Flexibility in Metal–Organic Frameworks: A fundamental understanding

Coordination Chemistry Reviews 358 (2018) 125–152 Contents lists available at ScienceDirect Coordination Chemistry Reviews journal homepage: www.elsevier.com/locate/ccr Review Flexibility in Metal–Organic Frameworks: A fundamental understanding Sameh K. Elsaidi a,b,⇑,1, Mona H. Mohamed a,1, Debasis Banerjee b, Praveen K. Thallapally b,⇑ a b Chemistry Department, Faculty of Science, Alexandria University, P. O. Box 426, Ibrahimia, Alexandria 21321, Egypt Physical and Computational Science Directorate, Pacific Northwest National Laboratory, Richland, WA 99352, United States a r t i c l e i n f o Article history: Received 12 June 2017 Accepted 16 November 2017 Keywords: Metal–Organic Framework Porous Coordination Polymers Flexibility Soft crystals Breathing Gate opening Gas adsorption a b s t r a c t Expansion and contraction of structurally flexible Metal–Organic Frameworks (MOFs) or Porous Coordination Polymers (PCPs) have been extensively studied from different structural and application perspectives including crystal engineering, structural characterization, and gas adsorption-separation applications. The flexibility of the MOFs or PCPs depends on a number of factors including the nature of secondary building units (SBUs), organic linkers, pore geometry and/or solvent molecules. The flexibility can lead to unique properties, especially in gas-adsorption-separation related applications that is not observed in rigid frameworks. In this review, we offer a brief summary regarding a fundamental understanding of framework expansion–contraction in flexible porous MOFs in terms of their design and structure tunability by covering representative examples in the literature. Ó 2017 Elsevier B.V. All rights reserved. Contents 1. 2. 3. 4. 5. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Intra-framework motives for the contraction or expansion in MOFs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.1. Molecular Building Blocks (MBBs) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.2. The impact of organic linker. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2.3. The impact of topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . External stimuli for the flexible MOFs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Removal or incorporation of guest molecules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. External pressure. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Thermal-induction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Photo-induction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Phenomena of flexible MOFs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1. Gate opening . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2. Phase change . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.1. Displacive phase transitions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.2. Reconstructive phase transitions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.3. Negative gas adsorption (NGA) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Methods for controlling the flexibility of MOFs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.1. Metal ions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.2. .Functional groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3. Host–host interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . p–p interaction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.3.1. 5.3.2. Hydrogen bonding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.4. Post-synthetic modification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5.5. Crystal size. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ⇑ Corresponding authors at: Physical and Computational Science Directorate, Pacific Northwest National Laboratory, Richland, WA 99352, United States. 1 E-mail addresses: Sameh.Elsaidi@pnnl.gov (S.K. Elsaidi), praveen.thallapally@pnnl.gov (P.K. Thallapally). These authors equally contributed to this work. https://doi.org/10.1016/j.ccr.2017.11.022 0010-8545/Ó 2017 Elsevier B.V. All rights reserved. 126 126 128 128 129 129 129 129 131 131 131 132 138 138 139 141 141 141 141 142 142 142 142 145 126 6. 7. S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 5.6. Inclusion of molecular rotors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . List of abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1. Introduction Metal–Organic Materials (MOMs) or Metal–Organic Frameworks (MOFs) are typically assembled from metal cations or metal cluster (nodes, also known as SBUs) that are connected by organic molecules/inorganic anions which serve as linkers to afford periodic frameworks with controllable pore size and chemistry [1–5]. The IUPAC (International Union of Pure & Applied Chemistry) task group on coordination polymers (CPs) and MOFs have given precise recommendations for the terminology and nomenclature of porous materials as numerous terminologies were introduced in the area by various research groups that could lead to unreasonable misunderstanding [6]. When we exclude the zerodimensional structures (e.g. nanoballs, cubes and metal–organic polyhedra) from the family of MOMs/MOFs and cover polymeric one-dimensional, two-dimensional and three-dimensional structures, these are termed as CPs [7–11], if all linkers are organic in nature, the materials are termed as MOFs [12], as exemplified by HKUST-1 (Hong Kong University of Science & Technology) [13] and MOF-5 [14]. The tunable and tailorable structures [12,15– 18], extra-high surface area [19–23] and modular pore functionality [7,24–28] of MOFs affords prodigious control over physicochemical properties, in comparison to their purely inorganic, porous materials such as aluminosilicate zeolites and have made them an attractive class of materials with potential for a wide range of applications such as gas storage and separation [29–33], heterogeneous catalysis [34–38], drug delivery [39,40], and conductivity [41,42]. MOMs (and MOFs) have received particular attention when the concept of crystal engineering was popularized in the early 1990’s [43] and subsequently, their scientific interest has been augmented especially after pioneering work of Williams [13], Yaghi [44] and Kitagawa [45]. It was demonstrated that systematic design of the framework structure can lead to a control over its properties by studying the impact of the molecular structure upon crystal packing, crystal structure and physicochemical properties ‘‘Form for Function” [46–48]. Specifically, the concepts of crystal engineering and selfassembly for the design and synthesis of MOMs (or MOFs) were intensely grown, enabling the systematic study of structure/function relationship in an unprecedented way. This is distinctly different from more random, high-throughput screening approach that is traditionally used in materials discovery and development [48,46]. Though MOFs have been envisioned as being made of rigid, aromatic linkers as strut and metal clusters as nodes, a number of MOFs are known to be flexible in nature in presence of external stimuli such as pressure, temperature, and light [49–51]. The design of MOFs with expansion and contraction or ‘‘breathing” properties is considered as one of a pertinent idea in order to target specific applications such as gas separation, gas storage, sensing and drug delivery [50,52]. These specific types of MOFs have flexible frameworks and lack the rigidity, therefore they are termed as soft porous crystal [53], flexible MOFs, sponge-like MOFs [54,55], spring-like [56] or dynamic MOFs [57] in literature. The