Controlling flexibility of metal–organic frameworks

Controlling flexibility of metal–organic frameworks Abstract Framework flexibility is one of the most important characteristics of metal–organic frameworks (MOFs), which is not only interesting, but also useful for a variety of applications. Designing, tailoring or controlling framework flexibility of MOFs is much more difficult than for static structural features such as the framework topology and pore size/shape. Nevertheless, with in-depth understanding of the relationship between framework flexibility and the host framework structure, guest loading and other aspects such as the crystal size/morphology and external physical environment, some strategies have been developed for controlling the flexibility of MOFs and the corresponding properties, which are summarized and discussed in this review. porous coordination polymers, metal–organic frameworks, flexibility, dynamism, adsorption, structural transformation INTRODUCTION Porous coordination polymers (PCPs), or more commonly quoted as metal–organic frameworks (MOFs), are highly ordered polymeric open network structures based on coordination bonding connections [1,2]. By virtue of the vast diversity of metal ions and organic ligands, as well as a couple of successful crystal engineering principles, MOFs are well known for their modular designability, enabling the framework topology, pore size/shape and pore surface characteristics to be constructed on demand [2]. Besides structures and applications [3], MOFs are also famous for their flexible structures responsive to external stimuli, such as guest change, temperature change, light irradiation and mechanical forces, which have attracted extensive interests especially after the pioneering works by Kitagawa and Férey [4–9]. In most MOFs, different parts of the host framework can rotate around single bonds within an organic ligand and/or between a metal ion and an organic ligand. The reversible nature of coordination bonds and weaker supramolecular interactions also enables the internal connections within the MOF crystals to reversibly break and reform. These structural features give rise to multiple, easily accessible thermodynamic metastable states. Consequently, flexible MOFs can show diverse types and magnitudes of structural dynamism. Importantly, many of the structure changes occur with the retention of the periodicity and even the single-crystallinity, which enable the transformed structures and host–guest interactions to be visualized at the atomic level by crystallography techniques including the convenient in-house single-crystal X-ray diffraction [10–13]. Compared with rigid adsorbents, flexible MOFs can have advantages for many applications such as storage and separation. Flexible MOFs usually show S-shaped or multi-step adsorption isotherms [14], which can be very useful for practical gas storage/delivery because these shapes can enlarge the uptake difference between the charge and discharge pressures [15,16]. Further, the structural transformation of a MOF may consume/provide the heat generated/required during the adsorption/desorption process, respectively, reducing the requirement of fast heat exchange between the container and environment [16,17]. While rigid adsorbents distinguish different guest species by either the differential binding (matching of size, shape and functional group) or the size exclusion (molecular sieving) mechanism, flexible MOFs can simultaneously utilize all available mechanisms and alter their structures to emphasize the differences of guests [18–20]. For example, the gate-opening pressure of adsorption depends on both the binding affinity and molecular size of the guest. Therefore, very large uptake differences similar to the molecular sieving effect can be easily observed for flexible MOFs [21,22]. Unlike rigid molecular sieves, precise adjustment of the aperture size is not necessary for flexible MOFs. The changeable framework structures and pore sizes/shapes of flexible MOFs can be also disadvantages. For example, SOD-[Zn(mim)2] (MAF-4/ZIF-8, Hmim = 2-methylimidazole) [23], possessing well-defined small pore aperture size (3.2 Å [24]) and originally thought to be rigid [25], had been expected to show good molecular sieving effect but the experimental results were disappointed [26], because it is actually flexible and can readily adsorb much larger molecules such as toluene and p-xylene [27,28]. Accompanying the rapidly growing research on flexible MOFs, many review articles focusing on their interesting phenomena, useful properties, characterization methods, etc., have been published in the past few years [8–10,29–34]. Considering the critical roles of framework flexibility for related applications, this review intends to discuss strategies for controlling the flexibility of MOFs. It should be noted that flexible MOFs have intrinsic abilities to show different structural transformations or dynamic behaviors toward different external stimuli, which should be regarded as controlling/changing structure rather than controlling/changing flexibility. For clarity, controlling flexibility of MOFs is defined as changing structural response or dynamic behavior of a MOF sample toward a given external stimulus, upon changing the structure of the MOF sample and/or changing its external environment (Fig. 1). Obviously, unlike static structural features (such as the framework topology and pore size/shape), controlling framework flexibility is much more difficult. Actually, very different flexibility can be observed for MOFs showing trivial structural differences [35]. And drastic structural transformations, especially those involving breakage/reformation of coordination bonds, are generally serendipitous. Figure 1. View largeDownload slide Controlling flexibility of MOFs: from observing different structural responses toward different external stimuli to tuning structural response toward a given external stimulus. Figure 1. View largeDownload slide Controlling flexibility of MOFs: from observing different structural responses toward different external stimuli to tuning structural response toward a given external stimulus. In principle, framework flexibility is determined by the structure of the MOF sample and/or the external environment. The structure of a MOF sample has three levels of meanings, namely the host coordination network structure, the crystal structure and the particle structure, in which each definition is a subset of latter ones. The host coordination network structure, or simply the host framework structure, is the basic/fundamental component for the structure–property relationship, because MOF applications usually involve host–guest interplays, and most studies have been devoted to designing/modifying the host framework structure. For some applications, the structure of a MOF sample needs to be described by its crystal structure, which consists of not only the host framework, but also the guest included in the crystal. Actually, some components inside the MOF crystals, such as a hardly removable molecule reinforced by coordination bonding or other interactions, can be defined as either the host or the guest, depending on the experimental design. The particle structure of a MOF sample further includes the size, morphology, defect and surface modification of the crystal sample, which are all effective for the sample property but usually not regarded as the intrinsic structural parameters of MOFs. Outside the MOF sample, temperature, pressure and some other physical parameters can not only drive the structural transformation of the MOF, but also modulate the MOF flexibility toward other stimuli. A strategy for controlling MOF flexibility can rely on the host, guest, particle characters and/or external physical environment, in which the concept gradually changes from designing new materials to changing property of existing materials (Fig. 1). In this context, this review summarizes and discusses representative examples and strategies for designing/tailoring/controlling flexibility of MOFs in the three sections, focusing on the host framework structure, the guest loading and the other means including the particle characters and external physical environment. CONTROLLING FLEXIBILITY BY FRAMEWORK DESIGN Although flexible MOFs can be constructed by using flexible organic ligands, by connecting metal ions and organic ligands with single coordination bonds or by designing low-dimensional or interpenetrating coordination networks, these structural features have poor designability/controllability. On the other hand, truly designable MOFs are generally based on some well-defined framework prototypes/topologies and less flexible building units. Nevertheless, the rigidity/flexibility of these designable MOFs can be rationally controlled by judicious selections of organic ligands and metal ions/clusters. Besides the rigid/flexible backbones, the bridging length and non-coordinative side groups of organic ligands can play important roles to the framework rigidity/flexibility of MOFs. The long backbone of organic ligands can bend more easily, giving rise to more flexible MOFs [36,37]. [M2(dobdc)] (MOF-74/CPO-27, H4dobdc = 2,5-dioxido-1,4-benzenedicarboxylic acid) is a typical rigid MOF structure with 3D honeycomb-like coordination network and 1D channels [38]. Expanded versions of MOF-74 with mesopores can be rationally constructed by lengthening the ligand backbone (inserting 1–10 phenyl rings) [38,39]. Similar to other mesoporous materials, the expanded versions of MOF-74 can show a multi-step gas-adsorption isotherm. Interestingly, small-angle X-ray scattering demonstrated the formation of a superlattice structure associated with the multi-step isotherm, which was ascribed to the special location of gas molecules [40]. On the other hand, computational simulations suggested that framework distortion at the inorganic chain was the origin of a superlattice [41]. Actually, X-ray single-crystal and powder diffraction analyses showed that adding one more phenyl ring to the ligand backbone of MOF-74 is enough to induce framework flexibility with significant and distinct contraction/expansion toward N,N-dimethylformamide (DMF), ethylenediamine and CO2, in which the ligand backbone bends differently [39,42]. Long organic ligands can increase framework flexibility without backbone bending. Férey et al. showed that the breathing amplitude of acs-[M3(μ3-O)(ldc)3(LT)3] (MIL-88, H2ldc = linear dicarboxylic acid, LT = terminal ligand) can significantly increase along with the length of the linear dicarboxylate ligand (e.g. 125% for MIL-88B based on 1,4-benzenedicarboxylate and 230% for MIL-88D based on 4,4΄-biphenyldicarboxylate) [43]. Crystal-structure analyses showed that the framework breathing is mainly originated from bending of the M2=O2>C junctions rather than other parts. Because the bridging directions of the organic ligands do not point to the exact center of the trinuclear cluster, the shape/volume change of the network is smaller than that of the metal–carboxylate junction, which is more serious for the shorter dicarboxylate ligands. While changing the bridging length of organic ligand is usually regarded as designing new MOFs, changing the non-coordinative side group is somewhat closer to the concept of modification of known MOFs. For framework flexibility, non-coordinative side groups can play steric (repulsive), attractive and/or electronic roles [44]. [Co(bdp)] (H2bdp = 1,4-benzenedipyrazole) reported by Long et al. is a highly flexible (160% volume increase from guest-free to guest-saturated form) pillared-rod structure [45,46]. At 298 K, this MOF shows a gate-opening-type adsorption for CH4, which is critical for achieving the remarkably high working capacity of 155 cm3 cm–3 for CH4 storage at 5–35 bar [16]. To tune the gate-opening pressure, they further compared the gas-adsorption properties and corresponding framework transformations of a series of isostructural [Co(Rbdp)] frameworks, which demonstrated important roles for the electronic, steric and positional effects of the side groups. For CH4 adsorption, introduction of electron withdrawing F as a side group can reduce the gate-opening pressure [47], while methyls capable of forming stronger edge-to-face π−π interactions can increase the gate-opening pressure [48]. The attractive/repulsive forces of non-coordinative side groups toward guest species can be also utilized to control framework flexibility. Compared with other forms of framework flexibility, breakage/formation of coordination bonds is more difficult to design/control, as exemplified by the scarce examples of the change of the net interpenetration number of MOFs [49–52]. In a special case, namely the inter-conversion between 5-fold and 6-fold interpenetrated dia-f networks of [Ag6X(Rtz)4]OH·nH2O (X = Cl− or Br−; HRtz = Hatz/Hmtz = 3-amino-1,2,4-triazole/3-methyl-1,2,4-triazole) [49], the framework flexibility can be controlled by the hydrophlicity/hydrophobicity of the ligand side groups, which controls the guest accessibility of the metal ions [53]. When hydrophilic amino groups were employed to allow the guest OH–/H2O to attack the linearly coordinated Ag(I) ions that help the formation of low-energy intermediates during the bond-breaking/reformation processes, both [Ag6Cl(atz)4]OH·nH2O and [Ag6Br(atz)4]OH·nH2O can simultaneously show interpenetration reconstitution and remarkable breathing. On the other hand, both [Ag6Cl(mtz)4]OH·nH2O and [Ag6Br(mtz)4]OH·nH2O with hydrophobic methyl groups prevent such attacks and only show large framework breathing (Fig. 2). Figure 2. View largeDownload slide Controlling the framework breathing/reconstitution behaviors of [Ag6X(Rtz)4]OH·nH2O. Left: simultaneous framework distortion and interpenetration reconstitution (between 5- and 6-fold) for R = NH2. Right: framework distortion without interpenetration reconstitution for R = CH3. Adapted from [53] with permission of the Royal Society of Chemistry. Figure 2. View largeDownload slide Controlling the framework breathing/reconstitution behaviors of [Ag6X(Rtz)4]OH·nH2O. Left: simultaneous framework distortion and interpenetration reconstitution (between 5- and 6-fold) for R = NH2. Right: framework distortion without interpenetration reconstitution for R = CH3. Adapted from [53] with permission of the Royal Society of Chemistry. As a single atom, a metal ion can induce framework flexibility by the structural diversity of its coordination sphere, including the change of coordination bond length, coordination geometry and coordination number. The change of coordination bond length occurs for a few special metal ions in some special cases (e.g. spin crossover). The change of coordination number, either permanent removal/addition of coordinative guests or transient breaking/reforming coordination bonds, is the main origin for many structural transformation phenomena. There are very few metal ions capable of drastically changing coordination geometry without simultaneous alternation of the coordination number. The large breathing ability of [Co(bdp)], in which the Co(II) ions possess the normal tetrahedral geometry in the open form, was attributed to the abnormal square-planar geometry in the guest-free form, as demonstrated by molecular-simulation-assisted analyses of synchrotron powder diffraction, magnetic measurements, and IR and UV–Vis–NIR spectroscopies [45,46]. Many MOF structures can adopt different metal ions via direct synthesis or post-synthetic ion exchange, and alternation of metal ion can readily modulate the framework flexibility. For example, being isostructural with [Co(bdp)], [Zn(bdp)] and [Fe(bdp)] can show similar breathing motions, but their breathing amplitudes and/or involved energy changes are different and can be used to tune the gas-storage performance [16,54]. Besides preferences of coordination geometries, larger metal ions (i.e. longer coordination bonds) may facilitate structural transformation due to the reduced steric hindrance effect [55–57]. For example, N2, H2, C2H2 and CO2 gas-adsorption measurements and multi-level computational simulations for a series of isostructural ultramicroporous MOFs, [Zn3(vtz)6] (MAF-123-Zn, Hvtz = 1,2,3-triazole), [Mn3(vtz)6] (MAF-123-Mn) and [Cd3(vtz)6] (MAF-123-Cd), demonstrated that a 0.1-Å increase in the metal ion radius (0.9, 1.0 and 1.1 Å for divalent octahedral Zn, Mn and Cd ions, respectively) can readily turn on the framework flexibility to give multi-step or S-shaped isotherms useful for gas storage/delivery [58]. Nevertheless, in most other cases, quantitative correlation of the differences of metal ions and framework flexibilities still needs further investigation, as exemplified by the complicated breathing trends of [M(bdp)] [16,54] and [M(μ-OH)(bdc)] (MIL-53, H2bdc = 1,4-benzenedicarboxylic acid) [59]. Compared with individual metal ions, polynuclear metal clusters/chains