J. K.; Fujita, M. Angew. Chem., Int. Ed. 2009, 48, 3418. (c) Wang, S.; Sawada, T.; Ohara, K.; Yamaguchi, K.; Fujita, M. Angew. Chem., Int. Ed. 2016, 55, 2063. (d) Fujita, D.; Ueda, Y.; Sato, S.; Mizuno, N.; Kumasaka, T.; Fujita, M. Nature 2016, 540, 563. (e) Tian, D.; Chen, Q.; Li, Y.; Zhang, Y.-H.; Bu, X.-H.; Chang, Z. Angew. Chem., Int. Ed. 2014, 53, 837. (4) (a) McConnell, A. J.; Wood, C. S.; Neelakandan, P. P.; Nitschke, J. R. Chem. Rev. 2015, 115, 7729. (b) Zarra, S.; Wood, D. M.; Roberts, D. A.; Nitschke, J. R. Chem. Soc. Rev. 2015, 44, 419. (c) Ward, M. D.; Raithby, P. R. Chem. Soc. Rev. 2013, 42, 1619. (d) Vardhan, H.; Yusubov, M.; Verpoort, F. Coord. Chem. Rev. 2016, 306, 171. (e) Brown, C. J.; Toste, F. D.; Bergman, R. G.; Raymond, K. N. Chem. Rev. 2015, 115, 3012. (f) Turega, S.; Cullen, W.; Whitehead, M.; Hunter, C. A.; Ward, M. D. J. Am. Chem. Soc. 2014, 136, 8475. (g) Cullen, W.; Misuraca, M. C.; Hunter, C. A.; Williams, N. H.; Ward, M. D. Nat. Chem. 2016, 8, 231. (5) (a) Li, J.-R.; Zhou, H.-C. Nat. Chem. 2010, 2, 893. (b) Li, J.-R.; Yakovenko, A. A.; Lu, W.; Timmons, D. J.; Zhuang, W.; Yuan, D.; Zhou, H.-C. J. Am. Chem. Soc. 2010, 132, 17599. (c) Li, K.; Zhang, L.Y.; Yan, C.; Wei, S.-C.; Pan, M.; Zhang, L.; Su, C.-Y. J. Am. Chem. Soc. 2014, 136, 4456. (d) Niu, Z.; Fang, S.; Liu, X.; Ma, J.-G.; Ma, S. Q.; Cheng, P. J. Am. Chem. Soc. 2015, 137, 14873. (e) Zhai, Q.-G.; Mao, C.; Zhao, X.; Lin, Q.; Bu, F.; Chen, X.; Bu, X.; Feng, P. Angew. Chem., Int. Ed. 2015, 54, 7886. (f) Amouri, H.; Desmarets, C.; Moussa, J. Chem. Rev. 2012, 112, 2015. (6) (a) Dalgarno, S. J.; Power, N. P.; Atwood, J. L. Coord. Chem. Rev. 2008, 252, 825. (b) Bi, Y. F.; Du, S. C.; Liao, W. P. Coord. Chem. Rev. 2014, 276, 61. (c) Frischmann, P. D.; MacLachlan, M. J. Chem. Soc. Rev. 2013, 42, 871. (d) Chen, L.; Chen, Q.; Wu, M.; Jiang, F.; Hong,

Figure 11. CO2 (a) and CH4 (b) total adsorption isotherms for 2a, measured at 273 and 298 K.

applications of these fascinating new metal-coordinated cavitand based materials.



REFERENCES

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/jacs.7b03169. DFT calculations, synthetic procedures, characterization details, gas adsorption isotherms, figures related to crystals, Tables and other data (PDF) Crystallographic data for 1a (CIF) Crystallographic data for activated 1a (CIF) Crystallographic data for activated 1a with CO2 (CIF) Crystallographic data for 1b (CIF) Crystallographic data for 1c (CIF) Crystallographic data for 2a (CIF) Crystallographic data for 2b (CIF) Crystallographic data for 2c (CIF) Movie evolving lattice contraction of 1a (AVI) 7655