ordered crystal structure of these flexible MOFs has the ability to transform through different ways such as a phase change or gate opening [58,59]. Such flexibility is often observed during the adsorption– desorption process, where the interaction of adsorbate molecules with the pore-surface is believed to be the origin of such flexibility 146 149 149 149 149 [60]. The framework flexibility is apparent from the adsorption– desorption curve of the adsorbate molecules with a sharp change in uptake at a certain pressure or temperature, indicative of a change in pore surface properties or an associated phase change. A large number of flexible MOFs with flexible structural features have been reported till date, tailored for applications ranging from gas-adsorption and separation to catalysis and explosive sensing [50]. It should be noted that although flexible MOFs are shown to be excellent candidates for gas-separation for various industrially important gases such as small chain hydrocarbons, unless a detailed structure–property relationship for a particular system as a function of external parameters are worked out, the application of these types of materials in an industrial setting seems unlikely. In many cases, the host–guest interactions and associated flexibilities in MOFs are of fundamental importance and understanding the interaction and associated phase transition behaviours can open up a new avenue in materials design and exploration. Kitagawa’s and Ferey’s groups took the early lead through a variety of study on flexible MOFs such as metal paddlewheel-based pillared square grids, MIL-53 and MIL-88 (MI L = Materials Institute Lavoisier) [50,61–64]. Indeed, the flexibility in MOFs was also addressed from the computational modeling perspectives [65–69]. The earliest work on the molecular modeling of flexible MOFs was presented by Miyahara and his co-workers [68]. They conducted Grand Canonical Monte Carlo (GCMC) simulations on catenated jungle gym structures exhibiting structural transitions and gate adsorption behavior. The calculations and the grand free energy profiles revealed that the integration of the guest molecule during the adsorption process provide a stabilization of the structure that triggers the structural transition. Most recently, Smit and his co-workers have presented several studies on the flexibility of MOFs, focusing on the impact of the flexibility of MOFs on their selective adsorption of gases [65,67]. The authors have developed a simple model to calculate the flexible Henry’s coefficients and selectivity for Xe and Kr gases as a function of the intrinsic flexibility of MOFs. The study gave an insight that the optimal materials for the shape selective adsorption applications should have a synergetic effect between pore size and pore chemistry as well as a minimal structural flexibility. However, for other nonoptimal materials, the selectivity could be improved in the presence of flexibility. The molecular modeling of MOFs is very critical to correctly understand the adsorption behavior of these materials and contribute to the future design of dynamic MOFs that are beneficial in certain applications such as sensing, catalysis, gas storage, gas separation and drug delivery. Rather than covering the entire literature, in the next few sections of this review, we will briefly cover the design perspective of MOFs with expansion–contraction/dynamic behavior and its impact on their properties. 2. Intra-framework motives for the contraction or expansion in MOFs The flexibility of MOFs is influenced by many parameters. One notable factor affecting the contraction and/or expansion of MOFs is their structural compositions including the metal ion/cluster (Molecular Building Blocks or MBBs, also known as secondary building units or SBUs) and the organic linker. Another factor S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 127 Fig. 1. Top: Structure of MIL-53(Cr) in LP (left) and NP (right) forms. Bottom: important angles and interatomic distances in SBU of MIL-53-Cr; LP (left) and NP (right) forms. Reproduced with permission from Refs. [59,82]. Copyright 2002 American Chemical Society and 2014 Elsevier. Fig. 2. (a) The structure of [Zn2(1,4-bdc)2(dabco)]n viewed along fourfold axis of the framework structure in 1
4DMF
1/2H2O. (b) The evacuated framework in space-filling representation showing the open channels along the fourfold axis. (c) Side view of evacuated structure showing the windows linking the channels. (d) Space-filing representation of the [Zn2(1,4-bdc)2(dabco)]n structure in 1
2C6H6, showing rhombic-grid motif of [Zn2(1,4-bdc)2] layers. The guest molecules and dabco hydrogens are omitted for clarity. (Zn green; N blue; C gray; O red; H white). Reproduced with permission from Ref. [71]. Copyright 2004 Wiley-VCH. 128 S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 involves the way these constituents are connected together: ‘‘Topology”, also plays a significant role. These factors can substantially impact the flexibility in MOF structures due to their ability to affect the nature and strength of the bonds of the framework as well as the pore environment. 2.1. Molecular Building Blocks (MBBs) Not all MBBs allow the dynamic movement (i.e. expansion or contraction) of the framework. Férey and Serre [61] stated some of the empirical rules for the possibility of breathing, such as inorganic MBBs should have a mirror plane with the carboxylates in symmetrical positions around it. This was observed in as exemplified by the MIL-88 (Fe, Cr) framework [70]. Another rule is that the ratio of C/M (C: number of carbon atoms of the carboxylates around the cluster; M: number of metal atoms within the cluster) must be 2 so that the MBB can breathe. These two rules were shown to be valid for a number of benchmark flexible MOFs such as [Zn2(bdc)2(dabco)]n [71] (bdc = 1,4-benzenedicarboxylate, dabc o = 1,4-diazabicyclo[2.2.2]-octane) and previously mentioned MIL-88 (Fe, Cr) [70]. Schneemann and co-workers have recently reported a summary about the impact of the metal nodes on the flexibility of MOFs [51]. For example, MIL-53 of formula [M(OH) (bdc)2]n (M = Cr- [59,72], Al- [73], Fe [74], Sc- [75], In- [76], Ga[77]) show different flexibility behavior based on the nature of the metal center. The Cr-, Al- and Ga- analogues of MIL-53 behave the same, where upon activation at 300 °C the expansion of the pore takes place [transition from the narrow pore, as-synthesized (MIL-53-as) to large pore high-temperature structure (MIL-53ht)] (see Fig. 1). However, the Fe- and Sc-analogues behave differently as the pore contraction takes place after activation (narrow pore to closed pore transition). Another approach proposed is that the utilization of some SBUs can lead to the flexibility of the structure and consequently allow breathing [61]. For instance, MOFs assembled by the trigonal prismatic SBUs, M3O(COO)6(H2O)2