are more suitable for rational design of the framework structure and the control of framework flexibility. Metal clusters/chains used for MOF design/construction are usually terminated by carboxylate groups with two M–O bonds. Because the carboxylate oxygen atoms have very weak coordination bonding directionality, the M2=O2>C junction can usually bend to a large extent like a knee [43]. Nevertheless, the symmetries of such clusters/chains and their arrangements are also determinative for framework rigidity/flexibility [8,34]. Zn4(μ4-O)(RCOO)6 and M3(μ3-O)(RCOO)6(LT)3 are typical examples of rigid and flexible clusters, in which the octahedral and the trigonal-prismatic arrangements of metal–carboxylate junctions are the key difference, and only the latter one allows the six bending vectors to operate in parallel. Nevertheless, with the same [M3(μ3-O)(ldc)3(LT)3] composition, MIL-88 with a six-connected acs topology and MIL-101 with a six-connected mtn-e topology are highly flexible and rigid [60,61], because the clusters are arranged in parallel in a trigonal symmetry or isotropically in a cubic symmetry, respectively. Similar phenomena have also been observed in MOFs based on metal–carboxlyate chains [5,62]. The flexibility mode of M3(μ3-O)(RCOO)6(LT)3 clusters can be rationally tuned/controlled by framework design. In the free M3(μ3-O)(RCOO)6(LT)3 clusters, the metal–carboxylate junctions adopt the linear conformation. In MIL-101 and MIL-88, as well as other MOFs containing such clusters, the metal–carboxylate junctions all bend toward the 3-fold axis of the clusters in different degrees. As the less bended metal–carboxylate junction can provide a larger framework volume, the trigonal framework MIL-88 shortens along the c-axis and expands in the ab-plane to accommodate more guests [43]. By using tripodal pyridyldicarboxylate ligands with pyridyl ends to further occupy the three monodentate terminal (LT) sites at the equatorial plane of the M3(μ3-O)(RCOO)6(LT)3 cluster, a (3,9)-connected xmz network [M3(μ3-O)(dcpb)3] (MCF-18, H2dcpb = 2,6-di-p-carboxyphenyl-4,4΄-bipyridine) can be obtained [63]. MCF-18 adopts a trigonal symmetry with the trinuclear clusters arranged in the same manner as for MIL-88, and it is also highly flexible. However, MCF-18 expands along the c-axis and contracts in the ab-plane to accommodate more guests, being reversed in the manner of MIL-88. Consequently, although MCF-18 exhibits much smaller volumetric breathing amplitude (105%) than MIL-88 (230%), its axial breathing amplitude (121%) is larger than for MIL-88 (100%) [43]. Careful structural examination showed that the metal–carboxylate junctions in MCF-18 all bend toward the equatorial plane, which accounts for its special breathing direction (Fig. 3). The unique shape of the dcpb2– ligand is responsible for the special metal–carboxylate conformation in MCF-18. Some MOFs isoreticular with MCF-18 can be constructed by using other tripodal pyridyldicarboxylate ligands, but they are all rigid and their metal–carboxylate junctions all bend toward the 3-fold axis [63]. Figure 3. View largeDownload slide Two types of distortion manners of the classic M3(μ3-O)(ldc)3(LT)3 cluster, leading to the reversed breathing directions for MIL-88 and MCF-18. Adapted from [10] and [63] with permission of the Royal Society of Chemistry. Figure 3. View largeDownload slide Two types of distortion manners of the classic M3(μ3-O)(ldc)3(LT)3 cluster, leading to the reversed breathing directions for MIL-88 and MCF-18. Adapted from [10] and [63] with permission of the Royal Society of Chemistry. Since framework flexibility is highly sensitive to trivial structural difference, changing components of isostructural/isoreticular MOFs can readily alter the framework flexibility but can hardly tune it to meet the requirement of desired application. Fortunately, similarly to other inorganic crystalline materials, MOFs can also adopt two or more analogous building blocks at the same crystallographic position with random distribution in different unit cells, giving solid-solution structures with variable stoichiometries to enable continuous adjustment of their structures and properties [24,64,65]. Kitagawa et al. demonstrated the usefulness of the solid-solution strategy for on-demand tuning of framework flexibility, by using two isostructural MOFs [Zn(5-NO2-ip)(bpy)] (CID-5, bpy = 4,4΄-bipyridyl, 5-NO2-H2ip = 5-nitroisophthalic acid) and [Zn(5-MeO-ip)(bpy)] (CID-6, 5-MeO-H2ip = 5-methoxyisophthalic acid) [64]. With a smaller substituent group, CID-5 is flexible and shrinks to a nonporous state that can adsorb CO2 only above a relatively high gate-opening pressure, and cannot adsorb CH4 up to 1.0 MPa. On the other hand, CID-6 with a larger substituent group is rigid and can adsorb both CO2 and CH4. By mixing the two ligands with different ratios, isostructural solid-solution frameworks [Zn(5-NO2-ip)x(5-MeO-ip)1−x(bpy)] (CID-5/6; x = 0.06-0.82) were obtained, showing tunable gate-opening pressure for CO2 adsorption. More importantly, the flexibility can be tuned to a state suitable for CO2/CH4 separation, at which the framework can be opened by CO2 and adsorb large amounts of CO2, while remaining repellent to CH4. Further breakthrough experiments showed that this strategy is useful for not only CO2/CH4 separation, but also C2H6/CH4 [66]. While the above-discussed strategies and examples can be still regarded as designing and synthesizing new materials, post-synthetic modification is a method closer to the purpose of controlling or tailoring structures/properties of existing materials [67]. Cohen et al. modified [Zn2(NH2-bdc)2(dabco)] (DMOF-1, NH2-H2bdc = 2-amino-1,4-benzenedicarboxylic acid, dabco = 1,4-diazabicyclo[2.2.2]octane) by reacting it with alkyl anhydrides (O(CO(CH2)nCH3)2, n = 0∼5) in CHCl3, giving a series of MOF crystals with different flexibility depending on the length of the appended alkyl chain [68]. DMOF-1 is a rigid framework showing normal type-I adsorption for N2, Ar and CO2. After modification, the crystals shrink after desolvation for n = 0∼2 due to attractive interactions between the alkyl chains, but remain rigid for n ≥ 3 because of steric hindrance effect. Further, gate-opening-type gas-adsorption isotherms were observed for n = 1 and 2 rather than n = 0. Obviously, precise control over the composition of a solid-solution framework, by either direct synthesis or post-synthetic modification, is critical for on-demand control of its property. However, the product compositions are generally different from the feeding ratios, and most post-synthetic modification reactions involve liquid reactants/reagents, preventing precise control/monitoring of the reaction/modification degrees [67]. This problem may be solved by using a solvent-free solid-gas reaction mechanism [69]. For example, we reported a flexible metal azolate framework [Cu4(btm)2] (MAF-42, H2btm = bis(5-methyl-1,2,4-triazol-3-yl)methane) consisting of Cu(I) ions and a methylene-bridged bistriazolate ligand [70]. Because the Cu(I) ions are two- or three-coordinated by the triazolate N donors being similar to those of the copper proteins and the methylene group activated by two aromatic rings are adjacent to these low-coordinated Cu(I) ions, the crystal can be oxidized to form [Cu4(btk)2] (H2btk = bis(5-methyl-1,2,4-triazol-3-yl)methanone) by O2 or air even at room temperature, with H2O as the only byproduct (Fig. 4). The carbonyl-bridged ligand is more rigid and more hydrophilic than the methylene-bridged one, leading to higher gate-opening pressure and larger adsorption hysteresis, as well as larger pore volumes. More importantly, without using any liquids, the oxidation degree can be simply monitored by the sample weight, so that the framework flexibility and surface characteristic can be tuned on demand. As a result, the CH4, C2H6 and CO2 adsorption selectivities can be drastically changed up to four orders in magnitude and even inversed. Figure 4. View largeDownload slide Post-synthetic modification and controlling flexibility of MAF-42. (a) Guest-induced framework breathing of MAF-42. (b) The solvent-free solid–gas reaction mechanism for post-synthetic modification. (c) Increasing framework rigidity and breathing amplitude by changing the flexible methylene groups to rigid carbonyl groups. Adapted from [70] with permission of Nature Publishing Group. Figure 4. View largeDownload slide Post-synthetic modification and controlling flexibility of MAF-42. (a) Guest-induced framework breathing of MAF-42. (b) The solvent-free solid–gas reaction mechanism for post-synthetic modification. (c) Increasing framework rigidity and breathing amplitude by changing the flexible methylene groups to rigid carbonyl groups. Adapted from [70] with permission of Nature Publishing Group. CONTROLLING FRAMEWORK FLEXIBILITY BY GUEST Changing guest loading in MOF crystals can be regarded as a special type of post-synthetic modification method. It is well known that the structures of flexible MOFs are highly dependent on the type and amount of the adsorbed guest species. Also, the manner of MOF structure change is always dependent on the type of guest, as indicated by their different isotherm shapes, breathing amplitudes and even breathing modes [43,71]. Nevertheless, these phenomena should not be regarded as controlling the flexibility of MOFs (Fig. 1). On the other hand, guests can be used to rationally modulate not only the physical properties (such as magnetism and luminescence), but also the framework flexibility of MOFs induced by physical stimuli such as changes in temperature and/or pressure. Guest-modulated thermal expansion of MOFs has been well demonstrated (Fig. 5), although guest-induced framework breathing is much larger than that induced by temperature change and it is usually difficult to keep constant guest-loading amounts at different temperatures. In this context, ultramicroporous MOFs have shown great potential, since the small pores strongly interact with guest molecules and restrict guest escape. Usually, the steric hindrance of guest molecules and host–guest attractive interactions are responsible for the change in thermal expansion coefficients, which can impede or promote the contraction of the host framework at low temperatures, giving smaller and larger thermal expansion coefficients, respectively [55,72–74]. Figure 5. View largeDownload slide Strategies for utilizing guests to control the thermal expansion behaviors of MOFs. Figure 5. View largeDownload slide Strategies for utilizing guests to control the thermal expansion behaviors of MOFs. The dynamic behaviors of the guest molecules, such as the rotation of a molecule and thermal expansion of a guest cluster, can be also used to control the thermal expansion properties of MOF crystals. For example, we designed and synthesized an ultramicroporous MOF [Mn(pba34)2] (MCF-34, Hpba34 = 3-(pyridin-4-yl)benzoic acid) by using a bent ligand with easily changeable conformation [75]. Because its 1D channels (Φmin/max = 3.9/4.4 Å) are too small for N2/O2 adsorption at low temperatures, it exhibits constant and huge thermal expansion (α = +224 × 10−6 K−1/–107 × 10−6 K−1) over a wide temperature range (127–673 K) in both vacuum and air. More interestingly, when DMF is included in MCF-34, the thermal expansion profile of the crystal showed an abrupt jumping around the melting point of the guest, because the rotation motion of DMF triggers the conformation reversion of the organic ligand, which further transmits this action to the whole crystal. To utilize the thermal expansion of guest clusters, we further synthesized a 3D hinge-like framework [Cd(pba34)(pba44)] (MCF-82, Hpba44 = 4-(pyridin-4-yl)benzoic acid) with quasi-discrete ultramicropores running along its a-axis, which allow DMF and N,N-dimethylacetamide (DMA) supramolecular dimers to be strongly confined yet exchangeable. The vacant crystal of MCF-82 shows extremely large positive/negative thermal expansion across its bc-plane (α = +482 × 10−6 K−1/–218 × 10−6 K−1) [76]. Meanwhile, a small contraction was observed for the a-axis during temperature increase. The guest-included crystals (especially for DMF) not only show largely reduced thermal expansion across the bc-plane, but also large expansion along the a-axis during temperature increase. Single-crystal X-ray diffraction indicated that the confined guest dimers impede the framework deformation across the bc-plane via the steric hindrance effect. Moreover, the DMF dimers form close π–π stacking interactions, which display overwhelmingly large thermal expansion along the a-axis to change the thermal expansion anisotropy of the MOF crystal. On the other hand, the DMA dimers form weak edge-to-edge interaction and small thermal expansion, so that its effect is much weaker than the DMF dimers. Alternation of guest inclusion can also modulate the dynamic response of flexible MOFs toward adsorption of other guests. Because adsorption measurements generally require degassing treatments by heating and/or vacuuming, strong host–guest binding (e.g. via coordination bonds) is necessary for the guest to modulate the flexibility and adsorption property of the MOF crystal [77–83]. Recently, Brammer et al. showed that desolvation of Me2NH2[In(abdc)2] (SHF-61, H2abdc = 2-aminobenzene-1,4-dicarboxylic acid) in two different solvents leads to two polymorphic-activated forms with very different pore openings and markedly different gas-adsorption properties (Fig. 6) [83]. With CHCl3 as the low-polarity solvent, desolvation leads to little change in the host framework, which exhibits a high CO2-adsorption capacity and a normal type-I isotherm shape. With DMF/H2O as the high-polarity solvent, desolvation leads to significant framework contraction, which also exhibits a normal type-I isotherm shape but with a low CO2-adsorption capacity. Interestingly, the framework contraction during desolvation of DMF/H2O is continuous, and partial desolvation produces a partially contracted framework, which exhibits a gating pressure associated with CO2 adsorption. Such interesting phenomena were ascribed to the different strengths of host–guest interactions induced by the solvent molecules, in which the high-polarity solvent helps the host framework to contact but can be broken by CO2 at high pressure [83]. Figure 6. View largeDownload slide Proposed roles of solvent-controlled framework dynamism and the CO2 adsorption property of SHF-61. Figure 6. View largeDownload slide Proposed roles of solvent-controlled framework dynamism and the CO2 adsorption property of SHF-61. Guest molecules with large molecular weights can form strong host–guest interactions, which are useful for controlling the guest loading for further modulating the MOF properties [84]. For example, [Co2(ndc)2(bpy)] (H2ndc = 2,6-napthalenedicarboxylic acid) is a 2-fold interpenetrated pcu network showing interesting flexibility, which simply shrinks to a less porous state upon guest removal at moderate temperature, and can further transform to a 3-fold interpenetrated nonporous structure by heating at higher temperature [85]. Uemura et al. showed that, after loading polystyrene, [Co2(ndc)2(bpy)] can keep the as-synthesized open structure after removing small guest molecules, and prevent the transformation to the 3-fold interpenetrated structure at higher temperature [86]. The polymer-included samples were prepared by adsorption of styrene monomer, polymerization and then vacuum removal of the residue styrene monomer. Obviously, the long-chain polymer guest can be hardly removed from the host using conventional MOF activation methods, which enable the precise control of guest loading. Propane and CO2 adsorption isotherms were measured for the MOF loaded with different amounts of polystyrene, which showed uptake trends following 5% loading >16% loading >0% loading. Obviously, an appropriate loading of polystyrene can increase the pore volume by expanding the framework from the less porous state to the most porous state, while not occupying too much of the pore. CONTROLLING FRAMEWORK FLEXIBILITY BY OTHER MEANS While the structure–property relationship is regarded as one of most important aspects for chemistry, the structural information of a MOF is generally referred to its chemical composition and crystal structure only. However, other particle parameters, such as size, morphology and surface modification, can also play important roles, being similar to conventional inorganic nanomaterials. For example, SOD-[Zn(bim)2] (MAF-3/ZIF-7, Hbim = benzimidazole [87]) is a flexible structure showing gate-opening-type hysteretic sorption isotherms for a variety of gases [25,88]. Choi et al. showed that MAF-3 with differing shapes and sizes (∼100 nm spherical, ∼400 nm rhombic-dodecahedral and ∼1300 nm rod-shaped) exhibit different N2 and CO2 sorption behaviors, including uptakes, gate-opening pressures and hysteresis widths [89]. In particular, the spherical sample can start to adsorb gases at very low pressure, which was ascribed to the possible presence of disordered structures near the outer shell with more flexibility than the core MAF-3 structure. Crystal downsizing can generally accelerate guest diffusion and hence facilitate structural transformation of flexible MOFs [90], and even make rigid frameworks flexible. For example, Kitagawa et al. reported guest adsorption in nonporous [Pt(CN)4Fe(py)2] (py = pyridine) enabled by crystal downsizing [91]. In [Pt(CN)4Fe(py)2], metal ions are bridged by cyanide ligands to form 2D layers, which interdigitate together using the monodentate pyridine ligands to form a nonporous structure. In the bulk state (particle size 135 nm), [Pt(CN)4Fe(py)2] cannot adsorb EtOH. After simple mechanical grinding (particle size 9 nm), it can adsorb 0.67 EtOH per formula unit. In situ powder X-ray diffraction analyses showed obvious shifting of diffraction peaks, indicating that the guest adsorption originates from the structural transformation of the host. Crystal downsizing can also impede structural transformation of flexible MOFs. As a representative flexible MOF, MAF-4 exhibits two-step N2 adsorption isotherm, and the second step starts at higher pressures for the smaller crystals [24,25,92,93]. Nevertheless, the relationship between crystal size and framework flexibility was discovered first by Kitagawa et al. [94–96]. [Zn(ip)(bpy)]·DMF (CID-1, H2ip = isophthalic acid) is an interdigitated stacking structure of 2D layers [94]. For bulk CID-1 with crystal sizes ∼5 μm × 20 μm, desolvation under vacuum at 130°C led to slight framework shrinkage, as indicated by the shifting of several powder X-ray diffraction peaks to the higher 2θ values. When the crystal size was reduced to nanoscale (500 × 100 × 30 nm3), the desolvation-induced shrinking reduced a lot, which was attributed to the change in crystal size or crystallite surface structure (covered by surfactant). Single-crystal X-ray and synchrotron powder diffraction analyses showed that bulk and nano CID-1 undergo 3.2% and 2.1% reductions of the unit-cell volume after desolvation, respectively [95]. Kitagawa et al. further developed a shape memory effect by downsizing the MOF crystal [96]. [Cu2(bdc)2(bpy)] is a 2-fold interpenetrated pcu network containing a notable porosity (void = 35%). A series of samples with different crystal sizes (methanol as guest) were synthesized. After guest removal under vacuum at room temperature, the sample with 300 × 300 × 30 nm3 or larger size shrink to a narrow-pore state (void = 20%). Surprisingly, the smaller nanoscale crystals can completely (50 × 50 × 20 nm3 and 60 × 60 × 20 nm3) or partially (110 × 110 × 23 nm3 and 160 × 160 × 25 nm3) retain the original open framework as observed for the as-synthesized form. When these nanocrystals were heated to 473 K, they could also transform to the shrunk state. Further, methanol adsorption measurement at 303 K for the crystals with a size of 50 × 50 × 20 nm3 showed a gate-opening isotherm for the closed-state sample, but an ordinary type-I isotherm for the open-state sample. Similar size-dependent flexibility was also demonstrated for an analogous MOF [Cu2(bdc)2(bpe)] (bpe = 1,2-bis(4-pyridyl)ethylene). The size-dependent flexibility of the MOFs was ascribed to the different numbers of defects present in the crystal, which promote the phase transition. Some flexible MOFs have been reported to show activation-method-dependent structures and flexibility [97,98]. Because it is generally difficult to prove the identical framework compositions for the samples obtained from different activation methods, such phenomena might be actually induced by the different residual guests. In principle, framework flexibilities or dynamic behaviors of MOFs can be also influenced by the outer physical environment such as temperature, light irradiation, mechanical force and electric field [71,99–102]. For example, [Cu(detz)] (MAF-2, Hdetz = 3,5-diethyl-1,2,4-triazole) is a 3D nbo-a network containing a bcu-type channel system, in which large cavities are connected by very small apertures blocked by the flexible ethyl groups (Fig. 7). Interestingly, MAF-2 shows abnormal N2 adsorption at 195 K rather than 77 K, which can be explained by the different flexibility of the ethyl groups. Only at high temperatures that provide enough thermal energy is the thermal motion of ethyl groups large enough to open the apertures for the transient passage of the N2 molecules. Single-crystal X-ray diffraction studies confirmed the variation in ethyl thermal motion at different temperatures, and there is no significant temperature-induced framework or guest-induced structural alteration for the host framework [71]. In contrast, thermal expansion and difference in kinetic diameters are the main reasons for most adsorbents showing higher gas adsorption at higher temperatures (mostly adsorption of CO2 at 195 K and no adsorption of N2 at 77 K). Figure 7. View largeDownload slide Temperature-controlled dynamism of ethyl groups that controls the effective aperture size of MAF-2. Adapted from [73] with permission of the American Chemical Society. Figure 7. View largeDownload slide Temperature-controlled dynamism of ethyl groups that controls the effective aperture size of MAF-2. Adapted from [73] with permission of the American Chemical Society. The temperature-controlled dynamism of flexible pore aperture can be also realized by using rigid frameworks and mobile guest species. For example, Bu et al. reported an interesting MOF [Cu2(btr)2](NO3)2 (btr = 4,4΄-bis(1,2,4-triazole)) with a rigid 3D cationic coordination network, in which the small apertures are blocked by the counter anion NO3– [103]. Similarly to the N2 adsorption of MAF-2, [Cu2(btr)2](NO3)2 can adsorb CO2 at high temperature (231 K) rather than low temperature (195 K and 226 K). Variable-temperature single-crystal X-ray diffraction analyses of [Cu2(btr)2](NO3)2 demonstrated the significant increase in the thermal motion of the nitrate guests. Mechanical pressure change can usually induce structural transformations of flexible MOFs similar to those induced by temperature change, and even more drastic transformations including reconstitution of coordination connectivities [101,104,105]. Obviously, mechanical pressure can be also used to modulate the flexibility of MOFs toward other external stimuli, albeit reported examples are still scarce. Long et al. observed the increases in the gate-opening pressure for CH4 adsorption and the energy of the gate-opening structural transformation, via increasing the [Co(bdp)] sample packing density, corresponding to applying higher mechanical pressure [101]. Although the CH4 uptake and working capacity both decrease under higher mechanical pressure, the higher energy consumption for the gate-opening structural transformation is beneficial for realizing a free-energy adsorption/desorption process. CONCLUSIONS Undoubtedly, framework flexibility and dynamic behavior are unique characteristics of MOFs (compared with other types of porous materials) and extremely important for their adsorption, thermal expansion and other properties. After the discovery of many interesting types of framework flexibilities, rational design/control of these behaviors is emerging as an attractive topic of research. As discussed above, the flexibility of MOFs can be designed/modified/controlled by many strategies, at different conceptual levels from designing/synthesizing new materials to tailoring/controlling properties of known materials. It should be pointed out that designing, tailoring or controlling framework flexibility is useful not only for understanding the structure–property relationship of MOFs, but also as a new dimension for developing MOF materials with excellent performances for molecular recognition, high storage/delivery capacity, selective separation, abnormal/controllable thermal expansion and so on. FUNDING This work was supported by the National Basic Research Program of China (973 Project, 2014CB845602) and the National Natural Science Foundation of China (21290173 and 21473260). REFERENCES 1. Kitagawa S, Kondo M. Functional micropore chemistry of crystalline metal complex-assembled compounds. Bull Chem Soc Jpn  1998; 71: 1739– 53. https://doi.org/10.1246/bcsj.71.1739 Google Scholar CrossRef Search ADS   2. Furukawa H, Cordova KE, O’Keeffe M et al. The chemistry and applications of metal-organic frameworks. Science  2013; 341: 1230444. https://doi.org/10.1126/science.1230444 Google Scholar CrossRef Search ADS PubMed  3. MacGillivray LR. Metal-Organic Frameworks: Design and Application . Hoboken, NJ, USA: John Wiley & Sons, Inc., 2010. Google Scholar CrossRef Search ADS   4. Kitaura R, Fujimoto K, Noro S et al.   A pillared-layer coordination polymer network displaying hysteretic sorption: [Cu2(pzdc)2(dpyg)]n (pzdc= Pyrazine-2,3-dicarboxylate; dpyg=1,2-Di(4-pyridyl)glycol). Angew Chem Int Ed  2002; 41: 133– 5. https://doi.org/10.1002/1521-3773(20020104)41:1%3c133::AID-ANIE133%3e3.0.CO;2-R Google Scholar CrossRef Search ADS   5. Serre C, Millange F, Thouvenot C et al. Very large breathing effect in the first nanoporous chromium(III)-Based Solids: MIL-53 or CrIII(OH)·{O2C-C6H4-CO2}·{HO2C-C6H4-CO2H}x H2Oy. J Am Chem Soc  2002; 124: 13519– 26. https://doi.org/10.1021/ja0276974 Google Scholar CrossRef Search ADS PubMed  6. Kitagawa S, Kitaura R, Noro S. Functional porous coordination polymers. Angew Chem Int Ed  2004; 43: 2334– 75. https://doi.org/10.1002/anie.200300610 Google Scholar CrossRef Search ADS   7. Uemura K, Matsuda R, Kitagawa S. Flexible microporous coordination polymers. J Solid State Chem  2005; 178: 2420– 9. https://doi.org/10.1016/j.jssc.2005.05.036 Google Scholar CrossRef Search ADS   8. Férey G, Serre C. Large breathing effects in three-dimensional porous hybrid matter: facts, analyses, rules and consequences. Chem Soc Rev  2009; 38: 1380– 99. https://doi.org/10.1039/b804302g Google Scholar CrossRef Search ADS PubMed  9. Horike S, Shimomura S, Kitagawa S. Soft porous crystals. Nat Chem  2009; 1: 695– 704. https://doi.org/10.1038/nchem.444 Google Scholar CrossRef Search ADS PubMed  10. Zhang J-P, Liao P-Q, Zhou H-L et al. Single-crystal X-ray diffraction studies on structural transformations of porous coordination polymers. Chem Soc Rev  2014; 43: 5789– 814. https://doi.org/10.1039/C4CS00129J Google Scholar CrossRef Search ADS PubMed  11. Suh MP, Cheon YE. Recent advances in the dynamics of single crystal to single crystal transformations in metal-organic open frameworks. Aust J Chem  2006; 59: 605– 12. https://doi.org/10.1071/CH06016 Google Scholar CrossRef Search ADS   12. Halder GJ, Kepert CJ. Single crystal to single crystal structural transformations in molecular framework materials. Aust J Chem  2006; 59: 597– 604. https://doi.org/10.1071/CH06322 Google Scholar CrossRef Search ADS   13. Kawano M, Fujita M. Direct observation of crystalline-state guest exchange in coordination networks. Coordin Chem Rev  2007; 251: 2592– 605. https://doi.org/10.1016/j.ccr.2007.07.022 Google Scholar CrossRef Search ADS   14. Coudert F, Jeffroy M, Fuchs AH et al. Thermodynamics of guest-induced structural transitions in hybrid organic–inorganic frameworks. J Am Chem Soc  2008; 130: 14294– 302. https://doi.org/10.1021/ja805129c Google Scholar CrossRef Search ADS PubMed  15. Zhang J-P, Chen X-M. Optimized acetylene/carbon dioxide sorption in a dynamic porous crystal. J Am Chem Soc  2009; 131: 5516– 21. https://doi.org/10.1021/ja8089872 Google Scholar CrossRef Search ADS PubMed  16. Mason JA, Oktawiec J, Taylor MK et al. Methane storage in flexible metal-organic frameworks with intrinsic thermal management. Nature  2015; 527: 357– 61. https://doi.org/10.1038/nature15732 Google Scholar CrossRef Search ADS PubMed  17. Lin R-B, Chen D, Lin Y-Y et al. A zeolite-like zinc triazolate framework with high gas adsorption and separation performance. Inorg Chem  2012; 51: 9950– 5. https://doi.org/10.1021/ic301463z Google Scholar CrossRef Search ADS PubMed  18. Liao P-Q, Huang N-Y, Zhang W-X et al. Controlling guest conformation for efficient purification of butadiene. Science  2017; 356: 1193– 6. https://doi.org/10.1126/science.aam7232 Google Scholar CrossRef Search ADS PubMed  19. Chen B, Liang C, Yang J et al. A microporous metal-organic framework for gas-chromatographic separation of alkanes. Angew Chem Int Ed  2006; 45: 1390– 3. https://doi.org/10.1002/anie.200502844 Google Scholar CrossRef Search ADS   20. Li L, Lin R, Krishna R et al. Flexible-robust metal-organic framework for efficient removal of propyne from propylene. J Am Chem Soc  2017; 139: 7733– 6. https://doi.org/10.1021/jacs.7b04268 Google Scholar CrossRef Search ADS PubMed  21. Witman M, Ling S, Jawahery S et al. The influence of intrinsic framework flexibility on adsorption in nanoporous materials. J Am Chem Soc  2017; 139: 5547– 57. https://doi.org/10.1021/jacs.7b01688 Google Scholar CrossRef Search ADS PubMed  22. Li J, Sculley J, Zhou H. Metal-organic frameworks for separations. Chem Rev  2012; 112: 869– 932. https://doi.org/10.1021/cr200190s Google Scholar CrossRef Search ADS PubMed  23. Huang X-C, Lin Y-Y, Zhang J-P et al. Ligand-directed strategy for zeolite-type metal-organic frameworks: zinc(II) imidazolates with unusual zeolitic topologies. Angew Chem Int Ed  2006; 45: 1557– 9. https://doi.org/10.1002/anie.200503778 Google Scholar CrossRef Search ADS   24. Zhang J, Zhu A, Lin R et al. Pore surface tailored sod-type metal-organic zeolites. Adv Mater  2011; 23: 1268– 71. https://doi.org/10.1002/adma.201004028 Google Scholar CrossRef Search ADS PubMed  25. Park KS, Ni Z, Cote AP et al. Exceptional chemical and thermal stability of zeolitic imidazolate frameworks. Proc Natl Acad Sci U S A  2006; 103: 10186– 91. https://doi.org/10.1073/pnas.0602439103 Google Scholar CrossRef Search ADS PubMed  26. Bux H, Liang F, Li Y et al. Zeolitic imidazolate framework membrane with molecular sieving properties by microwave-assisted solvothermal synthesis. J Am Chem Soc  2009; 131: 16000– 1. https://doi.org/10.1021/ja907359t Google Scholar CrossRef Search ADS PubMed  27. Zhu A-X, Lin R-B, Qi X-L et al. Zeolitic metal azolate frameworks (MAFs) from ZnO/Zn(OH)2 and monoalkyl-substituted imidazoles and 1,2,4-triazoles: efficient syntheses and properties. Micropor Mesopor Mat  2012; 157: 42– 9. https://doi.org/10.1016/j.micromeso.2011.11.033 Google Scholar CrossRef Search ADS   28. Zhang J-P, Zhu A-X, Chen X-M. Single-crystal X-ray diffraction and Raman spectroscopy studies of isobaric N2 adsorption in SOD-type metal-organic zeolites. Chem Commun  2012; 48: 11395– 7. https://doi.org/10.1039/c2cc35544b Google Scholar CrossRef Search ADS   29. Schneemann A, Bon V, Schwedler I et al. Flexible metal-organic frameworks. Chem Soc Rev  2014; 43: 6062– 96. https://doi.org/10.1039/C4CS00101J Google Scholar CrossRef Search ADS PubMed  30. Nagarkar SS, Desai AV, Ghosh SK. Stimulus-responsive metal-organic frameworks. Chem Asian J  2014; 9: 2358– 76. https://doi.org/10.1002/asia.201402004 Google Scholar CrossRef Search ADS PubMed  31. Coudert F. Responsive metal-organic frameworks and framework materials: under pressure, taking the heat, in the spotlight, with friends. Chem Mater  2015; 27: 1905– 16. https://doi.org/10.1021/acs.chemmater.5b00046 Google Scholar CrossRef Search ADS   32. Li C, Du M. Role of solvents in coordination supramolecular systems. Chem Commun  2011; 47: 5958– 72. https://doi.org/10.1039/c1cc10935a Google Scholar CrossRef Search ADS   33. Chang Z, Yang D, Xu J et al. Flexible metal-organic frameworks: recent advances and potential applications. Adv Mater  2015; 27: 5432– 41. https://doi.org/10.1002/adma.201501523 Google Scholar CrossRef Search ADS PubMed  34. Zhang JP, Zhou HL, Zhou DD et al. Dynamic behavior of porous coordination polymers. In: Atwood JL (ed). Comprehensive Supramolecular Chemistry II , Vol. 7. Oxford: Elsevier, 2017, 425– 74. Google Scholar CrossRef Search ADS   35. Li B, Wen H, Wang H et al. A porous metal-organic framework with dynamic pyrimidine groups exhibiting record high methane storage working capacity. J Am Chem Soc  2014; 136: 6207– 10. https://doi.org/10.1021/ja501810r Google Scholar CrossRef Search ADS PubMed  36. Wang G, Leus K, Couck S et al. Enhanced gas sorption and breathing properties of the new sulfone functionalized COMOC-2 metal organic framework. Dalton Trans  2016; 45: 9485– 91. https://doi.org/10.1039/C6DT01355D Google Scholar CrossRef Search ADS PubMed  37. Wieme J, Vanduyfhuys L, Rogge SMJ et al. Exploring the flexibility of MIL-47(V)-type materials using force field molecular dynamics simulations. J Phys Chem C  2016; 120: 14934– 47. https://doi.org/10.1021/acs.jpcc.6b04422 Google Scholar CrossRef Search ADS   38. Deng H, Grunder S, Cordova KE et al. Large-pore apertures in a series of metal-organic frameworks. Science  2012; 336: 1018– 