DOI: 10.1021/jacs.7b03169 J. Am. Chem. Soc. 2017, 139, 7648−7656

Article

Journal of the American Chemical Society M. Acc. Chem. Res. 2015, 48, 201. (e) Liu, M.; Liao, W. P.; Hu, C.; Du, S. C.; Zhang, H. J. Angew. Chem., Int. Ed. 2012, 51, 1585. (f) Kumari, H.; Mossine, A. V.; Kline, S. R.; Dennis, C. L.; Fowler, D. A.; Teat, S. J.; Barnes, C. L.; Deakyne, C. A.; Atwood, J. L. Angew. Chem., Int. Ed. 2012, 51, 1452. (7) (a) Jin, P.; Dalgarno, S. J.; Atwood, J. L. Coord. Chem. Rev. 2010, 254, 1760. (b) Pinalli, R.; Dalcanale, E.; Ugozzoli, F.; Massera, C. CrystEngComm 2016, 18, 5788. (c) Ronson, T. K.; Fisher, J.; Harding, L. P.; Rizkallah, P. J.; Warren, J. E.; Hardie, M. J. Nat. Chem. 2009, 1, 212. (d) Kajiwara, T.; Iki, N.; Yamashita, M. Coord. Chem. Rev. 2007, 251, 1734. (8) (a) Hang, X. X.; Liu, B.; Zhu, X. F.; Wang, S. T.; Han, H. T.; Liao, W. P.; Liu, Y. L.; Hu, C. H. J. Am. Chem. Soc. 2016, 138, 2969. (b) Qiao, Y.; Zhang, L.; Li, J.; Lin, W.; Wang, Z. Angew. Chem., Int. Ed. 2016, 55, 12778. (c) Fowler, D. A.; Rathnayake, A. S.; Kennedy, S.; Kumari, H.; Beavers, C. M.; Teat, S. J.; Atwood, J. L. J. Am. Chem. Soc. 2013, 135, 12184. (d) Dai, F.-R.; Sambasivam, U.; Hammerstrom, A. J.; Wang, Z. J. Am. Chem. Soc. 2014, 136, 7480. (e) Xiong, K.; Jiang, F.; Gai, Y.; Yuan, D.; Chen, L.; Wu, M.; Su, K.; Hong, M. Chem. Sci. 2012, 3, 2321. (f) Bi, Y. F.; Wang, X. T.; Liao, W. P.; Wang, X. F.; Wang, X. W.; Zhang, H. J.; Gao, S. J. Am. Chem. Soc. 2009, 131, 11650. (g) Zhang, S.-T.; Yang, J.; Wu, H.; Liu, Y.-Y.; Ma, J.-F. Chem. - Eur. J. 2015, 21, 15806. (h) Lv, L.−L.; Yang, J.; Zhang, H.-M.; Liu, Y.-Y.; Ma, J.-F. Inorg. Chem. 2015, 54, 1744. (9) (a) Patil, R. S.; Banerjee, D.; Zhang, C.; Thallapally, P. K.; Atwood, J. L. Angew. Chem., Int. Ed. 2016, 55, 4523. (b) Morohashi, N.; Narumi, F.; Iki, N.; Hattori, T.; Miyano, S. Chem. Rev. 2006, 106, 5291. (c) Kumari, H.; Dennis, C. L.; Mossine, A. V.; Deakyne, C. A.; Atwood, J. L. J. Am. Chem. Soc. 2013, 135, 7110. (d) Iyer, K. S.; Norret, M.; Dalgarno, S. J.; Atwood, J. L.; Raston, C. L. Angew. Chem., Int. Ed. 2008, 47, 6362. (10) (a) Kane, C. M.; Banisafar, A.; Dougherty, T. P.; Barbour, L. J.; Holman, K. T. J. Am. Chem. Soc. 2016, 138, 4377. (b) Kane, C. M.; Ugono, O.; Barbour, L. J.; Holman, K. T. Chem. Mater. 2015, 27, 7337. (c) Dondoni, A.; Marra, A. Chem. Rev. 2010, 110, 4949. (11) (a) Asadi, A.; Ajami, D.; Rebek, J., Jr. J. Am. Chem. Soc. 2011, 133, 10682. (b) Biros, S. M.; Rebek, J., Jr. Chem. Soc. Rev. 2007, 36, 93. (12) (a) Kobayashi, K.; Yamanaka, M. Chem. Soc. Rev. 2015, 44, 449. (b) Yamanaka, M.; Kawaharada, M.; Nito, Y.; Takaya, H.; Kobayashi, K. J. Am. Chem. Soc. 2011, 133, 16650. (c) Zuccaccia, D.; Pirondini, L.; Pinalli, R.; Dalcanale, E.; Macchioni, A. J. Am. Chem. Soc. 2005, 127, 7025. (13) (a) Horike, S.; Shimomura, S.; Kitagawa, S. Nat. Chem. 2009, 1, 695. (b) Gould, S. L.; Tranchemontagne, D.; Yaghi, O. M.; GarciaGaribay, M. A. J. Am. Chem. Soc. 2008, 130, 3246. (c) Férey, G.; Serre, C. Chem. Soc. Rev. 2009, 38, 1380. (d) Yuan, S.; Zou, L.; Li, H.; Chen, Y.-P.; Qin, J.; Zhang, Q.; Lu, W.; Hall, M. B.; Zhou, H.-C. Angew. Chem., Int. Ed. 2016, 55, 10776. (e) Yang, S.; Lin, X.; Lewis, W.; Suyetin, M.; Bichoutskaia, E.; Parker, J. E.; Tang, C. C.; Allan, D. R.; Rizkallah, P. J.; Hubberstey, P.; Champness, N. R.; Thomas, K. M.; Blake, A. J.; Schröder, M. Nat. Mater. 2012, 11, 710. (f) Mason, J. A.; Oktawiec, J.; Taylor, M. K.; Hudson, M. R.; Rodriguez, J.; Bachman, J. E.; Gonzalez, M. I.; Cervellino, A.; Guagliardi, A.; Brown, C. M.; Llewellyn, P. L.; Masciocchi, N.; Long, J. R. Nature 2015, 527, 357. (g) Taylor, M. K.; Runčevski, T.; Oktawiec, J.; Gonzalez, M. I.; Siegelman, R. L.; Mason, J. A.; Ye, J.; Brown, C. M.; Long, J. R. J. Am. Chem. Soc. 2016, 138, 15019. (h) Schneemann, A.; Bon, V.; Schwedler, I.; Senkovska, I.; Kaskel, S.; Fischer, R. A. Chem. Soc. Rev. 2014, 43, 6062. (14) (a) Zhang, J. P.; Liao, P. Q.; Zhou, H. L.; Lin, R. B.; Chen, X. M. Chem. Soc. Rev. 2014, 43, 5789. (b) Chang, Z.; Yang, D.-H.; Xu, J.; Hu, T.-L.; Bu, X.-H. Adv. Mater. 2015, 27, 5432. (c) Deria, P.; GómezGualdrón, D. A.; Bury, W.; Schaef, H. T.; Wang, T. C.; Thallapally, P. K.; Sarjeant, A. A.; Snurr, R. Q.; Hupp, J. T.; Farha, O. K. J. Am. Chem. Soc. 2015, 137, 13183. (d) Lama, P.; Aggarwal, H.; Bezuidenhout, C. X.; Barbour, L. J. Angew. Chem., Int. Ed. 2016, 55, 13271. (e) Zhou, H. L.; Lin, R. B.; He, C. T.; Zhang, Y. B.; Feng, N.; Wang, Q.; Deng, F.; Zhang, J. P.; Chen, X. M. Nat. Commun. 2013, 4, 2534. (f) MellotDraznieks, C.; Serre, C.; Surblé, S.; Audebrand, N.; Férey, G. J. Am.