X (M = Cr, Fe, Al; X = F, OH), such as one from MIL-88 family, exhibit a change in the orientation of the metal trimer and the phenyl rings due to their rotation around O–O axis of the carboxylate groups (kneecap-like rotational axis) [70]. Another SBU that can allow breathing of the structure is the metal-paddlewheel SBU [64]. The archetypal example, [M2(bdc)2(dabco)]n (M = Zn, Co, Cu) [71,78,79], has a square paddlewheel SBU, constructed by metal dimers, which are in-turn connected with four carboxylate groups from the linkers, forming layered square grid sheets. The layered sheets are then connected by a pillaring linker (dabco) in the axial position to form the overall threedimensional structure. The Zn and Co analogues, [Zn2(bdc)2(dabco)]n [71] and [Co2(bdc)2(dabco)]n [78] show a phase transition during the activation process from narrow pore, as-synthesized to large pore, evacuated structure (see Fig. 2). However, the Cu analogue doesn’t show any structural change between the assynthesized and activated form [79]. Similar observation was found in case of isostructural DUT-8(M) (DUT = Dresden University of Technology), ([M2(2,6-ndc)2(dabco)]n (M = Zn, Co, Ni, Cu, ndc = naphthalenedicarboxylate), where the shorter 1,4-bdc is replaced by a comparatively longer 2,6-ndc linker [80,81]. DUT-8(Cu) showed more rigidity than its AZn, ACo and ANi analogues. All of these observations further emphasize the impact of the metal nodes/SBUs on the flexibility of the framework. 2.2. The impact of organic linker Several reviews have also presented the impact of the flexible linker on the of MOF properties either by explaining the modes of the flexible linkers that lead to the breathing of MOFs or the influence of the different functional groups in the linker that facil- itates the expansion/contraction in MOFs [51,61,82]. Férey proposed an empirical rule concerning the impact of the linkers on breathing as it was observed from the literature that the ditopic carboxylate ligands, that are linked to two metal clusters or SBUs are favourable for the design of flexible MOFs [61]. On the other hand, the use of the tri- or tetra-topic carboxylate linkers prohibit the breathing in MOFs [61]. Organic linkers that can rotate due to the presence of internal flexible bonds (or flexible metal-linker bonding) can lead to framework flexibility. Such rotation allows the expansion of the pore and thereby the unexpected adsorption of larger guests as exemplified by the prototypal structure, ZIF-8, [Zn(mIm)2]n (mIm, also Im) = 2-methylimidazole, ZIF = Zeolitic Imidazolate Framework) [83–85]. It was found that the application of very high pressure (14–700 bar) on ZIF-8 allows the reorientation of the 2-methylimidazole linker which increases the size of the pore window. The process was found to be reversible as by decreasing the pressure the structure is changed back to the original one [83–85]. Another example of the linker rotation in flexible MOFs was found in [Cd2(pzdc)2(BHE-bpb)]n (pzdc = 2,3-pyrazinedicarboxy late; BHE-bpb = 2,5-bis(2-hydroxyethoxy)-1,4-bis(4-pyridyl)ben zene)) where the cadmium metal centers are connected with the dicarboxylate linker (pzdc) to form the layers of [Cd2(pzdc)2]n. The adjacent layers are then connected by a pillaring BHE-bpb linker to form a 3-D framework [86]. The pillaring linker has an AOH group which can interact with the other AOH groups from the adjacent pillars, leading to a pore blockage situation. However, upon the adsorption of polar guest molecules such as water leads to the rotation of the linker and therefore the opening of the pore [86]. In terms of the linker functionalities, three noticeable structures should be particularly mentioned: MIL-53(Fe) [74], MIL-53(Al) [77,87] and [Zn2(fu-bdc)2(dabco)] (fu-bdc = functionalized 1,4benzenedicarboxylate) [87]. The non-functionalized bdc linkers of MIL-53(Fe) and MIL-53(Al) were substituted with different functional groups such as ANH2, ACH3, ACl, ABr, A(COO)2, A(CF3)2 and A(OH)2 of different hydrophobicity, polarity and acidities [60]. Intensive studies have been performed on MIL-53(Fe) using powder X-ray diffraction (PXRD) and computer simulation, which showed that the opening of the pore is mainly based upon the intra-framework interactions rather than the steric hindrance of the functional groups. The as-synthesized large-pore forms undergo a phase transition to the narrow pore form after activation. The activated sample of MIL-53(Fe)–(CF3)2 also undergoes the above-mentioned narrow pore transition, however it is the only structure that showed a measurable BET surface area (SBET = 100 m2/g, BET = Brunauer Emmett Teller) and a larger cell volume which might be attributed to the bulkiness of the ACF3 groups and its inability to form hydrogen bonds with the free AOH groups of the framework. The ease of such structural transition is: MIL-53 (Fe)–(CF3)2 > MIL-53(Fe)–CH3 > MIL-53(Fe) > MIL-53(Fe)–(OH)2 > MIL-53(Fe)–NH2 > MIL-53(Fe) Br > MIL-53(Fe)–Cl. This could be explained by the presence of the inorganic–organic intraframework interaction between the free AOH groups of the framework and the functional group of the organic linker that stabilizes the narrow pore form as exemplified by those variants possess A(OH)2, ACl and ABr. As a result, the structures with a noninteracting, free AOH groups undergoes phase transition with better ease. Rosseinsky and co-workers reported a flexible metaldipeptide framework of formula [Zn(Gly-Ala)2]
(solvent) (Gly-Ala = glycylalanine) [88]. The framework showed no porosity in absence of guest molecule. Nevertheless, the pores are gradually opening above the gate pressure of 2 bar upon the adsorption of small guest molecules with polar bonds such as CO2 as exemplified by the unusual sigmoidal adsorption behavior. The Gly-Ala dipeptide linker was shown to play a significant role in the adaptability S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 129 Fig. 3. A comparison between (a) narrow pore form of MIL-53, (b) large pore form of MIL-53 and (c) MIL-68 structure. Reproduced with permission from Refs. [89,91]. Copyright 2004 and 2013 Royal Society of Chemistry. 3. External stimuli for the flexible MOFs The need for external stimuli or initiator for generation of dynamic MOFs was found to have a great impact on the structural aspects of the framework architecture such as phase change, change in the cell parameters and gate opening. 3.1. Removal or incorporation of guest molecules Fig. 4. The dynamic movement of porous MOFs under the incorporation/removal of guest molecule. Reproduced with permission from Ref. [62]. Copyright 2005 Royal Society of Chemistry. of the framework to the variation in guest loadings due to its low energy torsional motions through a continuous repositioning of methyl-group that allow the adaptable porosity. 2.3. The impact of topology The structures of MIL-53 and MIL-68 reflects the role of topology on breathing, while both structures have the same chemical formula of MIIIOH[bdc]
xG [G = guest, M = Fe(MIL-53). V(MIL-68), bdc = 1,4-benzenedicarboxylate] [89]. MIL-53 has a 44 topology with lozenge-based channels while MIL-68 has a 6.3.6.3 topology with hexagonal and triangular channels (see Fig. 3). These triangular, rigid channels in MIL-68 make the structure un-breathable, where it was found that there is no change of the cell parameters up to the decomposition temperature (350 °C) of the material. Therefore the topology that leads to even cycles allows the breathing to take place (e.g., MIL-53) while the structures with odd cycles are usually rigid. Another example which indicates that odd cycles in the topology of the framework are unfavourable for the breathing is MIL-101(Cr) which has cages of pentagon-shaped windows, prohibiting breathing [61,90]. During the incorporation or removal of guest molecules, dynamic frameworks can expand or contract which subsequently allow a variety of guest exchange [49,62]. For example, Kitagawa proposed three cases that could happen in dynamic MOFs when the guest species are either removed or incorporated (Fig. 4) [62]. The first case includes pillared MOFs, where the pillaring linkers of the frameworks are flexible. Therefore, the frameworks are either elongated or shortened during the incorporation or removal of the guest molecule [92]. Another case appeared in what is called