23. https://doi.org/10.1126/science.1220131 Google Scholar CrossRef Search ADS PubMed  39. Mcdonald TM, Lee WR, Mason JA et al. Capture of carbon dioxide from air and flue gas in the alkylamine-appended metal-organic framework mmen-Mg2 (dobpdc). J Am Chem Soc  2012; 134: 7056– 65. https://doi.org/10.1021/ja300034j Google Scholar CrossRef Search ADS PubMed  40. Sung Cho H, Deng H, Miyasaka K et al. Extra adsorption and adsorbate superlattice formation in metal-organic frameworks. Nature  2015; 527: 503– 7. https://doi.org/10.1038/nature15734 Google Scholar CrossRef Search ADS PubMed  41. Jawahery S, Simon CM, Braun E et al. Adsorbate-induced lattice deformation in IRMOF-74 series. Nat Commun  2017; 8: 13945. https://doi.org/10.1038/ncomms13945 Google Scholar CrossRef Search ADS PubMed  42. Lee WR, Hwang SY, Ryu DW et al. Diamine-functionalized metal-organic framework: exceptionally high CO2 capacities from ambient air and flue gas, ultrafast CO2 uptake rate, and adsorption mechanism. Energy Environ Sci  2014; 7: 744– 51. https://doi.org/10.1039/C3EE42328J Google Scholar CrossRef Search ADS   43. Serre C, Mellot-Draznieks C, Surble S et al. Role of solvent-host interactions that lead to very large swelling of hybrid frameworks. Science  2007; 315: 1828– 31. https://doi.org/10.1126/science.1137975 Google Scholar CrossRef Search ADS PubMed  44. Horcajada P, Salles F, Wuttke S et al. How Linker's modification controls swelling properties of highly flexible Iron(III) dicarboxylates MIL-88. J Am Chem Soc  2011; 133: 17839– 47. https://doi.org/10.1021/ja206936e Google Scholar CrossRef Search ADS PubMed  45. Choi HJ, Dinca M, Long JR. Broadly hysteretic H2 adsorption in the microporous metal-organic framework Co(1,4-benzenedipyrazolate). J Am Chem Soc  2008; 130: 7848– 50. https://doi.org/10.1021/ja8024092 Google Scholar CrossRef Search ADS PubMed  46. Salles F, Maurin G, Serre C et al. Multistep N2 breathing in the metal-organic framework Co(1,4-benzenedipyrazolate). J Am Chem Soc  2010; 132: 13782– 8. https://doi.org/10.1021/ja104357r Google Scholar CrossRef Search ADS PubMed  47. Ma Y, Matsuda R, Sato H et al. A convenient strategy for designing a soft nanospace: an atomic exchange in a ligand with isostructural frameworks. J Am Chem Soc  2015; 137: 15825– 32. https://doi.org/10.1021/jacs.5b09666 Google Scholar CrossRef Search ADS PubMed  48. Taylor MK, Runcevski T, Oktawiec J et al. Tuning the adsorption-induced phase change in the flexible metal-organic framework Co(bdp). J Am Chem Soc  2016; 138: 15019– 26. https://doi.org/10.1021/jacs.6b09155 Google Scholar CrossRef Search ADS PubMed  49. Zhang J-P, Lin Y-Y, Zhang W-X et al. Temperature- or guest-induced drastic single-crystal-to-single-crystal transformations of a nanoporous coordination polymer. J Am Chem Soc  2005; 127: 14162– 3. https://doi.org/10.1021/ja054913a Google Scholar CrossRef Search ADS PubMed  50. Choi SB, Furukawa H, Nam HJ et al. Reversible interpenetration in a metal-organic framework triggered by ligand removal and addition. Angew Chem Int Ed  2012; 51: 8791– 5. https://doi.org/10.1002/anie.201202925 Google Scholar CrossRef Search ADS   51. Aggarwal H, Bhatt PM, Bezuidenhout CX et al. Direct evidence for single-crystal to single-crystal switching of degree of interpenetration in a metal-organic framework. J Am Chem Soc  2014; 136: 3776– 9. https://doi.org/10.1021/ja500530y Google Scholar CrossRef Search ADS PubMed  52. Ferguson A, Liu L, Tapperwijn SJ et al. Controlled partial interpenetration in metal-organic frameworks. Nat Chem  2016; 8: 250– 7. https://doi.org/10.1038/nchem.2430 Google Scholar CrossRef Search ADS PubMed  53. Zhou D-D, Liu Z-J, He C-T et al. Controlling the flexibility and single-crystal to single-crystal interpenetration reconstitution of metal-organic frameworks. Chem Commun  2015; 51: 12665– 8. https://doi.org/10.1039/C5CC05036G Google Scholar CrossRef Search ADS   54. Choi HJ, Dinca M, Dailly A et al. Hydrogenstorage in water-stable metal-organic frameworks incorporating 1,3- and 1,4-benzenedipyrazolate. Energy Environ Sci  2010; 3: 117– 23. https://doi.org/10.1039/B917512A Google Scholar CrossRef Search ADS   55. Goodwin AL, Chapman KW, Kepert CJ. Guest-dependent negative thermal expansion in nanoporous prussian blue analogues MII PtIV(CN)6 xH2O (0 ≤ x ≤ 2; M = Zn, Cd). J Am Chem Soc  2005; 127: 17980– 1. https://doi.org/10.1021/ja056460f Google Scholar CrossRef Search ADS PubMed  56. Goodwin AL, Kepert CJ. Negative thermal expansion and low-frequency modes in cyanide-bridged framework materials. Phys Rev B  2005; 71: 140301. https://doi.org/10.1103/PhysRevB.71.140301 Google Scholar CrossRef Search ADS   57. Serra-Crespo P, Gobechiya E, Ramos-Fernandez EV et al. Interplay of metal node and amine functionality in NH2-MIL-53: modulating breathing behavior through intra-framework interactions. Langmuir  2012; 28: 12916– 22. https://doi.org/10.1021/la302824j Google Scholar CrossRef Search ADS PubMed  58. He C-T, Ye Z-M, Xu Y-T et al. Hyperfine adjustment of flexible pore-surface pockets enables smart recognition of gas size and quadrupole moment. Chem Sci  2017; 8: 7560– 5. Google Scholar CrossRef Search ADS PubMed  59. Millange F, Guillou N, Walton RI et al. Effect of the nature of the metal on the breathing steps in MOFs with dynamic frameworks. Chem Commun  2008; 4732– 4. https://doi.org/10.1039/b809419e 60. Serre C, Millange F, Surble S et al. A route to the synthesis of trivalent transition-metal porous carboxylates with trimeric secondary building units. Angew Chem Int Ed  2004; 43: 6285– 9. https://doi.org/10.1002/anie.200454250 Google Scholar CrossRef Search ADS   61. Férey G, Mellot-Draznieks C, Serre C et al.   A chromium terephthalate-based solid with unusually large pore volumes and surface area. Science  2005; 309: 2040– 2. https://doi.org/10.1126/science.1116275 Google Scholar CrossRef Search ADS PubMed  62. Barthelet K, Marrot J, Férey G et al. VIII(OH){O2C-C6H4-CO2} (HO2C-C6H4-CO2H)x(DMF)y(H2O)z (or MIL-68), a new vanadocarboxylate with a large pore hybrid topology: reticular synthesis with infinite inorganic building blocks? Chem Commun  2004; 520– 1. https://doi.org/10.1039/B312589K 63. Wei Y-S, Chen K-J, Liao P-Q et al. Turning on the flexibility of isoreticular porous coordination frameworks for drastically tunable framework breathing and thermal expansion. Chem Sci  2013; 4: 1539– 46. https://doi.org/10.1039/c3sc22222e Google Scholar CrossRef Search ADS   64. Fukushima T, Horike S, Inubushi Y et al. Solid solutions of soft porous coordination polymers: fine-tuning of gas adsorption properties. Angew Chem Int Ed  2010; 49: 4820– 4. https://doi.org/10.1002/anie.201000989 Google Scholar CrossRef Search ADS   65. Nouar F, Devic T, Chevreau H et al. Tuning the breathing behaviour of MIL-53 by cation mixing. Chem Commun  2012; 48: 10237– 9. https://doi.org/10.1039/c2cc35348b Google Scholar CrossRef Search ADS   66. Horike S, Inubushi Y, Hori T et al. A solid solution approach to 2D coordination polymers for CH4/CO2 and CH4/C2H6 gas separation: equilibrium and kinetic studies. Chem Sci  2012; 3: 116– 20. https://doi.org/10.1039/C1SC00591J Google Scholar CrossRef Search ADS   67. Cohen SM. Postsynthetic methods for the functionalization of metal-organic frameworks. Chem Rev  2012; 112: 970– 1000. https://doi.org/10.1021/cr200179u Google Scholar CrossRef Search ADS PubMed  68. Wang Z, Cohen SM. Modulating metal-organic frameworks to breathe: a postsynthetic covalent modification approach. J Am Chem Soc  2009; 131: 16675– 7. https://doi.org/10.1021/ja907742z Google Scholar CrossRef Search ADS PubMed  69. Leclerc H, Devic T, Devautour-Vinot S et al. Influence of the oxidation state of the metal center on the flexibility and adsorption properties of a porous metal organic framework: MIL-47(V). J Phys Chem C  2011; 115: 19828– 40. https://doi.org/10.1021/jp206655y Google Scholar CrossRef Search ADS   70. Liao P-Q, Zhu A-X, Zhang W-X et al. Self-catalysed aerobic oxidization of organic linker in porous crystal for on-demand regulation of sorption behaviours. Nat Commun  2015; 6: 6350. https://doi.org/10.1038/ncomms7350 Google Scholar CrossRef Search ADS PubMed  71. Zhang J-P, Chen X-M. Exceptional framework flexibility and sorption behavior of a multifunctional porous cuprous triazolate framework. J Am Chem Soc  2008; 130: 6010– 7. https://doi.org/10.1021/ja800550a Google Scholar CrossRef Search ADS PubMed  72. Phillips AE, Goodwin AL, Halder GJ et al. Nanoporosity and exceptional negative thermal expansion in single-network cadmium cyanide. Angew Chem Int Ed  2008; 47: 1396– 9. https://doi.org/10.1002/anie.200704421 Google Scholar CrossRef Search ADS   73. Grobler I, Smith VJ, Bhatt PM et al. Tunable anisotropic thermal expansion of a porous Zinc(II) metal–organic framework. J Am Chem Soc  2013; 135: 6411– 4. https://doi.org/10.1021/ja401671p Google Scholar CrossRef Search ADS PubMed  74. Kim Y, Haldar R, Kim H et al. The guest-dependent thermal response of the flexible MOF Zn2(BDC)2(DABCO). Dalton Trans  2016; 45: 4187– 92. https://doi.org/10.1039/C5DT03710G Google Scholar CrossRef Search ADS PubMed  75. Zhou H-L, Lin R-B, He C-T et al. Direct visualization of a guest-triggered crystal deformation based on a flexible ultramicroporous framework. Nat Commun  2013; 4: 2534. https://doi.org/10.1038/ncomms3534 Google Scholar PubMed  76. Zhou H-L, Zhang Y-B, Zhang J-P et al. Supramolecular-jack-like guest in ultramicroporous crystal for exceptional thermal expansion behaviour. Nat Commun  2015; 6: 6917. https://doi.org/10.1038/ncomms7917 Google Scholar CrossRef Search ADS PubMed  77. Keene TD, Rankine D, Evans JD et al. Solvent-modified dynamic porosity in chiral 3D kagome frameworks. Dalton Trans  2013; 42: 7871– 9. https://doi.org/10.1039/c3dt00096f Google Scholar CrossRef Search ADS PubMed  78. Wei Y-S, Zhang M, Liao P-Q et al. Coordination templated [2+2+2] cyclotrimerization in a porous coordination framework. Nat Commun  2015; 6: 8348. https://doi.org/10.1038/ncomms9348 Google Scholar CrossRef Search ADS PubMed  79. Zhai Q-G, Bu X, Mao C et al. An ultra-tunable platform for molecular engineering of high-performance crystalline porous materials. Nat Commun  2016; 7: 13645. Google Scholar CrossRef Search ADS PubMed  80. Zhai Q, Bu X, Zhao X et al. Pore space partition in metal-organic frameworks. Acc Chem Res  2017; 50: 407– 17. https://doi.org/10.1021/acs.accounts.6b00526 Google Scholar CrossRef Search ADS PubMed  81. Chen Q, Chang Z, Song W et al. A controllable gate effect in Cobalt(II) organic frameworks by reversible structure transformations. Angew Chem Int Ed  2013; 52: 11550– 3. https://doi.org/10.1002/anie.201306304 Google Scholar CrossRef Search ADS   82. Wang H, Xu J, Zhang D et al. Crystalline capsules: metal-organic frameworks locked by size-matching ligand bolts. Angew Chem Int Ed  2015; 54: 5966– 70. https://doi.org/10.1002/anie.201500468 Google Scholar CrossRef Search ADS   83. Carrington EJ, Mcanally CA, Fletcher AJ et al. Solvent-switchable continuous-breathing behaviour in a diamondoid metal-organic framework and its influence on CO2 versus CH4 selectivity. Nat Chem  2017; 9: 882– 9. https://doi.org/10.1038/nchem.2747 Google Scholar CrossRef Search ADS PubMed  84. Yanai N, Kitayama K, Hijikata Y et al. Gas detection by structural variations of fluorescent guest molecules in a flexible porous coordination polymer. Nat Mater  2011; 10: 787– 93. https://doi.org/10.1038/nmat3104 Google Scholar CrossRef Search ADS PubMed  85. Aggarwal H, Das RK, Bhatt PM et al. Isolation of a structural intermediate during switching of degree of interpenetration in a metal-organic framework. Chem Sci  2015; 6: 4986– 92. https://doi.org/10.1039/C5SC01796C Google Scholar CrossRef Search ADS   86. Le Ouay B, Kitagawa S, Uemura T. Opening of an accessible microporosity in an otherwise nonporous metal-organic framework by polymeric guests. J Am Chem Soc  2017; 139: 7886– 92. https://doi.org/10.1021/jacs.7b02402 Google Scholar CrossRef Search ADS PubMed  87. Huang X-C, Zhang J-P, Chen X-M. [Zn(bim)2]·(H2O)1.67: a metal-organic open-framework with sodalite topology. Chinese Sci Bull  2003; 48: 1531– 4. Google Scholar CrossRef Search ADS   88. Gucuyener C, Van Den Bergh J, Gascon J et al. Ethane/ethene separation turned on its head: selective ethane adsorption on the metal-organic framework zif-7 through a gate-opening mechanism. J Am Chem Soc  2010; 132: 17704– 6. https://doi.org/10.1021/ja1089765 Google Scholar CrossRef Search ADS PubMed  89. Cai W, Lee T, Lee M et al. Thermal structural transitions and carbon dioxide adsorption properties of zeolitic imidazolate framework-7 (ZIF-7). J Am Chem Soc  2014; 136: 7961– 71. https://doi.org/10.1021/ja5016298 Google Scholar CrossRef Search ADS PubMed  90. Linder-Patton OM, Bloch WM, Coghlan CJ et al. Particle size effects in the kinetic trapping of a structurally-locked form of a flexible MOF. CrystEngComm  2016; 18: 4172– 9. https://doi.org/10.1039/C6CE00082G Google Scholar CrossRef Search ADS   91. Sakaida S, Otsubo K, Sakata O et al. Crystalline coordination framework endowed with dynamic gate-opening behaviour by being downsized to a thin film. Nat Chem  2016; 8: 377– 83. https://doi.org/10.1038/nchem.2469 Google Scholar CrossRef Search ADS PubMed  92. Zhang C, Gee JA, Sholl DS et al. Crystal-size-dependent structural transitions in nanoporous crystals: adsorption-induced transitions in ZIF-8. J Phys Chem C  2014; 118: 20727– 33. https://doi.org/10.1021/jp5081466 Google Scholar CrossRef Search ADS   93. Tanaka S, Fujita K, Miyake Y et al. Adsorption and diffusion phenomena in crystal size engineered ZIF-8 MOF. J Phys Chem C  2015; 119: 28430– 9. https://doi.org/10.1021/acs.jpcc.5b09520 Google Scholar CrossRef Search ADS   94. Tanaka D, Henke A, Albrecht K et al. Rapid preparation of flexible porous coordination polymer nanocrystals with accelerated guest adsorption kinetics. Nat Chem  2010; 2: 410– 6. https://doi.org/10.1038/nchem.627 Google Scholar CrossRef Search ADS PubMed  95. Hijikata Y, Horike S, Tanaka D et al. Differences of crystal structure and dynamics between a soft porous nanocrystal and a bulk crystal. Chem Commun  2011; 47: 7632– 4. https://doi.org/10.1039/c1cc10983a Google Scholar CrossRef Search ADS   96. Sakata Y, Furukawa S, Kondo M et al. Shape-memory nanopores induced in coordination frameworks by crystal downsizing. Science  2013; 339: 193– 6. https://doi.org/10.1126/science.1231451 Google Scholar CrossRef Search ADS PubMed  97. Kanoo P, Haldar R, Reddy SK et al. Crystal dynamics in multi-stimuli-responsive entangled metal-organic frameworks. Chem Eur J  2016; 22: 15864– 73. https://doi.org/10.1002/chem.201602087 Google Scholar CrossRef Search ADS PubMed  98. Engel ER, Jouaiti A, Bezuidenhout CX et al. Activation-dependent breathing in a flexible metal-organic framework and the effects of repeated sorption/desorption cycling. Angew Chem Int Ed  2017; 56: 8874– 8. https://doi.org/10.1002/anie.201704044 Google Scholar CrossRef Search ADS   99. Yanai N, Uemura T, Inoue M et al. Guest-to-host transmission of structural changes for stimuli-responsive adsorption property. J Am Chem Soc  2012; 134: 4501– 4. https://doi.org/10.1021/ja2115713 Google Scholar CrossRef Search ADS PubMed  100. Ma S, Sun D, Wang X et al. A mesh-adjustable molecular sieve for general use in gas separation. Angew Chem Int Ed  2007; 46: 2458– 62. https://doi.org/10.1002/anie.200604353 Google Scholar CrossRef Search ADS   101. Mason JA, Oktawiec J, Taylor MK et al. Methane storage in flexible metal-organic frameworks with intrinsic thermal management. Nature  2015; 527: 357– 61. https://doi.org/10.1038/nature15732 Google Scholar CrossRef Search ADS PubMed  102. Ghoufi A, Benhamed K, Boukli-Hacene L et al. Electrically induced breathing of the MIL-53(Cr) metal-organic framework. ACS Cent Sci  2017; 3: 394– 8. https://doi.org/10.1021/acscentsci.6b00392 Google Scholar CrossRef Search ADS PubMed  103. Gao Q, Xu J, Cao D et al. A rigid nested metal-organic framework featuring a thermoresponsive gating effect dominated by counterions. Angew Chem Int Ed  2016; 55: 15027– 30. https://doi.org/10.1002/anie.201608250 Google Scholar CrossRef Search ADS   104. Serra-Crespo P, Dikhtiarenko A, Stavitski E et al. Experimental evidence of negative linear compressibility in the MIL-53 metal-organic framework family. CrystEngComm  2015; 17: 276– 80. https://doi.org/10.1039/C4CE00436A Google Scholar CrossRef Search ADS PubMed  105. Spencer EC, Angel RJ, Ross NL et al. Pressure-induced cooperative bond rearrangement in a zinc imidazolate framework: a high-pressure single-crystal X-Ray diffraction study. J Am Chem Soc  2009; 131: 4022– 6. https://doi.org/10.1021/ja808531m Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of China Science Publishing & Media Ltd. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png National Science Review Oxford University Press

Controlling flexibility of metal–organic frameworks

Loading next page...