Chem. Soc. 2005, 127, 16273. (g) 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. (h) Dybtsev, D. N.; Chun, H.; Kim, K. Angew. Chem., Int. Ed. 2004, 43, 5033. (j) Foo, M. L.; Matsuda, R.; Hijikata, Y.; Krishna, R.; Sato, H.; Horike, S.; Hori, A.; Duan, J.; Sato, Y.; Kubota, Y.; Takata, M.; Kitagawa, S. J. Am. Chem. Soc. 2016, 138, 3022. (15) (a) Jacobs, T.; Lloyd, G. O.; Gertenbach, J.-A.; MüllerNedebock, K. K.; Esterhuysen, C.; Barbour, L. J. Angew. Chem., Int. Ed. 2012, 51, 4913. (b) Park, J.; Sun, L.-B.; Chen, Y.-P.; Perry, Z.; Zhou, H.-C. Angew. Chem., Int. Ed. 2014, 53, 5842. (c) Gao, Q.; Xu, J.; Cao, D.-P.; Chang, Z.; Bu, X.-H. Angew. Chem., Int. Ed. 2016, 55, 15027. (16) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7. (17) Wang, H.; Li, B.; Wu, H.; Hu, T.-L.; Yao, Z.; Zhou, W.; Xiang, S.; Chen, B. J. Am. Chem. Soc. 2015, 137, 9963. (18) (a) Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M. Science 2013, 341, 6149. (b) Furukawa, S.; Reboul, J.; Diring, S.; Sumida, K.; Kitagawa, S. Chem. Soc. Rev. 2014, 43, 5700. (c) He, Y.; Zhou, W.; Qian, G.; Chen, B. Chem. Soc. Rev. 2014, 43, 5657.

7656

DOI: 10.1021/jacs.7b03169 J. Am. Chem. Soc. 2017, 139, 7648−7656