sponge-like MOFs [54,55], which can expand or shrink with a noticeable change in the cell volume due to strong framework– guest interaction [59]. The last reported case appeared in case of interpenetrated grids, where the introduction of guest molecules causes the sliding of the adjacent interpenetrated net [93,94]. The effect of the guest is evident in case of MIL-53(Cr), which has a smaller unit cell rather than as-synthesized (MIL-53-as) upon adsorption of humid air to form a contracted low-temperature form (MIL-53-lt). The cell volume decreases by 32% during the adsorption process. The process was found to be fully reversible and upon dehydration the opening of the channels takes place. The incorporation of the DMF solvent to MIL-53-lt leads to the exchange of the water molecules by DMF and the resulting structure is expanded with a unit cell parameter in between MIL-53as and MIL-53-lt. Another exquisite example shows the effect of using different solvents on the breathing or swelling of the framework: the unit cell volume of evacuated MIL-88(Fe) increases upon the incorporation of different guests such as n-butanol, ethanol, methanol [70]. 3.2. External pressure Pressure is another thermodynamic parameter that is closely related to the flexibility of materials [95–100]. The effect of pressure on a material is directly related to its mechanical stability, critical for many types of commercial applications. MOFs, as softmaterials with metal polyhedra, bridged by the organic linkers, can rotate or relax more easily in response to external pressure. MOFs demonstrate a wide variety of behavior in their response to pressure. For small-scale deformations, examples of anomalous elastic mechanical properties include negative linear 130 S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 Fig. 6. Adsorption/desorption isotherms of CO2, CH4, O2 and N2 in [Cu(4,40 -bipy) (dhbc)2]
2H2O indicating the relation between the gate opening effect and the nature of the gas. Reproduced with permission from Ref. [94]. Copyright 2003 Wiley-VCH. Fig. 5. (a), (e) Coordination sphere for the Er3+ ions; (b), (f) phase transformation process; (c), (d), (g), (h) framework structure viewed along the b-axis and connectivity of the erbium centers of the framework at ambient pressure and 3.524(9) Gpa (Gpa = gigapascal). The tmen2+ guests and hydrogen atoms are omitted for clarity. In (d) and (h), black struts represent Er–(HCOO)–Er links that lie approximately within the bc plane and generate a sheet motif, and parallel to the a-axis and act as pillars between neighboring sheets. Red struts show the new linkages that are formed as the crystal converts from phase I into phase II. The formate ligands, which change from a chelating mode in phase I to an anti–anti bridging coordination mode after the transition (phase II), have been enlarged and the bond rearrangement is indicated by arrows. Er3+ green, O red, C black, N blue. Reproduced with permission from Ref. [95]. Copyright 2014 Wiley-VCH. compressibility and negative area compressibility. Larger scale deformation includes pressure-induced phase transformations including pressure-induced amorphization. Several reports on MOF flexibility as a function of pressure in reported in the literature. Examples with features like a decrease in unit cell volume with increase in applied pressure and irreversible bond-rearrangement to amorphization are reported [96,101,102]. For example, dense Zn(im)2 (im = imidazolate) was shown to undergo an irreversible cooperative bond rearrangement under applied pressure, as depicted by single crystal X-ray diffraction studies using a high-pressure diamond anvil cell [96]. Cheetham and co-workers [101] reported ZIF-8; a prototypical MOF with rho net topology forms an amorphous form upon ball milling, with a continuous random network (CRN) topology. The material so formed has a higher crystal density and a lower porosity than its crystalline counterpart with a decrease in thermal stability. However, a truly flexible material should show reversible behavior under external parameters such as applied pressure. Cheetham and co-workers reported such an example, where a lanthanide-based MOF, namely [tmenH2][Er(HCOO)4]2 (tmenH2 = N,N,N0 ,N0 -tetrame thylethylenediammonium) undergoes a reversible pressureinduced phase transition (PIPTs), associated with bond rearrangement, as identified by high-pressure single crystal diffraction and nanoindentation technique [95]. The framework has a pillaredlayered three-dimensional anionic [Er(HCOO)4] structure with the tmemH2+ 2 cations located in the channel. The erbium metal center is eight-coordinated, with each ErO8 polyhedra is connected to five neighboring erbium atoms by five formate bridges to form the erbium formate layers that lie parallel to the crystallographic bc plane. The adjacent layers are pillared parallel to the a axis by a sixth formate bridge (see Fig. 5). The framework undergoes the transition as a function of pressure: A first-order phase transition occurs at 0.6 GPa. The framework retains its monoclinic symmetry and space group C2/c, but the unit cell volume of the higher pressure phase (phase II) is 10% smaller than that of the low pressure (phase I) one. The length of the crystallographic a axis undergoes a pronounced reduction of about 11% through the phase transition, while the crystallographic b and c axes expand by about 0.5% and 0.7% respectively. Both phases exhibit significant differences in compressibility, indicating that the phase transition occurs through extensive alteration of the framework structure. The transition from phase I to phase II was found to be reversible, as confirmed by means of independent nanoindentation measurements on single crystals. Furthermore, the reversibility of the phase transition confirms the flexibility of the framework. Ortiz and co-workers reported another elegant example of MOF flexibility under pressure: Two Zinc Alkyl Gate (ZAG) frameworks, namely ZAG-4 and ZAG-6, MOFs with common wine rack geometric motif with alkyl chains as the linker molecules, show negative linear compressibility (NLC) as a function of applied pressure, as evidenced by high-pressure single-crystal X-ray crystallography and quantum mechanical calculations. The S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 NLC behavior originates due to the occurrence of a pressureinduced structural transition involving a reversible proton transfer between an included water molecule and the linker’s phosphonate group. Moreover, while the 4-carbon alkyl chain of ZAG-4 acts as a rigid linker, it was shown that the 6-carbon alkyl chain of ZAG-6 exhibits a coiling transition under pressure. Again, the reversibility of the transition confirms the flexibility of the material itself. If the pore opening is closed or blocked by guest molecules then the external pressure is often necessary for its opening. [Cu(4,40 bipy)(dhbc)2]
2H2O (bipy = bipyridine, dhbc = 2,5-dihydroxybenzo ate) [94] represents a perfect example of the application of applied pressure as a stimulus for the breathing of MOFs. The study takes place using different common gases such as N2, CO2, O2 and CH4. As shown in Fig. 6 each gas has different pressure at which the pores open and the adsorption takes place. This pressure (termed as threshold pressure, also called gate-opening pressure) is inversely proportional to the gas-framework interaction, represented by the enthalpy or heat of adsorption (Qst). For example, [Cu(4,40 bipy)(dhbc)2]