 
/lp/ou_press/controlling-flexibility-of-metal-organic-frameworks-rXVGUSAUCm
Publisher
Oxford University Press
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of China Science Publishing & Media Ltd. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
2095-5138
eISSN
2053-714X
D.O.I.
10.1093/nsr/nwx127
Publisher site
See Article on Publisher Site

Abstract

Abstract Framework flexibility is one of the most important characteristics of metal–organic frameworks (MOFs), which is not only interesting, but also useful for a variety of applications. Designing, tailoring or controlling framework flexibility of MOFs is much more difficult than for static structural features such as the framework topology and pore size/shape. Nevertheless, with in-depth understanding of the relationship between framework flexibility and the host framework structure, guest loading and other aspects such as the crystal size/morphology and external physical environment, some strategies have been developed for controlling the flexibility of MOFs and the corresponding properties, which are summarized and discussed in this review. porous coordination polymers, metal–organic frameworks, flexibility, dynamism, adsorption, structural transformation INTRODUCTION Porous coordination polymers (PCPs), or more commonly quoted as metal–organic frameworks (MOFs), are highly ordered polymeric open network structures based on coordination bonding connections [1,2]. By virtue of the vast diversity of metal ions and organic ligands, as well as a couple of successful crystal engineering principles, MOFs are well known for their modular designability, enabling the framework topology, pore size/shape and pore surface characteristics to be constructed on demand [2]. Besides structures and applications [3], MOFs are also famous for their flexible structures responsive to external stimuli, such as guest change, temperature change, light irradiation and mechanical forces, which have attracted extensive interests especially after the pioneering works by Kitagawa and Férey [4–9]. In most MOFs, different parts of the host framework can rotate around single bonds within an organic ligand and/or between a metal ion and an organic ligand. The reversible nature of coordination bonds and weaker supramolecular interactions also enables the internal connections within the MOF crystals to reversibly break and reform. These structural features give rise to multiple, easily accessible thermodynamic metastable states. Consequently, flexible MOFs can show diverse types and magnitudes of structural dynamism. Importantly, many of the structure changes occur with the retention of the periodicity and even the single-crystallinity, which enable the transformed structures and host–guest interactions to be visualized at the atomic level by crystallography techniques including the convenient in-house single-crystal X-ray diffraction [10–13]. Compared with rigid adsorbents, flexible MOFs can have advantages for many applications such as storage and separation. Flexible MOFs usually show S-shaped or multi-step adsorption isotherms [14], which can be very useful for practical gas storage/delivery because these shapes can enlarge the uptake difference between the charge and discharge pressures [15,16]. Further, the structural transformation of a MOF may consume/provide the heat generated/required during the adsorption/desorption process, respectively, reducing the requirement of fast heat exchange between the container and environment [16,17]. While rigid adsorbents distinguish different guest species by either the differential binding (matching of size, shape and functional group) or the size exclusion (molecular sieving) mechanism, flexible MOFs can simultaneously utilize all available mechanisms and alter their structures to emphasize the differences of guests [18–20]. For example, the gate-opening pressure of adsorption depends on both the binding affinity and molecular size of the guest. Therefore, very large uptake differences similar to the molecular sieving effect can be easily observed for flexible MOFs [21,22]. Unlike rigid molecular sieves, precise adjustment of the aperture size is not necessary for flexible MOFs. The changeable framework structures and pore sizes/shapes of flexible MOFs can be also disadvantages. For example, SOD-[Zn(mim)2] (MAF-4/ZIF-8, Hmim = 2-methylimidazole) [23], possessing well-defined small pore aperture size (3.2 Å [24]) and originally thought to be rigid [25], had been expected to show good molecular sieving effect but the experimental results were disappointed [26], because it is actually flexible and can readily adsorb much larger molecules such as toluene and p-xylene [27,28]. Accompanying the rapidly growing research on flexible MOFs, many review articles focusing on their interesting phenomena, useful properties, characterization methods, etc., have been published in the past few years [8–10,29–34]. Considering the critical roles of framework flexibility for related applications, this review intends to discuss strategies for controlling the flexibility of MOFs. It should be noted that flexible MOFs have intrinsic abilities to show different structural transformations or dynamic behaviors toward different external stimuli, which should be regarded as controlling/changing structure rather than controlling/changing flexibility. For clarity, controlling flexibility of MOFs is defined as changing structural response or dynamic behavior of a MOF sample toward a given external stimulus, upon changing the structure of the MOF sample and/or changing its external environment (Fig. 1). Obviously, unlike static structural features (such as the framework topology and pore size/shape), controlling framework flexibility is much more difficult. Actually, very different flexibility can be observed for MOFs showing trivial structural differences [35]. And drastic structural transformations, especially those involving breakage/reformation of coordination bonds, are generally serendipitous. Figure 1. View largeDownload slide Controlling flexibility of MOFs: from observing different structural responses toward different external stimuli to tuning structural response toward a given external stimulus. Figure 1. View largeDownload slide Controlling flexibility of MOFs: from observing different structural responses toward different external stimuli to tuning structural response toward a given external stimulus. In principle, framework flexibility is determined by the structure of the MOF sample and/or the external environment. The structure of a MOF sample has three levels of meanings, namely the host coordination network structure, the crystal structure and the particle structure, in which each definition is a subset of latter ones. The host coordination network structure, or simply the host framework structure, is the basic/fundamental component for the structure–property relationship, because MOF applications usually involve host–guest interplays, and most studies have been devoted to designing/modifying the host framework structure. For some applications, the structure of a MOF sample needs to be described by its crystal structure, which consists of not only the host framework, but also the guest included in the crystal. Actually, some components inside the MOF crystals, such as a hardly removable molecule reinforced by coordination bonding or other interactions, can be defined as either the host or the guest, depending on the experimental design. The particle structure of a MOF sample further includes the size, morphology, defect and surface modification of the crystal sample, which are all effective for the sample property but usually not regarded as the intrinsic structural parameters of MOFs. Outside the MOF sample, temperature, pressure and some other physical parameters can not only drive the structural transformation of the MOF, but also modulate the MOF flexibility toward other stimuli. A strategy for controlling MOF flexibility can rely on the host, guest, particle characters and/or external physical environment, in which the concept gradually changes from designing new materials to changing property of existing materials (Fig. 1). In this context, this review summarizes and discusses representative examples and strategies for designing/tailoring/controlling flexibility of MOFs in the three sections, focusing on the host framework structure, the guest loading and the other means including the particle characters and external physical environment. CONTROLLING FLEXIBILITY BY FRAMEWORK DESIGN Although flexible MOFs can be constructed by using flexible organic ligands, by connecting metal ions and organic ligands with single coordination bonds or by designing low-dimensional or interpenetrating coordination networks, these structural features have poor designability/controllability. On the other hand, truly designable MOFs are generally based on some well-defined framework prototypes/topologies and less flexible building units. Nevertheless, the rigidity/flexibility of these designable MOFs can be rationally controlled by judicious selections of organic ligands and metal ions/clusters. Besides the rigid/flexible backbones, the bridging length and non-coordinative side groups of organic ligands can play important roles to the framework rigidity/flexibility of MOFs. The long backbone of organic ligands can bend more easily, giving rise to more flexible MOFs [36,37]. [M2(dobdc)] (MOF-74/CPO-27, H4dobdc = 2,5-dioxido-1,4-benzenedicarboxylic acid) is a typical rigid MOF structure with 3D honeycomb-like coordination network and 1D channels [38]. Expanded versions of MOF-74 with mesopores can be rationally constructed by lengthening the ligand backbone (inserting 1–10 phenyl rings) [38,39]. Similar to other mesoporous materials, the expanded versions of MOF-74 can show a multi-step gas-adsorption isotherm. Interestingly, small-angle X-ray scattering demonstrated the formation of a superlattice structure associated with the multi-step isotherm, which was ascribed to the special location of gas molecules [40]. On the other hand, computational simulations suggested that framework distortion at the inorganic chain was the origin of a superlattice [41]. Actually, X-ray single-crystal and powder diffraction analyses showed that adding one more phenyl ring to the ligand backbone of MOF-74 is enough to induce framework flexibility with significant and distinct contraction/expansion toward N,N-dimethylformamide (DMF), ethylenediamine and CO2, in which the ligand backbone bends differently [39,42]. Long organic ligands can increase framework flexibility without backbone bending. Férey et al. showed that the breathing amplitude of acs-[M3(μ3-O)(ldc)3(LT)3] (MIL-88, H2ldc = linear dicarboxylic acid, LT = terminal ligand) can significantly increase along with the length of the linear dicarboxylate ligand (e.g. 125% for MIL-88B based on 1,4-benzenedicarboxylate and 230% for MIL-88D based on 4,4΄-biphenyldicarboxylate) [43]. Crystal-structure analyses showed that the framework breathing is mainly originated from bending of the M2=O2>C junctions rather than other parts. Because the bridging directions of the organic ligands do not point to the exact center of the trinuclear cluster, the shape/volume change of the network is smaller than that of the metal–carboxylate junction, which is more serious for the shorter dicarboxylate ligands. While changing the bridging length of organic ligand is usually regarded as designing new MOFs, changing the non-coordinative side group is somewhat closer to the concept of modification of known MOFs. For framework flexibility, non-coordinative side groups can play steric (repulsive), attractive and/or electronic roles [44]. [Co(bdp)] (H2bdp = 1,4-benzenedipyrazole) reported by Long et al. is a highly flexible (160% volume increase from guest-free to guest-saturated form) pillared-rod structure [45,46]. At 298 K, this MOF shows a gate-opening-type adsorption for CH4, which is critical for achieving the remarkably high working capacity of 155 cm3 cm–3 for CH4 storage at 5–35 bar [16]. To tune the gate-opening pressure, they further compared the gas-adsorption properties and corresponding framework transformations of a series of isostructural [Co(Rbdp)] frameworks, which demonstrated important roles for the electronic, steric and positional effects of the side groups. For CH4 adsorption, introduction of electron withdrawing F as a side group can reduce the gate-opening pressure [47], while methyls capable of forming stronger edge-to-face π−π interactions can increase the gate-opening pressure [48]. The attractive/repulsive forces of non-coordinative side groups toward guest species can be also utilized to control framework flexibility. Compared with other forms of framework flexibility, breakage/formation of coordination bonds is more difficult to design/control, as exemplified by the scarce examples of the change of the net interpenetration number of MOFs [49–52]. In a special case, namely the inter-conversion between 5-fold and 6-fold interpenetrated dia-f networks of [Ag6X(Rtz)4]OH·nH2O (X = Cl− or Br−; HRtz = Hatz/Hmtz = 3-amino-1,2,4-triazole/3-methyl-1,2,4-triazole) [49], the framework flexibility can be controlled by the hydrophlicity/hydrophobicity of the ligand side groups, which controls the guest accessibility of the metal ions [53]. When hydrophilic amino groups were employed to allow the guest OH–/H2O to attack the linearly coordinated Ag(I) ions that help the formation of low-energy intermediates during the bond-breaking/reformation processes, both [Ag6Cl(atz)4]OH·nH2O and [Ag6Br(atz)4]OH·nH2O can simultaneously show interpenetration reconstitution and remarkable breathing. On the other hand, both [Ag6Cl(mtz)4]OH·nH2O and [Ag6Br(mtz)4]OH·nH2O with hydrophobic methyl groups prevent such attacks and only show large framework breathing (Fig. 2). Figure 2. View largeDownload slide Controlling the framework breathing/reconstitution behaviors of [Ag6X(Rtz)4]OH·nH2O. Left: simultaneous framework distortion and interpenetration reconstitution (between 5- and 6-fold) for R = NH2. Right: framework distortion without interpenetration reconstitution for R = CH3. Adapted from [53] with permission of the Royal Society of Chemistry. Figure 2. View largeDownload slide Controlling the framework breathing/reconstitution behaviors of [Ag6X(Rtz)4]OH·nH2O. Left: simultaneous framework distortion and interpenetration reconstitution (between 5- and 6-fold) for R = NH2. Right: framework distortion without interpenetration reconstitution for R = CH3. Adapted from [53] with permission of the Royal Society of Chemistry. As a single atom, a metal ion can induce framework flexibility by the structural diversity of its coordination sphere, including the change of coordination bond length, coordination geometry and coordination number. The change of coordination bond length occurs for a few special metal ions in some special cases (e.g. spin crossover). The change of coordination number, either permanent removal/addition of coordinative guests or transient breaking/reforming coordination bonds, is the main origin for many structural transformation phenomena. There are very few metal ions capable of drastically changing coordination geometry without simultaneous alternation of the coordination number. The large breathing ability of [Co(bdp)], in which the Co(II) ions possess the normal tetrahedral geometry in the open form, was attributed to the abnormal square-planar geometry in the guest-free form, as demonstrated by molecular-simulation-assisted analyses of synchrotron powder diffraction, magnetic measurements, and IR and UV–Vis–NIR spectroscopies [45,46]. Many MOF structures can adopt different metal ions via direct synthesis or post-synthetic ion exchange, and alternation of metal ion can readily modulate the framework flexibility. For example, being isostructural with [Co(bdp)], [Zn(bdp)] and [Fe(bdp)] can show similar breathing motions, but their breathing amplitudes and/or involved energy changes are different and can be used to tune the gas-storage performance [16,54]. Besides preferences of coordination geometries, larger metal ions (i.e. longer coordination bonds) may facilitate structural transformation due to the reduced steric hindrance effect [55–57]. For example, N2, H2, C2H2 and CO2 gas-adsorption measurements and multi-level computational simulations for a series of isostructural ultramicroporous MOFs, [Zn3(vtz)6] (MAF-123-Zn, Hvtz = 1,2,3-triazole), [Mn3(vtz)6] (MAF-123-Mn) and [Cd3(vtz)6] (MAF-123-Cd), demonstrated that a 0.1-Å increase in the metal ion radius (0.9, 1.0 and 1.1 Å for divalent octahedral Zn, Mn and Cd ions, respectively) can readily turn on the framework flexibility to give multi-step or S-shaped isotherms useful for gas storage/delivery [58]. Nevertheless, in most other cases, quantitative correlation of the differences of metal ions and framework flexibilities still needs further investigation, as exemplified