2H2O has a very high affinity and Qst for CO2 which allow the opening of the pore (gate-effect) at very low pressure. The framework has a lower affinity towards other gases especially N2 and thus needs a relatively high pressure in order to open the pore [94]. Along the same line, a high-pressure structural study on ZIF-8 has been performed using single crystal X-ray diffraction in a Diamond anvil cell with a methanol/ethanol mixture as hydrostatic medium (at 0.18, 0.52, 0.96, and 1.47 GPa pressure) [102]. The high-pressure diffraction data showed that by applying the pressure up to 0.18 GPa, more solvent molecules squeeze inside the pore, thereby increasing the pore size from 2.465 to 2.556 Å (unit cell volume from 4900.5 to 4999.6 Å3). The increase in pressure to 0.96 GPa, however, leads to a decrease in the unit cell and the contraction of the pore takes place to be close to that of the original structure. Interestingly, a phase transition to form a high-pressure phase (ZIF-8-II) occurs at a pressure of 1.47 GPa. The high-pressure phase (ZIF-8-II) has the same symmetry as the original structure but has a larger accessible pore volume by means of an increase of pore aperture due to the rotation of the linker. Such pressureinduced transformation was found in ZAG-4 (Zinc Alkyl Gate), [Zn(BB-pc)
2H2O] (BB-pc = 1,4-butane bis(phosphonic acid) [100], where the cell volume decreases by 27% upon the increase in pressure to 1.65 Gpa. Further increase in the pressure lead to a decrease of the OAZnAO angle, as a result, while the length of crystallographic a and c axes decrease, the length of the b-axis slightly increases. MIL-47(V(IV)), [V(O)(bdc)]n, also showed a similar phase transition from open large pore to small or narrow pore phase under applied pressure [103]. 3.3. Thermal-induction Temperature plays an important role as stimuli for expansion or contraction in MOFs, without a change in the overall structural composition [51,104,105]. For example, change in temperature can lead to a reversible or irreversible phase transition [75]. The increase in temperature leads to the removal of the solvent in the channels of the framework that cause the dynamic movement in the structure [104,105]. The disordered guest bdc molecules in the channel of the MIL-53-as (unit cell volume: 1440 Å3) can be removed by heating at 573 K to generate MIL-53-ht (unit-cell volume of 1486 Å3) and empty channels [59,72]. However this change is irreversible as the MIL-53-ht form another phase, namely MIL53-lt upon cooling. Temperature also influences the thermal motions of the linkers or the side chain attached functional groups, which leads to breathing. Isostructural frameworks of formula, [Zn2(F/BME/DB-bdc)2 (dabco)]n [F-bdc = alkoxy functionalized 1,4-benezenedicarboxylate, 131 BME-bdc = 2,5-bis(2-methoxyethoxy)benzenedicarboxylate, DB-bdc = 2,5-dibutoxybenzenedicarboxylate] experience swelling and subsequent pore opening by increase in temperature due to the thermal movement of side chains [87,106]. The transition temperature from narrow pore to large pore depends on the nature of the linker functionality [85,97]. Similar examples were reported in literature where linkers or its side chains can undergo a thermal movement to provide a larger pore such as in HMOF-1 (HMOF = Hinged Metal–Organic Framework) [107] and [Zn(NIY-bc)(OH)]n (NIY-bc = 4-(1H-naphtho[2,3-d]imidazol-1-yl)benzoate) [108]. HMOF-1 is a thermo-responsive three-dimensional MOF constructed from CdI2 and flexible meso-tetra(4-pyridyl)porphine based linker [107]. The thermos-responsive nature of the material arises because of the movement of the axial –I atom as a function of temperature. For [Zn(NIY-bc)(OH)]n, breathing motion was observed as a function of temperature, because of the movement of the flexible organic linker. The linker rotates to avoid the collision of the naphthalene groups from the adjacent linkers, leading to the opening of the pore (Fig. 7) [108]. 3.4. Photo-induction Light can also be used as a stimulus for the dynamic movement of MOFs that mainly occurs by the movement or change in conformation of the photosensitive functional groups of the linker molecule upon irradiation [51]. For example, azobenzene and related organic molecules can change their conformation from trans- to cis- after exposing to light (k = 365 nm). The first example of a photo-responsive MOF was reported by Stock and co-workers: [Zn2(2,6-ndc)2(azo-bipy)] (azo-bipy = 3-azo-phenyl-4,40 -bipyri dine) is a three-dimensional framework, where the layers of the Zn2(2,6-ndc)2 are pillared by the azo-bipy linkers. The 3-azophenyl side chain is exposed towards the channel [109]. In the as-synthesized structure 3-azo-phenyl groups are present in the stable trans-form, which transforms to the cis-form upon exposure to the light (k = 365 nm), increasing the pore accessibility (see Fig. 8). The process was found to be fully reversible, where the cis-form can be transformed back to the more stable trans-form by heating or exposing the material to a higher wavelength light (k = 440 nm). Several other MOFs are also synthesized using similar azofunctionalized linkers such as PCN-123 [110], IRMOF-74 III [111], and [Zn2(AzDC)2(4,40 -bpe)]n (AzDC = azobenzene-4,40 -dicarboxy late; 4,40 -bpe = trans-1,2-bis(4-pyridyl)ethylene) [112]. Another route to form flexible, photo-responsive MOF is by incorporation of photoactive organic guest molecules inside the channel. An example illustrating this case is the three-dimensional MOF, [Zn2(bdc)2(dabco)] loaded with azo-benzene [113]: when the incorporated azo-benzene molecules are present in the cis-form, a framework contraction takes place around the azo-benzene molecule, while the expansion of the framework happens in the presence of the trans-form under the effect of light (see Fig. 8). 4. Phenomena of flexible MOFs Breathing (or expansion–contraction) is the ability of the framework to reversibly contact or expand which is the general definition for the dynamic movement in MOFs. This phenomenon is exemplified by the prototypal flexible MOFs, such as the family of MIL-53(M), [MIII(OH)(1,4-bdc)] (M = Cr [59,72], Fe [74], Al [73], Sc [75], Ga [77], In [76]). Similarly, MIL-88 represents a great example of the breathing effect in MOFs and it has the largest expansion ever seen in hybrid materials [70]. The general formula III 3+ 3+ of MIL-88 family is [MIII 3 O(X)3[R(COO)2]3
xG (M = Fe , Cr ; X = CH3OH, H2O, F), which is based on trimers of inorganic cluster 132 S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 Fig. 7. The structure of MIL-53(Sc) viewed down the channel axis in three different transitions: cp (above, middle) and np (below). Reproduced with permission from Ref. [75]. Copyright 2012 Royal Society of Chemistry. MIII 3 O (three metal ions sharing a m3-oxygen) connected by the dicarboxylate linkers where each cluster has three carboxylates groups are positioned above the plane of the trimer and three below. The dicarboxylate linkers are rigid in this family of MOFs: fumaric acid, (H2fa, MIL-88A), 1,4-bdc (MIL-88B), 2,6-ndc (MIL-88C) or 4,40 -biphe nyldicarboxylic acid (H2bpdc, MIL-88D) [114]. MIL-53(Al) and MIL-53(Cr) are interesting materials because of their ‘‘breathing” forms are induced by adsorption of moisture as shown in Fig. 1 [54,61]. The pores are slightly deformed due to hydrogen-bonding interactions between the hydrogen atoms of the water molecules and the oxygen atoms of the carboxylates and the l2-hydroxyl groups in the hydrated form. Once the water was removed from the pore upon heating, the materials go back to its open pore form with higher overall porosity. Bourrelly and coworkers found that the CO2 molecules will initially be adsorbed near the hydroxyl groups in MIL-53(Al) and that leads to the shrinkage of the structure [115]. A further increase in the CO2 pressure will lead to reopening of the pore structure. These phenomenon not only leads to change in the cell parameter of the MOFs, but there are some unique related phenomenon associated with the breathing such as gate-opening and phase change. 