by the complicated breathing trends of [M(bdp)] [16,54] and [M(μ-OH)(bdc)] (MIL-53, H2bdc = 1,4-benzenedicarboxylic acid) [59]. Compared with individual metal ions, polynuclear metal clusters/chains are more suitable for rational design of the framework structure and the control of framework flexibility. Metal clusters/chains used for MOF design/construction are usually terminated by carboxylate groups with two M–O bonds. Because the carboxylate oxygen atoms have very weak coordination bonding directionality, the M2=O2>C junction can usually bend to a large extent like a knee [43]. Nevertheless, the symmetries of such clusters/chains and their arrangements are also determinative for framework rigidity/flexibility [8,34]. Zn4(μ4-O)(RCOO)6 and M3(μ3-O)(RCOO)6(LT)3 are typical examples of rigid and flexible clusters, in which the octahedral and the trigonal-prismatic arrangements of metal–carboxylate junctions are the key difference, and only the latter one allows the six bending vectors to operate in parallel. Nevertheless, with the same [M3(μ3-O)(ldc)3(LT)3] composition, MIL-88 with a six-connected acs topology and MIL-101 with a six-connected mtn-e topology are highly flexible and rigid [60,61], because the clusters are arranged in parallel in a trigonal symmetry or isotropically in a cubic symmetry, respectively. Similar phenomena have also been observed in MOFs based on metal–carboxlyate chains [5,62]. The flexibility mode of M3(μ3-O)(RCOO)6(LT)3 clusters can be rationally tuned/controlled by framework design. In the free M3(μ3-O)(RCOO)6(LT)3 clusters, the metal–carboxylate junctions adopt the linear conformation. In MIL-101 and MIL-88, as well as other MOFs containing such clusters, the metal–carboxylate junctions all bend toward the 3-fold axis of the clusters in different degrees. As the less bended metal–carboxylate junction can provide a larger framework volume, the trigonal framework MIL-88 shortens along the c-axis and expands in the ab-plane to accommodate more guests [43]. By using tripodal pyridyldicarboxylate ligands with pyridyl ends to further occupy the three monodentate terminal (LT) sites at the equatorial plane of the M3(μ3-O)(RCOO)6(LT)3 cluster, a (3,9)-connected xmz network [M3(μ3-O)(dcpb)3] (MCF-18, H2dcpb = 2,6-di-p-carboxyphenyl-4,4΄-bipyridine) can be obtained [63]. MCF-18 adopts a trigonal symmetry with the trinuclear clusters arranged in the same manner as for MIL-88, and it is also highly flexible. However, MCF-18 expands along the c-axis and contracts in the ab-plane to accommodate more guests, being reversed in the manner of MIL-88. Consequently, although MCF-18 exhibits much smaller volumetric breathing amplitude (105%) than MIL-88 (230%), its axial breathing amplitude (121%) is larger than for MIL-88 (100%) [43]. Careful structural examination showed that the metal–carboxylate junctions in MCF-18 all bend toward the equatorial plane, which accounts for its special breathing direction (Fig. 3). The unique shape of the dcpb2– ligand is responsible for the special metal–carboxylate conformation in MCF-18. Some MOFs isoreticular with MCF-18 can be constructed by using other tripodal pyridyldicarboxylate ligands, but they are all rigid and their metal–carboxylate junctions all bend toward the 3-fold axis [63]. Figure 3. View largeDownload slide Two types of distortion manners of the classic M3(μ3-O)(ldc)3(LT)3 cluster, leading to the reversed breathing directions for MIL-88 and MCF-18. Adapted from [10] and [63] with permission of the Royal Society of Chemistry. Figure 3. View largeDownload slide Two types of distortion manners of the classic M3(μ3-O)(ldc)3(LT)3 cluster, leading to the reversed breathing directions for MIL-88 and MCF-18. Adapted from [10] and [63] with permission of the Royal Society of Chemistry. Since framework flexibility is highly sensitive to trivial structural difference, changing components of isostructural/isoreticular MOFs can readily alter the framework flexibility but can hardly tune it to meet the requirement of desired application. Fortunately, similarly to other inorganic crystalline materials, MOFs can also adopt two or more analogous building blocks at the same crystallographic position with random distribution in different unit cells, giving solid-solution structures with variable stoichiometries to enable continuous adjustment of their structures and properties [24,64,65]. Kitagawa et al. demonstrated the usefulness of the solid-solution strategy for on-demand tuning of framework flexibility, by using two isostructural MOFs [Zn(5-NO2-ip)(bpy)] (CID-5, bpy = 4,4΄-bipyridyl, 5-NO2-H2ip = 5-nitroisophthalic acid) and [Zn(5-MeO-ip)(bpy)] (CID-6, 5-MeO-H2ip = 5-methoxyisophthalic acid) [64]. With a smaller substituent group, CID-5 is flexible and shrinks to a nonporous state that can adsorb CO2 only above a relatively high gate-opening pressure, and cannot adsorb CH4 up to 1.0 MPa. On the other hand, CID-6 with a larger substituent group is rigid and can adsorb both CO2 and CH4. By mixing the two ligands with different ratios, isostructural solid-solution frameworks [Zn(5-NO2-ip)x(5-MeO-ip)1−x(bpy)] (CID-5/6; x = 0.06-0.82) were obtained, showing tunable gate-opening pressure for CO2 adsorption. More importantly, the flexibility can be tuned to a state suitable for CO2/CH4 separation, at which the framework can be opened by CO2 and adsorb large amounts of CO2, while remaining repellent to CH4. Further breakthrough experiments showed that this strategy is useful for not only CO2/CH4 separation, but also C2H6/CH4 [66]. While the above-discussed strategies and examples can be still regarded as designing and synthesizing new materials, post-synthetic modification is a method closer to the purpose of controlling or tailoring structures/properties of existing materials [67]. Cohen et al. modified [Zn2(NH2-bdc)2(dabco)] (DMOF-1, NH2-H2bdc = 2-amino-1,4-benzenedicarboxylic acid, dabco = 1,4-diazabicyclo[2.2.2]octane) by reacting it with alkyl anhydrides (O(CO(CH2)nCH3)2, n = 0∼5) in CHCl3, giving a series of MOF crystals with different flexibility depending on the length of the appended alkyl chain [68]. DMOF-1 is a rigid framework showing normal type-I adsorption for N2, Ar and CO2. After modification, the crystals shrink after desolvation for n = 0∼2 due to attractive interactions between the alkyl chains, but remain rigid for n ≥ 3 because of steric hindrance effect. Further, gate-opening-type gas-adsorption isotherms were observed for n = 1 and 2 rather than n = 0. Obviously, precise control over the composition of a solid-solution framework, by either direct synthesis or post-synthetic modification, is critical for on-demand control of its property. However, the product compositions are generally different from the feeding ratios, and most post-synthetic modification reactions involve liquid reactants/reagents, preventing precise control/monitoring of the reaction/modification degrees [67]. This problem may be solved by using a solvent-free solid-gas reaction mechanism [69]. For example, we reported a flexible metal azolate framework [Cu4(btm)2] (MAF-42, H2btm = bis(5-methyl-1,2,4-triazol-3-yl)methane) consisting of Cu(I) ions and a methylene-bridged bistriazolate ligand [70]. Because the Cu(I) ions are two- or three-coordinated by the triazolate N donors being similar to those of the copper proteins and the methylene group activated by two aromatic rings are adjacent to these low-coordinated Cu(I) ions, the crystal can be oxidized to form [Cu4(btk)2] (H2btk = bis(5-methyl-1,2,4-triazol-3-yl)methanone) by O2 or air even at room temperature, with H2O as the only byproduct (Fig. 4). The carbonyl-bridged ligand is more rigid and more hydrophilic than the methylene-bridged one, leading to higher gate-opening pressure and larger adsorption hysteresis, as well as larger pore volumes. More importantly, without using any liquids, the oxidation degree can be simply monitored by the sample weight, so that the framework flexibility and surface characteristic can be tuned on demand. As a result, the CH4, C2H6 and CO2 adsorption selectivities can be drastically changed up to four orders in magnitude and even inversed. Figure 4. View largeDownload slide Post-synthetic modification and controlling flexibility of MAF-42. (a) Guest-induced framework breathing of MAF-42. (b) The solvent-free solid–gas reaction mechanism for post-synthetic modification. (c) Increasing framework rigidity and breathing amplitude by changing the flexible methylene groups to rigid carbonyl groups. Adapted from [70] with permission of Nature Publishing Group. Figure 4. View largeDownload slide Post-synthetic modification and controlling flexibility of MAF-42. (a) Guest-induced framework breathing of MAF-42. (b) The solvent-free solid–gas reaction mechanism for post-synthetic modification. (c) Increasing framework rigidity and breathing amplitude by changing the flexible methylene groups to rigid carbonyl groups. Adapted from [70] with permission of Nature Publishing Group. CONTROLLING FRAMEWORK FLEXIBILITY BY GUEST Changing guest loading in MOF crystals can be regarded as a special type of post-synthetic modification method. It is well known that the structures of flexible MOFs are highly dependent on the type and amount of the adsorbed guest species. Also, the manner of MOF structure change is always dependent on the type of guest, as indicated by their different isotherm shapes, breathing amplitudes and even breathing modes [43,71]. Nevertheless, these phenomena should not be regarded as controlling the flexibility of MOFs (Fig. 1). On the other hand, guests can be used to rationally modulate not only the physical properties (such as magnetism and luminescence), but also the framework flexibility of MOFs induced by physical stimuli such as changes in temperature and/or pressure. Guest-modulated thermal expansion of MOFs has been well demonstrated (Fig. 5), although guest-induced framework breathing is much larger than that induced by temperature change and it is usually difficult to keep constant guest-loading amounts at different temperatures. In this context, ultramicroporous MOFs have shown great potential, since the small pores strongly interact with guest molecules and restrict guest escape. Usually, the steric hindrance of guest molecules and host–guest attractive interactions are responsible for the change in thermal expansion coefficients, which can impede or promote the contraction of the host framework at low temperatures, giving smaller and larger thermal expansion coefficients, respectively [55,72–74]. Figure 5. View largeDownload slide Strategies for utilizing guests to control the thermal expansion behaviors of MOFs. Figure 5. View largeDownload slide Strategies for utilizing guests to control the thermal expansion behaviors of MOFs. The dynamic behaviors of the guest molecules, such as the rotation of a molecule and thermal expansion of a guest cluster, can be also used to control the thermal expansion properties of MOF crystals. For example, we designed and synthesized an ultramicroporous MOF [Mn(pba34)2] (MCF-34, Hpba34 = 3-(pyridin-4-yl)benzoic acid) by using a bent ligand with easily changeable conformation [75]. Because its 1D channels (Φmin/max = 3.9/4.4 Å) are too small for N2/O2 adsorption at low temperatures, it exhibits constant and huge thermal expansion (α = +224 × 10−6 K−1/–107 × 10−6 K−1) over a wide temperature range (127–673 K) in both vacuum and air. More interestingly, when DMF is included in MCF-34, the thermal expansion profile of the crystal showed an abrupt jumping around the melting point of the guest, because the rotation motion of DMF triggers the conformation reversion of the organic ligand, which further transmits this action to the whole crystal. To utilize the thermal expansion of guest clusters, we further synthesized a 3D hinge-like framework [Cd(pba34)(pba44)] (MCF-82, Hpba44 = 4-(pyridin-4-yl)benzoic acid) with quasi-discrete ultramicropores running along its a-axis, which allow DMF and N,N-dimethylacetamide (DMA) supramolecular dimers to be strongly confined yet exchangeable. The vacant crystal of MCF-82 shows extremely large positive/negative thermal expansion across its bc-plane (α = +482 × 10−6 K−1/–218 × 10−6 K−1) [76]. Meanwhile, a small contraction was observed for the a-axis during temperature increase. The guest-included crystals (especially for DMF) not only show largely reduced thermal expansion across the bc-plane, but also large expansion along the a-axis during temperature increase. Single-crystal X-ray diffraction indicated that the confined guest dimers impede the framework deformation across the bc-plane via the steric hindrance effect. Moreover, the DMF dimers form close π–π stacking interactions, which display overwhelmingly large thermal expansion along the a-axis to change the thermal expansion anisotropy of the MOF crystal. On the other hand, the DMA dimers form weak edge-to-edge interaction and small thermal expansion, so that its effect is much weaker than the DMF dimers. Alternation of guest inclusion can also modulate the dynamic response of flexible MOFs toward adsorption of other guests. Because adsorption measurements generally require degassing treatments by heating and/or vacuuming, strong host–guest binding (e.g. via coordination bonds) is necessary for the guest to modulate the flexibility and adsorption property of the MOF crystal [77–83]. Recently, Brammer et al. showed that desolvation of Me2NH2[In(abdc)2] (SHF-61, H2abdc = 2-aminobenzene-1,4-dicarboxylic acid) in two different solvents leads to two polymorphic-activated forms with very different pore openings and markedly different gas-adsorption properties (Fig. 6) [83]. With CHCl3 as the low-polarity solvent, desolvation leads to little change in the host framework, which exhibits a high CO2-adsorption capacity and a normal type-I isotherm shape. With DMF/H2O as the high-polarity solvent, desolvation leads to significant framework contraction, which also exhibits a normal type-I isotherm shape but with a low CO2-adsorption capacity. Interestingly, the framework contraction during desolvation of DMF/H2O is continuous, and partial desolvation produces a partially contracted framework, which exhibits a gating pressure associated with CO2 adsorption. Such interesting phenomena were ascribed to the different strengths of host–guest interactions induced by the solvent molecules, in which the high-polarity solvent helps the host framework to contact but can be broken by CO2 at high pressure [83]. Figure 6. View largeDownload slide Proposed roles of solvent-controlled framework dynamism and the CO2 adsorption property of SHF-61. Figure 6. View largeDownload slide Proposed roles of solvent-controlled framework dynamism and the CO2 adsorption property of SHF-61. Guest molecules with large molecular weights can form strong host–guest interactions, which are useful for controlling the guest loading for further modulating the MOF properties [84]. For example, [Co2(ndc)2(bpy)] (H2ndc = 2,6-napthalenedicarboxylic acid) is a 2-fold interpenetrated pcu network showing interesting flexibility, which simply shrinks to a less porous state upon guest removal at moderate temperature, and can further transform to a 3-fold interpenetrated nonporous structure by heating at higher temperature [85]. Uemura et al. showed that, after loading polystyrene, [Co2(ndc)2(bpy)] can keep the as-synthesized open structure after removing small guest molecules, and prevent the transformation to the 3-fold interpenetrated structure at higher temperature [86]. The polymer-included samples were prepared by adsorption of styrene monomer, polymerization and then vacuum removal of the residue styrene monomer. Obviously, the long-chain polymer guest can be hardly removed from the host using conventional MOF activation methods, which enable the precise control of guest loading. Propane and CO2 adsorption isotherms were measured for the MOF loaded with different amounts of polystyrene, which showed uptake trends following 5% loading >16% loading >0% loading. Obviously, an appropriate loading of polystyrene can increase the pore volume by expanding the framework from the less porous state to the most porous state, while not occupying too much of the pore. CONTROLLING FRAMEWORK FLEXIBILITY BY OTHER MEANS While the structure–property relationship is regarded as one of most important aspects for chemistry, the structural information of a MOF is generally referred to its chemical composition and crystal structure only. However, other particle parameters, such as size, morphology and surface modification, can also play important roles, being similar to conventional inorganic nanomaterials. For example, SOD-[Zn(bim)2] (MAF-3/ZIF-7, Hbim = benzimidazole [87]) is a flexible structure showing gate-opening-type hysteretic sorption isotherms for a variety of gases [25,88]. Choi et al. showed that MAF-3 with differing shapes and sizes (∼100 nm spherical, ∼400 nm rhombic-dodecahedral and ∼1300 nm rod-shaped) exhibit different N2 and CO2 sorption behaviors, including uptakes, gate-opening pressures and hysteresis widths [89]. In particular, the spherical sample can start to adsorb gases at very low pressure, which was ascribed to the possible presence of disordered structures near the outer shell with more flexibility than the core MAF-3 structure. Crystal downsizing can generally accelerate guest diffusion and hence facilitate structural transformation of flexible MOFs [90], and even make rigid frameworks flexible. For example, Kitagawa et al. reported guest adsorption in nonporous [Pt(CN)4Fe(py)2] (py = pyridine) enabled by crystal downsizing [91]. In [Pt(CN)4Fe(py)2], metal ions are bridged by cyanide ligands to form 2D layers, which interdigitate together using the monodentate pyridine ligands to form a nonporous structure. In the bulk state (particle size 135 nm), [Pt(CN)4Fe(py)2] cannot adsorb EtOH. After simple mechanical grinding (particle size 9 nm), it can adsorb 0.67 EtOH per formula unit. In situ powder X-ray diffraction analyses showed obvious shifting of diffraction peaks, indicating that the guest adsorption originates from the structural transformation of the host. Crystal downsizing can also impede structural transformation of flexible MOFs. As a representative flexible MOF, MAF-4 exhibits two-step N2 adsorption isotherm, and the second step starts at higher pressures for the smaller crystals [24,25,92,93]. Nevertheless, the relationship between crystal size and framework flexibility was discovered first by Kitagawa et al. [94–96]. [Zn(ip)(bpy)]·DMF (CID-1, H2ip = isophthalic acid) is an interdigitated stacking structure of 2D layers [94]. For bulk CID-1 with crystal sizes ∼5 μm × 20 μm, desolvation under vacuum at 130°C led to slight framework shrinkage, as indicated by the shifting of several powder X-ray diffraction peaks to the higher 2θ values. When the crystal size was reduced to nanoscale (500 × 100 × 30 nm3), the desolvation-induced shrinking reduced a lot, which was attributed to the change in crystal size or crystallite surface structure (covered by surfactant). Single-crystal X-ray and synchrotron powder diffraction analyses showed that bulk and nano CID-1 undergo 3.2% and 2.1% reductions of the unit-cell volume after desolvation, respectively [95]. Kitagawa et al. further developed a shape memory effect by downsizing the MOF crystal [96]. [Cu2(bdc)2(bpy)] is a 2-fold interpenetrated pcu network containing a notable porosity (void = 35%). A series of samples with different crystal sizes (methanol as guest) were synthesized. After guest removal under vacuum at room temperature, the sample with 300 × 300 × 30 nm3 or larger size shrink to a narrow-pore state (void = 20%). Surprisingly, the smaller nanoscale crystals can completely (50 × 50 × 20 nm3 and 60 × 60 × 20 nm3) or partially (110 × 110 × 23 nm3 and 160 × 160 × 25 nm3) retain the original open framework as observed for the as-synthesized form. When these nanocrystals were heated to 473 K, they could also transform to the shrunk state. Further, methanol adsorption measurement at 303 K for the crystals with a size of 50 × 50 × 20 nm3 showed a gate-opening isotherm for the closed-state sample, but an ordinary type-I isotherm for the open-state sample. Similar size-dependent flexibility was also demonstrated for an analogous MOF [Cu2(bdc)2(bpe)] (bpe = 1,2-bis(4-pyridyl)ethylene). The size-dependent flexibility of the MOFs was ascribed to the different numbers of defects present in the crystal, which promote the phase transition. Some flexible MOFs have been reported to show activation-method-dependent structures and flexibility [97,98]. Because it is generally difficult to prove the identical framework compositions for the samples obtained from different activation methods, such phenomena might be actually induced by the different residual guests. In principle, framework flexibilities or dynamic behaviors of MOFs can be also influenced by the outer physical environment such as temperature, light irradiation, mechanical force and electric field [71,99–102]. For example, [Cu(detz)] (MAF-2, Hdetz = 3,5-diethyl-1,2,4-triazole) is a 3D nbo-a network containing a bcu-type channel system, in which large cavities are connected by very small apertures blocked by the flexible ethyl groups (Fig. 7). Interestingly, MAF-2 shows abnormal N2 adsorption at 195 K rather than 77 K, which can be explained by the different flexibility of the ethyl groups. Only at high temperatures that provide enough thermal energy is the thermal motion of ethyl groups large enough to open the apertures for the transient passage of the N2 molecules. Single-crystal X-ray diffraction studies confirmed the variation in ethyl thermal motion at different temperatures, and there is no significant temperature-induced framework or guest-induced structural alteration for the host framework [71]. In contrast, thermal expansion and difference in kinetic diameters are the main reasons for most adsorbents showing higher gas adsorption at higher temperatures (mostly adsorption of CO2 at 195 K and no adsorption of N2 at 77 K). Figure 7. View largeDownload slide Temperature-controlled dynamism of ethyl groups that controls the effective aperture size of MAF-2. Adapted from [73] with permission of the American Chemical Society. Figure 7. View largeDownload slide Temperature-controlled dynamism of ethyl groups that controls the effective aperture size of MAF-2. Adapted from [73] with permission of the American Chemical Society. The temperature-controlled dynamism of flexible pore aperture can be also realized by using rigid frameworks and mobile guest species. For example, Bu et al. reported an interesting MOF [Cu2(btr)2](NO3)2 (btr = 4,4΄-bis(1,2,4-triazole)) with a rigid 3D cationic coordination network, in which the small apertures are blocked by the counter anion NO3– [103]. Similarly to the N2 adsorption of MAF-2, [Cu2(btr)2](NO3)2 can adsorb CO2 at high temperature (231 K) rather than low temperature (195 K and 226 K). Variable-temperature single-crystal X-ray diffraction analyses of [Cu2(btr)2](NO3)2 demonstrated the significant increase in the thermal motion of the nitrate guests. Mechanical pressure change can usually induce structural transformations of flexible MOFs similar to those induced by temperature change, and even more drastic transformations including reconstitution of coordination connectivities [101,104,105]. Obviously, mechanical pressure can be also used to modulate the flexibility of MOFs toward other external stimuli, albeit reported examples are still scarce. Long et al. observed the increases in the gate-opening pressure for CH4 adsorption and the energy of the gate-opening structural transformation, via increasing the [Co(bdp)] sample packing density, corresponding to applying higher mechanical pressure [101]. Although the CH4 uptake and working capacity both decrease under higher mechanical pressure, the higher energy consumption for the gate-opening structural transformation is beneficial for realizing a free-energy adsorption/desorption process. CONCLUSIONS Undoubtedly, framework flexibility and dynamic behavior are unique characteristics of MOFs (compared with other types of porous materials) and extremely important for their adsorption, thermal expansion and other properties. After the discovery of many interesting types of framework flexibilities, rational design/control of these behaviors is emerging as an attractive topic of research. As discussed above, the flexibility of MOFs can be designed/modified/controlled by many strategies, at different conceptual levels from designing/synthesizing new materials to tailoring/controlling properties of known materials. It should be pointed out that designing, tailoring or controlling framework flexibility is useful not only for understanding the structure–property relationship of MOFs, but also as a new dimension for developing MOF materials with excellent performances for molecular recognition, high storage/delivery capacity, selective separation, abnormal/controllable thermal expansion and so on. FUNDING This work was supported by the National Basic Research Program of China (973 Project, 2014CB845602) and the National Natural Science Foundation of China (21290173 and 21473260). REFERENCES 1. Kitagawa S, Kondo M. Functional micropore chemistry of crystalline metal complex-assembled compounds. Bull Chem Soc Jpn  1998; 71: 1739– 53. https://doi.org/10.1246/bcsj.71.1739 Google Scholar CrossRef Search ADS   2. Furukawa H, Cordova KE, O’Keeffe M et al. The chemistry and applications of metal-organic frameworks. Science  2013; 341: 1230444. https://doi.org/10.1126/science.1230444 Google Scholar CrossRef Search ADS PubMed  3. MacGillivray LR. Metal-Organic Frameworks: Design and Application . Hoboken, NJ, USA: John Wiley & Sons, Inc., 2010. Google Scholar CrossRef Search ADS   4. Kitaura R, Fujimoto K, Noro S et al.   A pillared-layer coordination polymer network displaying hysteretic sorption: [Cu2(pzdc)2(dpyg)]n (pzdc= Pyrazine-2,3-dicarboxylate; dpyg=1,2-Di(4-pyridyl)glycol). Angew Chem Int Ed  2002; 41: 133– 5. https://doi.org/10.1002/1521-3773(20020104)41:1%3c133::AID-ANIE133%3e3.0.CO;2-R Google Scholar CrossRef Search ADS   5. Serre C, Millange F, Thouvenot C et al. Very large breathing effect in the first nanoporous chromium(III)-Based Solids: MIL-53 or CrIII(OH)·{O2C-C6H4-CO2}·{HO2C-C6H4-CO2H}x H2Oy. J Am Chem Soc  2002; 124: 13519– 26. https://doi.org/10.1021/ja0276974 Google Scholar CrossRef Search ADS PubMed  6. Kitagawa S, Kitaura R, Noro S. Functional porous coordination polymers. Angew Chem Int Ed  2004; 43: 2334– 75. https://doi.org/10.1002/anie.200300610 Google Scholar CrossRef Search ADS   7. Uemura K, Matsuda R, Kitagawa S. Flexible microporous coordination polymers. J Solid State Chem  2005; 178: 2420– 9. https://doi.org/10.1016/j.jssc.2005.05.036 Google Scholar CrossRef Search ADS   8. Férey G, Serre C. Large breathing effects in three-dimensional porous hybrid matter: facts, analyses, rules and consequences. Chem Soc Rev  2009; 38: 1380– 99. https://doi.org/10.1039/b804302g Google Scholar CrossRef Search ADS PubMed  9. Horike S, Shimomura S, Kitagawa S. Soft porous crystals. Nat Chem  2009; 1: 695– 704. https://doi.org/10.1038/nchem.444 Google Scholar CrossRef Search ADS PubMed  10. Zhang J-P, Liao P-Q, Zhou H-L et al. Single-crystal X-ray diffraction studies on structural transformations of porous coordination polymers. Chem Soc Rev  2014; 43: 5789– 814. https://doi.org/10.1039/C4CS00129J Google Scholar CrossRef Search ADS PubMed  11. Suh MP, Cheon YE. Recent advances in the dynamics of single crystal to single crystal transformations in metal-organic open frameworks. Aust J Chem  2006; 59: 605– 12. https://doi.org/10.1071/CH06016 Google Scholar CrossRef Search ADS   12. Halder GJ, Kepert CJ. Single crystal to single crystal structural transformations in molecular framework materials. Aust J Chem  2006; 59: 597– 604. https://doi.org/10.1071/CH06322 Google Scholar CrossRef Search ADS   13. Kawano M, Fujita M. Direct observation of crystalline-state guest exchange in coordination networks. Coordin Chem Rev  2007; 251: 2592– 605. https://doi.org/10.1016/j.ccr.2007.07.022 Google Scholar CrossRef Search ADS   14. Coudert F, Jeffroy M, Fuchs AH et al. Thermodynamics of guest-induced structural transitions in hybrid organic–inorganic frameworks. J Am Chem Soc  2008; 130: 14294– 302. https://doi.org/10.1021/ja805129c Google Scholar CrossRef Search ADS PubMed  15. Zhang J-P, Chen X-M. Optimized acetylene/carbon dioxide sorption in a dynamic porous crystal. J Am Chem Soc  2009; 131: 5516– 21. https://doi.org/10.1021/ja8089872 Google Scholar CrossRef Search ADS PubMed  16. Mason JA, Oktawiec J, Taylor MK et al. Methane storage in flexible metal-organic frameworks with intrinsic thermal management. Nature  2015; 527: 357– 61. https://doi.org/10.1038/nature15732 Google Scholar CrossRef Search ADS PubMed  17. Lin R-B, Chen D, Lin Y-Y et al. A zeolite-like zinc triazolate framework with high gas adsorption and separation performance. Inorg Chem  2012; 51: 9950– 5. https://doi.org/10.1021/ic301463z Google Scholar CrossRef Search ADS PubMed  18. Liao P-Q, Huang N-Y, Zhang W-X et al. Controlling guest conformation for efficient purification of butadiene. Science  2017; 356: 1193– 6. https://doi.org/10.1126/science.aam7232 Google Scholar CrossRef Search ADS PubMed  19. Chen B, Liang C, Yang J et al. A microporous metal-organic framework for gas-chromatographic separation of alkanes. Angew Chem Int Ed  2006; 45: 1390– 3. https://doi.org/10.1002/anie.200502844 Google Scholar CrossRef Search ADS   20. Li L, Lin R, Krishna R et al. Flexible-robust metal-organic framework for efficient removal of propyne from propylene. J Am Chem Soc  2017; 139: 7733– 6. https://doi.org/10.1021/jacs.7b04268 Google Scholar CrossRef Search ADS PubMed  21. Witman M, Ling S, Jawahery S et al. The influence of intrinsic framework flexibility on adsorption in nanoporous materials. J Am Chem Soc  2017; 139: 5547– 57. https://doi.org/10.1021/jacs.7b01688 Google Scholar CrossRef Search ADS PubMed  22. Li J, Sculley J, Zhou H. Metal-organic frameworks for separations. Chem Rev  2012; 112: 869– 932. https://doi.org/10.1021/cr200190s Google Scholar CrossRef Search ADS PubMed  23. Huang X-C, Lin Y-Y, Zhang J-P et al. Ligand-directed strategy for zeolite-type metal-organic frameworks: zinc(II) imidazolates with unusual zeolitic topologies. Angew Chem Int Ed  2006; 45: 1557– 9. https://doi.org/10.1002/anie.200503778 Google Scholar CrossRef Search ADS   24. Zhang J, Zhu A, Lin R et al. Pore surface tailored sod-type metal-organic zeolites. Adv Mater  2011; 23: 1268– 71. https://doi.org/10.1002/adma.201004028 Google Scholar CrossRef Search ADS PubMed  25. Park KS, Ni Z, Cote AP et al. Exceptional chemical and thermal stability of zeolitic imidazolate frameworks. Proc Natl Acad Sci U S A  2006; 103: 10186– 91. https://doi.org/10.1073/pnas.0602439103 Google Scholar CrossRef Search ADS PubMed  26. Bux H, Liang F, Li Y et al. Zeolitic imidazolate framework membrane with molecular sieving properties by microwave-assisted solvothermal synthesis. J Am Chem Soc  2009; 131: 16000– 1. https://doi.org/10.1021/ja907359t Google Scholar CrossRef Search ADS PubMed  27. Zhu A-X, Lin R-B, Qi X-L et al. Zeolitic metal azolate frameworks (MAFs) from ZnO/Zn(OH)2 and monoalkyl-substituted imidazoles and 1,2,4-triazoles: efficient syntheses and properties. Micropor Mesopor Mat  2012; 157: 42– 9. https://doi.org/10.1016/j.micromeso.2011.11.033 Google Scholar CrossRef Search ADS   28. Zhang J-P, Zhu A-X, Chen X-M. Single-crystal X-ray diffraction and Raman spectroscopy studies of isobaric N2 adsorption in SOD-type metal-organic zeolites. Chem Commun  2012; 48: 11395– 7. https://doi.org/10.1039/c2cc35544b Google Scholar CrossRef Search ADS   29. Schneemann A, Bon V, Schwedler I et al. Flexible metal-organic frameworks. Chem Soc Rev  2014; 43: 6062– 96. https://doi.org/10.1039/C4CS00101J Google Scholar CrossRef Search ADS PubMed  30. Nagarkar SS, Desai AV, Ghosh SK. Stimulus-responsive metal-organic frameworks. Chem Asian J  2014; 9: 2358– 76. https://doi.org/10.1002/asia.201402004 Google Scholar CrossRef Search ADS PubMed  31. Coudert F. Responsive metal-organic frameworks and framework materials: under pressure, taking the heat, in the spotlight, with friends. Chem Mater  2015; 27: 1905– 16. https://doi.org/10.1021/acs.chemmater.5b00046 Google Scholar CrossRef Search ADS   32. Li C, Du M. Role of solvents in coordination supramolecular systems. Chem Commun  2011; 47: 5958– 72. https://doi.org/10.1039/c1cc10935a Google Scholar CrossRef Search ADS   33. Chang Z, Yang D, Xu J et al. Flexible metal-organic frameworks: recent advances and potential applications. Adv Mater  2015; 27: 5432– 41. https://doi.org/10.1002/adma.201501523 Google Scholar CrossRef Search ADS PubMed  34. Zhang JP, Zhou HL, Zhou DD et al. Dynamic behavior of porous coordination polymers. In: Atwood JL (ed). Comprehensive Supramolecular Chemistry II , Vol. 7. Oxford: Elsevier, 2017, 425– 74. Google Scholar CrossRef Search ADS   35. Li B, Wen H, Wang H et al. A porous metal-organic framework with dynamic pyrimidine groups exhibiting record high methane storage working capacity. J Am Chem Soc  2014; 136: 6207– 10. https://doi.org/10.1021/ja501810r Google Scholar CrossRef Search ADS PubMed  36. Wang G, Leus K, Couck S et al. Enhanced gas sorption and breathing properties of the new sulfone functionalized COMOC-2 metal organic framework. Dalton Trans  2016; 45: 9485– 91. https://doi.org/10.1039/C6DT01355D Google Scholar CrossRef Search ADS PubMed  37. Wieme J, Vanduyfhuys L, Rogge SMJ et al. Exploring the flexibility of MIL-47(V)-type materials using force field molecular dynamics simulations. J Phys Chem C  2016; 120: 14934– 47. https://doi.org/10.1021/acs.jpcc.6b04422 Google Scholar CrossRef Search ADS   38. Deng H, Grunder S, Cordova KE et al. Large-pore apertures in a series