4.1. Gate opening Gate-opening is a physical phenomenon where the pore (or gate) of a framework opens in presence of a stimuli especially an external adsorbate molecules [116,117]. As gate-opening of a particular MOF is dependent on parameters such as nature of the adsorbate, the separation of two or more adsorbates is possible when there exist a difference in the gate-opening pressure for the respective adsorbates. Breathing and gate-opening are often used interchangeably in literature as both are associated with flexible MOFs. The phenomenon of gate-opening and breathing can be differentiated by the fact that gate-opening is characterized by transition from an essentially non-porous state to a porous, open phase, while breathing is associated with two successive structural S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 133 Fig. 8. Schematic representations of two ways for introducing the photo-responsivity to generate flexible MOFs (a) introducing azo-group covalently attached to the linker and exposed into the pore (b) schematic (b) inclusion of photo-responsive guest molecule into the pore of the MOF (Red and orange represent trans-AB and cis-AB, respectively) Under the effect of external stimuli such as light the conformational of guest molecule undergoes to impact the porosity of the structure. Reproduced with permission from Refs. [109,113]. Copyright 2011 Royal Society of Chemistry and 2012 American Chemical Society. Fig. 9. Several hydrocarbon adsorption (closed symbols)/desorption (opened symbols) isotherms collected at 25 °C on ZIF-7. Reproduced with permission from Ref. [122]. Copyright 2010 American Chemical Society. transitions or swelling upon adsorption [118]. Gate-opening have been utilized in separation of industrially important gas mixtures in several prototypical MOFs. Several members of the ZIF [ZIF = Zeo litic Imidazolate Framework], notably ZIF-7 and -8 have exhibited pronounced gate opening in presence of different gases, leading to excellent selectivity [83,119–126]. ZIFs are a subfamily of MOFs with tetrahedrally coordinated divalent metal centers, connected by imidazolate based linkers in such a fashion that the metal –N (from Imidazolate linker) – metal bond angle resembles to the SiAOASi angle of traditional zeolite molecular sieves. For a separation purpose, they are particularly interesting as they are more selective towards paraffin (i.e., ethane, propane) over olefins (e.g., ethylene, propylene) [122]. ZIF-7, which forms a microporous framework of sodalite (sod) topology, with a pore width of 3 Å shows preferential uptake of ethane over ethylene (Fig. 9) [122]. It is to be noted that the pore width of ZIF-7 is much smaller than for both ethane and ethylene, but it can accommodate both of them by means of framework flexibility. Both exhibit a three step type IV adsorption isotherm at 298 K for ZIF-7, with minimal uptake before the gate-opening pressure. The uptake increases after reaching gate-opening pressure and saturates at 1.75 mmol/g loading at 1 atm. The adsorption before gate-opening pressure occurs mostly on the external surface of the ZIF particles, followed by adsorption in the framework cavity after gate opening. The gate opening pressure for ethane is significantly lower than that of ethylene, indicating a preferential interaction between ethane and the framework. Such a preferential interaction at different pressure points leads to an ethane/ethylene’ reverse selectivity scenario, confirmed by subsequent breakthrough experiments. It is postulated that the threefold symmetry of the methyl groups of ethane allow the adsorbate molecule to interact favourably with the triangular pore window of ZIF-7, which allows ethane to adsorb at lower pressure. The lower gate-opening pressure for ethane is alternatively attributed due to the rotation of organic linkers. ZIF-8, another member of ZIF family, formed by a combination of tetrahedrally coordinated zinc metal centers and 2methylimidazole as a linker exhibit similar adsorption behavior with ethane more preferentially adsorbed than ethylene [123,126,127]. A follow-up in-situ powder XRD study shows that the imidazolate linkers within the framework have a swing effect at the gate-opening pressure point, causing the pore opening phenomenon. The higher adsorption amount of ethane over ethylene means the adsorbent has a preference for the higher molecular weight hydrocarbon, which indicates an absence of specific adoption sites with the framework (see Fig. 9). The higher loading of ethane is attributed towards to a CH


p type weak nonbonding interaction. Among flexible MOFs involving mixed linkers, a flexible MMOF (Microporous Metal–Organic Framework), namely Zn2(bpdc)2(bpee) (also known as RPM-3, RPM = Rutgers Recyclable Porous Material, bpdc = 4,40 -biphenyldicarboxylate; bpee = 1,2-bipyriyle thylene, also known as bpe) is shown to be capable of separating small hydrocarbon molecules (carbon number <4) by means of a pronounced gate opening behavior [128,129]. [Zn2(bpdc)2(bpee)] possess a three-dimensional flexible framework with permanent microporosity [128]. The material was shown to be exceptionally selective towards CO2 over N2 (selectivity >50) based on a gateopening mechanism [130]. The material exhibit enhanced separation of C1–C4 paraffins and both C2 isomer pairs (C2H2–C2H4 and C2H4–C2H6) at room temperature [129]. In-situ Raman spectroscopy and ab initio DFT (DFT = Density Functional Theory) calculations show that the gas-induced structural change and its associated gate-opening pressure originates as a result of hydrogen bonding between the adsorbate and the free C@O group of the 134 S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 Fig. 10. (a) Representation of PtS network of TetZB (b) CO2 kinetics at high pressure and 25 °C: 4 and 5 represent absorption and desorption. Selectivity of CO2 over N2 at 1 bar was shown (inset). Reproduced with permission from Ref. [132]. Copyright 2008 American Chemical Society. bpdc linker in the framework. Furthermore, the silver exchanged form of RPM-3 has also been shown to exhibit higher ethylene detection ability by p complexation pathway, as monitored by fluorescence and Raman spectroscopic methods [131]. Along the same line, Thallapally and co-workers reported a series of flexible MOFs (and a non-porous molecular solid) with interesting gas-adsorption properties [132–138]. For example, TetZB, a flexible MOF with PtS topology, composed of zinc paddle wheel cluster and a flexible tetrahedral organic linker, tetrakis [4-(carbox yphenyl)-oxamethyl]methane shows a stepwise CO2 adsorption at high pressure (10 bar) indicative of framework flexibility (Fig. 10) [132,137,138]. Variable temperature powder XRD under vacuum shows a considerable change in the powder pattern, possibly due to framework contraction upon solvent removal. The framework transforms backs to the as-synthesized form in presence of 200 psi of CO2 at room temperature. TetZB also exhibits pronounced breathing phenomenon for alkanes (n  4) below 313 K [138]. For example, after the first uptake of butane at low relative pressure the isotherm reaches its first plateau (at P/P0 = 0.04), followed by a step adsorption with approximately three times higher capacity of butane at P/P0 = 0.2. This step-wise adsorption