of metal-organic frameworks. Science  2012; 336: 1018– 23. https://doi.org/10.1126/science.1220131 Google Scholar CrossRef Search ADS PubMed  39. Mcdonald TM, Lee WR, Mason JA et al. Capture of carbon dioxide from air and flue gas in the alkylamine-appended metal-organic framework mmen-Mg2 (dobpdc). J Am Chem Soc  2012; 134: 7056– 65. https://doi.org/10.1021/ja300034j Google Scholar CrossRef Search ADS PubMed  40. Sung Cho H, Deng H, Miyasaka K et al. Extra adsorption and adsorbate superlattice formation in metal-organic frameworks. Nature  2015; 527: 503– 7. https://doi.org/10.1038/nature15734 Google Scholar CrossRef Search ADS PubMed  41. Jawahery S, Simon CM, Braun E et al. Adsorbate-induced lattice deformation in IRMOF-74 series. Nat Commun  2017; 8: 13945. https://doi.org/10.1038/ncomms13945 Google Scholar CrossRef Search ADS PubMed  42. Lee WR, Hwang SY, Ryu DW et al. Diamine-functionalized metal-organic framework: exceptionally high CO2 capacities from ambient air and flue gas, ultrafast CO2 uptake rate, and adsorption mechanism. Energy Environ Sci  2014; 7: 744– 51. https://doi.org/10.1039/C3EE42328J Google Scholar CrossRef Search ADS   43. Serre C, Mellot-Draznieks C, Surble S et al. Role of solvent-host interactions that lead to very large swelling of hybrid frameworks. Science  2007; 315: 1828– 31. https://doi.org/10.1126/science.1137975 Google Scholar CrossRef Search ADS PubMed  44. Horcajada P, Salles F, Wuttke S et al. How Linker's modification controls swelling properties of highly flexible Iron(III) dicarboxylates MIL-88. J Am Chem Soc  2011; 133: 17839– 47. https://doi.org/10.1021/ja206936e Google Scholar CrossRef Search ADS PubMed  45. Choi HJ, Dinca M, Long JR. Broadly hysteretic H2 adsorption in the microporous metal-organic framework Co(1,4-benzenedipyrazolate). J Am Chem Soc  2008; 130: 7848– 50. https://doi.org/10.1021/ja8024092 Google Scholar CrossRef Search ADS PubMed  46. Salles F, Maurin G, Serre C et al. Multistep N2 breathing in the metal-organic framework Co(1,4-benzenedipyrazolate). J Am Chem Soc  2010; 132: 13782– 8. https://doi.org/10.1021/ja104357r Google Scholar CrossRef Search ADS PubMed  47. Ma Y, Matsuda R, Sato H et al. A convenient strategy for designing a soft nanospace: an atomic exchange in a ligand with isostructural frameworks. J Am Chem Soc  2015; 137: 15825– 32. https://doi.org/10.1021/jacs.5b09666 Google Scholar CrossRef Search ADS PubMed  48. Taylor MK, Runcevski T, Oktawiec J et al. Tuning the adsorption-induced phase change in the flexible metal-organic framework Co(bdp). J Am Chem Soc  2016; 138: 15019– 26. https://doi.org/10.1021/jacs.6b09155 Google Scholar CrossRef Search ADS PubMed  49. Zhang J-P, Lin Y-Y, Zhang W-X et al. Temperature- or guest-induced drastic single-crystal-to-single-crystal transformations of a nanoporous coordination polymer. J Am Chem Soc  2005; 127: 14162– 3. https://doi.org/10.1021/ja054913a Google Scholar CrossRef Search ADS PubMed  50. Choi SB, Furukawa H, Nam HJ et al. Reversible interpenetration in a metal-organic framework triggered by ligand removal and addition. Angew Chem Int Ed  2012; 51: 8791– 5. https://doi.org/10.1002/anie.201202925 Google Scholar CrossRef Search ADS   51. Aggarwal H, Bhatt PM, Bezuidenhout CX et al. Direct evidence for single-crystal to single-crystal switching of degree of interpenetration in a metal-organic framework. J Am Chem Soc  2014; 136: 3776– 9. https://doi.org/10.1021/ja500530y Google Scholar CrossRef Search ADS PubMed  52. Ferguson A, Liu L, Tapperwijn SJ et al. Controlled partial interpenetration in metal-organic frameworks. Nat Chem  2016; 8: 250– 7. https://doi.org/10.1038/nchem.2430 Google Scholar CrossRef Search ADS PubMed  53. Zhou D-D, Liu Z-J, He C-T et al. Controlling the flexibility and single-crystal to single-crystal interpenetration reconstitution of metal-organic frameworks. Chem Commun  2015; 51: 12665– 8. https://doi.org/10.1039/C5CC05036G Google Scholar CrossRef Search ADS   54. Choi HJ, Dinca M, Dailly A et al. Hydrogenstorage in water-stable metal-organic frameworks incorporating 1,3- and 1,4-benzenedipyrazolate. Energy Environ Sci  2010; 3: 117– 23. https://doi.org/10.1039/B917512A Google Scholar CrossRef Search ADS   55. Goodwin AL, Chapman KW, Kepert CJ. Guest-dependent negative thermal expansion in nanoporous prussian blue analogues MII PtIV(CN)6 xH2O (0 ≤ x ≤ 2; M = Zn, Cd). J Am Chem Soc  2005; 127: 17980– 1. https://doi.org/10.1021/ja056460f Google Scholar CrossRef Search ADS PubMed  56. Goodwin AL, Kepert CJ. Negative thermal expansion and low-frequency modes in cyanide-bridged framework materials. Phys Rev B  2005; 71: 140301. https://doi.org/10.1103/PhysRevB.71.140301 Google Scholar CrossRef Search ADS   57. Serra-Crespo P, Gobechiya E, Ramos-Fernandez EV et al. Interplay of metal node and amine functionality in NH2-MIL-53: modulating breathing behavior through intra-framework interactions. Langmuir  2012; 28: 12916– 22. https://doi.org/10.1021/la302824j Google Scholar CrossRef Search ADS PubMed  58. He C-T, Ye Z-M, Xu Y-T et al. Hyperfine adjustment of flexible pore-surface pockets enables smart recognition of gas size and quadrupole moment. Chem Sci  2017; 8: 7560– 5. Google Scholar CrossRef Search ADS PubMed  59. Millange F, Guillou N, Walton RI et al. Effect of the nature of the metal on the breathing steps in MOFs with dynamic frameworks. Chem Commun  2008; 4732– 4. https://doi.org/10.1039/b809419e 60. Serre C, Millange F, Surble S et al. A route to the synthesis of trivalent transition-metal porous carboxylates with trimeric secondary building units. Angew Chem Int Ed  2004; 43: 6285– 9. https://doi.org/10.1002/anie.200454250 Google Scholar CrossRef Search ADS   61. Férey G, Mellot-Draznieks C, Serre C et al.   A chromium terephthalate-based solid with unusually large pore volumes and surface area. Science  2005; 309: 2040– 2. https://doi.org/10.1126/science.1116275 Google Scholar CrossRef Search ADS PubMed  62. Barthelet K, Marrot J, Férey G et al. VIII(OH){O2C-C6H4-CO2} (HO2C-C6H4-CO2H)x(DMF)y(H2O)z (or MIL-68), a new vanadocarboxylate with a large pore hybrid topology: reticular synthesis with infinite inorganic building blocks? Chem Commun  2004; 520– 1. https://doi.org/10.1039/B312589K 63. Wei Y-S, Chen K-J, Liao P-Q et al. Turning on the flexibility of isoreticular porous coordination frameworks for drastically tunable framework breathing and thermal expansion. Chem Sci  2013; 4: 1539– 46. https://doi.org/10.1039/c3sc22222e Google Scholar CrossRef Search ADS   64. Fukushima T, Horike S, Inubushi Y et al. Solid solutions of soft porous coordination polymers: fine-tuning of gas adsorption properties. Angew Chem Int Ed  2010; 49: 4820– 4. https://doi.org/10.1002/anie.201000989 Google Scholar CrossRef Search ADS   65. Nouar F, Devic T, Chevreau H et al. Tuning the breathing behaviour of MIL-53 by cation mixing. Chem Commun  2012; 48: 10237– 9. https://doi.org/10.1039/c2cc35348b Google Scholar CrossRef Search ADS   66. Horike S, Inubushi Y, Hori T et al. A solid solution approach to 2D coordination polymers for CH4/CO2 and CH4/C2H6 gas separation: equilibrium and kinetic studies. Chem Sci  2012; 3: 116– 20. https://doi.org/10.1039/C1SC00591J Google Scholar CrossRef Search ADS   67. Cohen SM. Postsynthetic methods for the functionalization of metal-organic frameworks. Chem Rev  2012; 112: 970– 1000. https://doi.org/10.1021/cr200179u Google Scholar CrossRef Search ADS PubMed  68. Wang Z, Cohen SM. Modulating metal-organic frameworks to breathe: a postsynthetic covalent modification approach. J Am Chem Soc  2009; 131: 16675– 7. https://doi.org/10.1021/ja907742z Google Scholar CrossRef Search ADS PubMed  69. Leclerc H, Devic T, Devautour-Vinot S et al. Influence of the oxidation state of the metal center on the flexibility and adsorption properties of a porous metal organic framework: MIL-47(V). J Phys Chem C  2011; 115: 19828– 40. https://doi.org/10.1021/jp206655y Google Scholar CrossRef Search ADS   70. Liao P-Q, Zhu A-X, Zhang W-X et al. Self-catalysed aerobic oxidization of organic linker in porous crystal for on-demand regulation of sorption behaviours. Nat Commun  2015; 6: 6350. https://doi.org/10.1038/ncomms7350 Google Scholar CrossRef Search ADS PubMed  71. Zhang J-P, Chen X-M. Exceptional framework flexibility and sorption behavior of a multifunctional porous cuprous triazolate framework. J Am Chem Soc  2008; 130: 6010– 7. https://doi.org/10.1021/ja800550a Google Scholar CrossRef Search ADS PubMed  72. Phillips AE, Goodwin AL, Halder GJ et al. Nanoporosity and exceptional negative thermal expansion in single-network cadmium cyanide. Angew Chem Int Ed  2008; 47: 1396– 9. https://doi.org/10.1002/anie.200704421 Google Scholar CrossRef Search ADS   73. Grobler I, Smith VJ, Bhatt PM et al. Tunable anisotropic thermal expansion of a porous Zinc(II) metal–organic framework. J Am Chem Soc  2013; 135: 6411– 4. https://doi.org/10.1021/ja401671p Google Scholar CrossRef Search ADS PubMed  74. Kim Y, Haldar R, Kim H et al. The guest-dependent thermal response of the flexible MOF Zn2(BDC)2(DABCO). Dalton Trans  2016; 45: 4187– 92. https://doi.org/10.1039/C5DT03710G Google Scholar CrossRef Search ADS PubMed  75. Zhou H-L, Lin R-B, He C-T et al. Direct visualization of a guest-triggered crystal deformation based on a flexible ultramicroporous framework. Nat Commun  2013; 4: 2534. https://doi.org/10.1038/ncomms3534 Google Scholar PubMed  76. Zhou H-L, Zhang Y-B, Zhang J-P et al. Supramolecular-jack-like guest in ultramicroporous crystal for exceptional thermal expansion behaviour. Nat Commun  2015; 6: 6917. https://doi.org/10.1038/ncomms7917 Google Scholar CrossRef Search ADS PubMed  77. Keene TD, Rankine D, Evans JD et al. Solvent-modified dynamic porosity in chiral 3D kagome frameworks. Dalton Trans  2013; 42: 7871– 9. https://doi.org/10.1039/c3dt00096f Google Scholar CrossRef Search ADS PubMed  78. Wei Y-S, Zhang M, Liao P-Q et al. Coordination templated [2+2+2] cyclotrimerization in a porous coordination framework. Nat Commun  2015; 6: 8348. https://doi.org/10.1038/ncomms9348 Google Scholar CrossRef Search ADS PubMed  79. Zhai Q-G, Bu X, Mao C et al. An ultra-tunable platform for molecular engineering of high-performance crystalline porous materials. Nat Commun  2016; 7: 13645. Google Scholar CrossRef Search ADS PubMed  80. Zhai Q, Bu X, Zhao X et al. Pore space partition in metal-organic frameworks. Acc Chem Res  2017; 50: 407– 17. https://doi.org/10.1021/acs.accounts.6b00526 Google Scholar CrossRef Search ADS PubMed  81. Chen Q, Chang Z, Song W et al. A controllable gate effect in Cobalt(II) organic frameworks by reversible structure transformations. Angew Chem Int Ed  2013; 52: 11550– 3. https://doi.org/10.1002/anie.201306304 Google Scholar CrossRef Search ADS   82. Wang H, Xu J, Zhang D et al. Crystalline capsules: metal-organic frameworks locked by size-matching ligand bolts. Angew Chem Int Ed  2015; 54: 5966– 70. https://doi.org/10.1002/anie.201500468 Google Scholar CrossRef Search ADS   83. Carrington EJ, Mcanally CA, Fletcher AJ et al. Solvent-switchable continuous-breathing behaviour in a diamondoid metal-organic framework and its influence on CO2 versus CH4 selectivity. Nat Chem  2017; 9: 882– 9. https://doi.org/10.1038/nchem.2747 Google Scholar CrossRef Search ADS PubMed  84. Yanai N, Kitayama K, Hijikata Y et al. Gas detection by structural variations of fluorescent guest molecules in a flexible porous coordination polymer. Nat Mater  2011; 10: 787– 93. https://doi.org/10.1038/nmat3104 Google Scholar CrossRef Search ADS PubMed  85. Aggarwal H, Das RK, Bhatt PM et al. Isolation of a structural intermediate during switching of degree of interpenetration in a metal-organic framework. Chem Sci  2015; 6: 4986– 92. https://doi.org/10.1039/C5SC01796C Google Scholar CrossRef Search ADS   86. Le Ouay B, Kitagawa S, Uemura T. Opening of an accessible microporosity in an otherwise nonporous metal-organic framework by polymeric guests. J Am Chem Soc  2017; 139: 7886– 92. https://doi.org/10.1021/jacs.7b02402 Google Scholar CrossRef Search ADS PubMed  87. Huang X-C, Zhang J-P, Chen X-M. [Zn(bim)2]·(H2O)1.67: a metal-organic open-framework with sodalite topology. Chinese Sci Bull  2003; 48: 1531– 4. Google Scholar CrossRef Search ADS   88. Gucuyener C, Van Den Bergh J, Gascon J et al. Ethane/ethene separation turned on its head: selective ethane adsorption on the metal-organic framework zif-7 through a gate-opening mechanism. J Am Chem Soc  2010; 132: 17704– 6. https://doi.org/10.1021/ja1089765 Google Scholar CrossRef Search ADS PubMed  89. Cai W, Lee T, Lee M et al. Thermal structural transitions and carbon dioxide adsorption properties of zeolitic imidazolate framework-7 (ZIF-7). J Am Chem Soc  2014; 136: 7961– 71. https://doi.org/10.1021/ja5016298 Google Scholar CrossRef Search ADS PubMed  90. Linder-Patton OM, Bloch WM, Coghlan CJ et al. Particle size effects in the kinetic trapping of a structurally-locked form of a flexible MOF. CrystEngComm  2016; 18: 4172– 9. https://doi.org/10.1039/C6CE00082G Google Scholar CrossRef Search ADS   91. Sakaida S, Otsubo K, Sakata O et al. Crystalline coordination framework endowed with dynamic gate-opening behaviour by being downsized to a thin film. Nat Chem  2016; 8: 377– 83. https://doi.org/10.1038/nchem.2469 Google Scholar CrossRef Search ADS PubMed  92. Zhang C, Gee JA, Sholl DS et al. Crystal-size-dependent structural transitions in nanoporous crystals: adsorption-induced transitions in ZIF-8. J Phys Chem C  2014; 118: 20727– 33. https://doi.org/10.1021/jp5081466 Google Scholar CrossRef Search ADS   93. Tanaka S, Fujita K, Miyake Y et al. Adsorption and diffusion phenomena in crystal size engineered ZIF-8 MOF. J Phys Chem C  2015; 119: 28430– 9. https://doi.org/10.1021/acs.jpcc.5b09520 Google Scholar CrossRef Search ADS   94. Tanaka D, Henke A, Albrecht K et al. Rapid preparation of flexible porous coordination polymer nanocrystals with accelerated guest adsorption kinetics. Nat Chem  2010; 2: 410– 6. https://doi.org/10.1038/nchem.627 Google Scholar CrossRef Search ADS PubMed  95. Hijikata Y, Horike S, Tanaka D et al. Differences of crystal structure and dynamics between a soft porous nanocrystal and a bulk crystal. Chem Commun  2011; 47: 7632– 4. https://doi.org/10.1039/c1cc10983a Google Scholar CrossRef Search ADS   96. Sakata Y, Furukawa S, Kondo M et al. Shape-memory nanopores induced in coordination frameworks by crystal downsizing. Science  2013; 339: 193– 6. https://doi.org/10.1126/science.1231451 Google Scholar CrossRef Search ADS PubMed  97. Kanoo P, Haldar R, Reddy SK et al. Crystal dynamics in multi-stimuli-responsive entangled metal-organic frameworks. Chem Eur J  2016; 22: 15864– 73. https://doi.org/10.1002/chem.201602087 Google Scholar CrossRef Search ADS PubMed  98. Engel ER, Jouaiti A, Bezuidenhout CX et al. Activation-dependent breathing in a flexible metal-organic framework and the effects of repeated sorption/desorption cycling. Angew Chem Int Ed  2017; 56: 8874– 8. https://doi.org/10.1002/anie.201704044 Google Scholar CrossRef Search ADS   99. Yanai N, Uemura T, Inoue M et al. Guest-to-host transmission of structural changes for stimuli-responsive adsorption property. J Am Chem Soc  2012; 134: 4501– 4. https://doi.org/10.1021/ja2115713 Google Scholar CrossRef Search ADS PubMed  100. Ma S, Sun D, Wang X et al. A mesh-adjustable molecular sieve for general use in gas separation. Angew Chem Int Ed  2007; 46: 2458– 62. https://doi.org/10.1002/anie.200604353 Google Scholar CrossRef Search ADS   101. Mason JA, Oktawiec J, Taylor MK et al. Methane storage in flexible metal-organic frameworks with intrinsic thermal management. Nature  2015; 527: 357– 61. https://doi.org/10.1038/nature15732 Google Scholar CrossRef Search ADS PubMed  102. Ghoufi A, Benhamed K, Boukli-Hacene L et al. Electrically induced breathing of the MIL-53(Cr) metal-organic framework. ACS Cent Sci  2017; 3: 394– 8. https://doi.org/10.1021/acscentsci.6b00392 Google Scholar CrossRef Search ADS PubMed  103. Gao Q, Xu J, Cao D et al. A rigid nested metal-organic framework featuring a thermoresponsive gating effect dominated by counterions. Angew Chem Int Ed  2016; 55: 15027– 30. https://doi.org/10.1002/anie.201608250 Google Scholar CrossRef Search ADS   104. Serra-Crespo P, Dikhtiarenko A, Stavitski E et al. Experimental evidence of negative linear compressibility in the MIL-53 metal-organic framework family. CrystEngComm  2015; 17: 276– 80. https://doi.org/10.1039/C4CE00436A Google Scholar CrossRef Search ADS PubMed  105. Spencer EC, Angel RJ, Ross NL et al. Pressure-induced cooperative bond rearrangement in a zinc imidazolate framework: a high-pressure single-crystal X-Ray diffraction study. J Am Chem Soc  2009; 131: 4022– 6. https://doi.org/10.1021/ja808531m Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of China Science Publishing & Media Ltd. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

National Science ReviewOxford University Press

Published: Oct 16, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off