and subsequent hysteresis during desorption steps is attributed to a narrow pore (np) to large pore (lp) transition: after the first fraction of adsorbate is adsorbed, the narrow pores in the framework are saturated or closed. Sorption of additional adsorbate at high relative pressures reopens the framework, thereby accommodating additional solvent molecules in the newly expanded pores. The np ? lp transition depends on the nature of adsorbate, as evident by the adsorption–desorption curves of larger alkanes, e.g., pentane, hexane, heptane and octane. Similar step-wise adsorption was also observed for polar adsorbates such as ethanol, methanol and isomers of propanol [137]. TetZB was shown to possess the capability of separating mixtures of these polar adsorbates (e.g., propanol isomers) by exploiting the differences in the saturation capacities of each component as evident from transient breakthrough simulations. Another flexible MOF reported by Thallapally and co-workers, namely FMOF-2 showed pronounced gas-induced expansion–contraction behavior (Fig. 11) [135,136]. FMOF-2 is composed of zinc based SBU, a V-shaped fluorinated linker, 2,2-bis(4-carboxyphe nyl)hexafluoropropane, forming a three-dimensional two fold interpenetrated framework. The material show stepwise adsorption for acidic gases such as SO2 and H2S between 0 and 1 bar and for CO2 at a higher pressure (10 bar). Similar flexible MOF with interesting gas-adsorption and separation behavior was recently reported by Li and co-workers [139]. They reported an interesting example of a gas-induced gateopening process in a microporous Metal–Organic Framework, [Mn(ina)2] (ina = isonicotinate) by a combination of several analytical techniques including single crystal X-ray diffraction, in situ powder X-ray diffraction coupled with differential scanning calorimetry (XRD-DSC), and gas adsorption–desorption methods. The as-synthesized framework has a non-cylindrical onedimensional channel of a pore width between 3.0 and 3.7 Å. Upon removal of solvent ethanol, the channel of the activated framework shrink to a maximum pore width of 3.0 Å due to rotation of the isonicotinate linker, thus making it theoretically inaccessible to even small gas molecules such as CO2. However, subsequent CO2 adsorption experiments show that the activated [Mn(ina)2] exhibit a S shaped gas-adsorption curve with hysteresis, typical for a gateopening behavior. Gate-opening was also observed for C3 hydrocarbons but the onset of gate-opening occurs at a much lower pressure (0.005 bar) than for CO2 (0.4 bar), indicating strong interaction between C3 molecules and pore surface. The framework shows exceptionally high CO2/CH4 separation properties: At 278 K and a total pressure of 1 bar for an equimolar binary mixture of CO2 and CH4, CO2/CH4 selectivity was calculated to be >600. The high selectivity is reasonable considering that CO2 is able to induce the gate-opening at a relatively low pressure, while CH4 is not able to trigger the gate-opening, leading to essentially no adsorption under similar experimental conditions. Single-crystal data of CO2 adsorbed and C3 (C3H8) adsorbed activated framework show that upon adsorption, organic linkers rotate back to their original position of the as-synthesized framework. The crystal structures of CO2 loaded and C3H8 loaded framework show that the adsorbate molecule was situated inside the channel roughly at the same location of the solvent ethanol molecule of the as-synthesized framework. Specifically, the crystal structure of the CO2 loaded activated framework suggest that the atoms of the adsorbed CO2 do not interact with the hydrogen atoms in the aromatic rings, in contrast to the C3H8-loaded material where H atoms from the propane adsorbate interact with carboxylate O atoms on the pore surface of the four-sided one-dimensional framework. Simultaneous insitu XRD-DSC method was further employed to elucidate the gate-opening mechanism in detail: For example, the largest exothermic peak was observed during the gate-opening transition at the CO2 pressure of 0.7–0.8 bar. Before gate-opening, each step appears to release similar amount of heat with a small shift of XRD peaks towards lower angle. The gradual shift toward lower angle suggests a continuous expansion of a unit cell with increasing adsorbate pressure up to 0.8 bar, above which the peaks split into two, indicating a structural transformation or gate-opening. S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 135 Fig. 11. (a) Twofold interpenetrated molecular box with helical channels with a void space of 133 Å3 (b) CO2 sorption kinetics of FMOF-2 at room temperature. Reproduced with permission from Ref. [136]. Copyright 2010 American Chemical Society. After the gate-opening step, the exothermic peak was much higher than before, suggesting that gate-opening process allows a larger amount of gas molecules to enter the pore space, consistent to isotherm measurements. Because of thermal motion of individual atoms, all linkers that are used for construction are somewhat flexible, even though the linker itself is colloquially considered rigid. For example, the rotational motion of rigid benzene based linker could play a role into the adsorption and separation properties of benchmark rigid MOFs such as IRMOFs and MIL-47(V). Indeed sophisticated spectroscopic methods such as proton NMR were applied to understand the dynamic motion of the organic linkers in solid state [140–143]. For example, solid-state 2H NMR unveils that the aromatic rings in MIL-47(V) and MIL-53(Cr) perform p flips about their symmetry axis [140]. The aromatic benzene rings flip faster and with lower activation energy in the flexible MIL-53(Cr) than in the rigid MIL-47(V). Such p flipping of the benzene ring was also observed in a number of isostructural MOFs, namely [M2(bdc)2(dabco)]
G [M = Co2+, Ni2+, Cu2+, Zn2+; dabco = 1,4-diazabicyclo[2.2.2]-octane; G = none or dimethylformamide, DMF] [142]. It was found that 136 S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 the dynamics of the benzene rings is insensitive to the variation of the metal cations, whereas the loading of the guest DMF molecules provides both a significant decrease of the rate of p-flips and an increase of the activation energy for the motion of the benzene rings. It is believed that for macroscopic measurements, the rotational dynamics of benzene could play a role in the adsorption and transport properties of adsorbate molecules. Similarly, 1H relaxation experiment was conducted to measure the lattice dynamics of IRMOF-3 an amino functionalized IRMOF [144]. These results indicate a low energy process associated with rotation of the ANH2 group, with activation energy of 1.8 ± 0.6 kcal/mol, and full 180° rotation of the benzene ring with activation energy of 5. 0 ± 0.2 kcal/mol. As mentioned in various places of this review, rotation of the flexible linkers upon activation and (or) adsorption of different adsorbates is the prime reason of detectable flexibility in MOF materials. A benchmark example of that is found in MIL series, especially those belonging to MIL-53 family [M(OH)(O2CC6H4CO2) (M = Al, Cr, Fe, Ga)]. MIL-53(Cr) has been shown to exhibit different crystalline states, corresponding to different pore openings based on the guest entrapped in the pores. The assynthesized form has terephthalic acid molecules in the pores and has a cell volume of 1440 Å3. Upon calcination, the free acid is removed and the cell volume increases to 1486 Å3, but decreases to 1012 Å3 on hydration. This transition between lp to np forms is found to be reversible, as evident by X-ray powder diffraction, adsorption microcalorimetry, and computer simulations. The flexibility is also dependent on the nature of the adsorbate, with MIL53(Cr) behaving as a rigid framework for adsorption of smaller adsorbate molecules such as H2 and CH4, while it exhibits flexibility upon adsorption of larger adsorbates such as Xe. Apart from MIL-53 series, a plethora of examples are known in literature where the flexibility arises due to nature of the organic linkers, metal–linker connectivity and nature of the secondary building unit (SBU). For example, Dai and co-workers [145] reported a flexible MOF, [Zn3(btca)2(OH)2]
(guest)n (H2btca = 1,2,3-benzotriazole5-carboxylic acid) that exhibits guest molecule controlled dynamic gas adsorption, where CO2 rather than N2, He, and Ar induce a structural transition with a corresponding appearance of additional steps in the single component gas-adsorption isotherms at different temperatures and pressures, confirmed by Fourier transform infrared spectroscopy and molecular simulations. Simulation details show that the structure opens quite easily for CO2, reaching a plateau where sorption becomes more difficult, and then opens even more at slightly higher pressure. The organic linker possesses few internal degrees of freedom and forms a rigid coordination to the metal center. In combination with two different binding modes at each end, the linker is postulated to be responsible for the observed structural transition. The rotational degrees of freedom of organic linkers, especially when it involves organic side chains play a significant role in the overall flexibility of the MOFs, as shown by Fisher and coworkers [106] on an important family of pillared-layered MOFs based on the parent structure [Zn2(bdc)2(dabco)]n [bdc = 1,4-benze nedicarboxylate; dabco = 1,4-diazabicyclo[2.2.2]octane]. A library of MOFs with functionalized bdc-type linkers (fu-bdc), which bear alkoxy groups of varying chain length with diverse functionalities and polarity at different position of the benzene core were synthesized. The parent [Zn2(bdc)2(dabco)]n is only weakly flexible but the substituted frameworks of [Zn2(fu-bdc)2(dabco)]n contract drastically upon guest removal and expand again upon adsorption of DMF, EtOH or CO2. On the contrary, N2 is not adsorbed and does not open the framework. As expected, the physical properties of the materials such as their gas-adsorption behavior are dependent on the nature of the side chain substituent of the linker. Furthermore, the use of two differently functionalized linkers in varying ratio leads to the formation of single-phased [Zn2(fu-bdc0 )2x(fu-b dc00 )22x(dabco)]n MOFs (0 < x < 1), that exhibit a non-linear dependence of their gas sorption properties on the applied ratio of components. As a result, the responsive behavior of these mixed components, single phased MOFs can be extensively tuned via an intelligent combination of functionalized linkers. So far our discussions on MOF flexibilities are centered around either on the inherent nature of the MOF structure or external parameters such as adsorbate, pressure, temperature (activation). An intriguing case of MOF flexibility arises for MOFs that leads to change in physical properties their adsorption related properties upon illumination [146]. A straightforward strategy is to introduce light-sensitive, photo-switchable linkers such as those with azobenzene moieties for MOF construction, that can switch from one isomeric form to the other by means of light exposure, leading to changes in adsorption–desorption properties. For example, azobenzene based organic moieties interconvert from trans (E) to cis (Z) configuration via rotation of the aromatic rings around the central AN@NA double bond upon irradiation. Similar structural effect can also be observed for light-sensitive guest molecules as well. For example, a three-dimensional MOF, namely [Zn2(terephthalate)2(triethylenediamine)]n, composed of two-dimensional square grids bridged by triethylenediamine, show flexibility depends on the presence of light-sensitive linker azobenzene. The cis/trans isomerization of the guest azobenzene by light triggers a structural transformation, resulting in a drastic switching of the adsorption property of the host–guest composite. N2 adsorption measurements exhibit a dramatic increase of the adsorption amount upon UV irradiation because of the pore expansion effect due to guest molecule isomerization (Fig. 12). Thomas and co-workers [147] reported a series of flexible 3-fold interpenetrated lanthanide-based MOFs with general formula Ln (HL)(DMA)2]
DMA
2H2O [Ln = La, Ce, Pr, Nd, Sm, Eu, Gd, Tb, Dy, Er DMA = dimethylacetamide, H4L = 5,50 -(2,3,5,6-tetramethyl-1,4phenylene)bis(methylene)bis(azanediyl)diisophthalic acid]. The authors further tested the Sm-analogue as a representative example of the series. The activation of the Sm-analogue leads to the closing of the one dimensional channel (13.8  14.8 Å) with coordinated DMA molecules rotating into the interior of the channels with a change from trans- to cis-Sm coordination (see Fig. 13). Such movement leads to unit volume shrinkage of 20%. The activated material shows gate-opening behavior for selected adsorbates like dichloromethane and CO2 but not for other common adsorbates such as CH4, H2, O2 and N2. Upon gate-opening the activated structure goes back to the trans-configuration similar to that of assynthesized framework as revealed by single-crystal X-ray diffraction (SC-XRD). Moreover, CO2 adsorption kinetics reveals that the gate-opening occurs in a co-operative manner with CO2 diffusion along the pore with the progress of structural change. Recently Long and co-workers [148,149] reported an elegant example of gate-opening MOFs, where flexible MOFs such as Co (bdp) and Fe(bdp) [bdp2 = 1,4-benzenedipyrazolate] were shown to achieve higher usable CH4 capacity than for classical, rigid adsorbents such as zeolites and activated carbon. CH4 is the principal component of natural gas. While natural gas is considered as a cleaner, cheaper and widely available fuel that has considerable environmental and economic advantages over petroleum as a source of energy for the transportation sector, the low volumetric energy density of natural gas at operating pressure and temperature leads to substantial challenge for passenger cars where little room is available for on-board fuel storage. In case of M(bdp) (M = Fe, Co), the framework undergoes a structural phase transition from ‘close’ to ‘open’ in response to specific CH4 pressure, resulting in adsorption and desorption isotherms that feature sharp steps. The solvated form of M(bdp) possess one-dimensional chains of tetrahedral M2+ cations, that are bridged by m2-pyrazolates, forming a three-dimensional S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 137 Fig. 12. Structure of pillared-layered [Zn2(fu-bdc)2(dabco)]n networks. (top-left) View along the linker (fu-bdc) axis. (top-right) View along the pillar (dabco) axis. The flexible substituents are represented as green spheres. Zn, O, N, and C atoms are shown in yellow, red, blue, and gray, respectively. Zn coordination polyhedra are shown in yellow (bottom) CO2 (195 K) and N2 (77 K) sorption isotherms of [Zn2(fu-bdc)2(dabco)]n networks as a function of linker side chains. Reproduced with permission from Ref. [106]. Copyright 2012 American Chemical Society. framework with square channels of edge length of 13 Å. N2 adsorption isotherm of the evacuated framework show five distinct steps, attributed to four structural transitions during ‘closed’ phase to ‘open’ phase form. The ‘open’ phase has a Langmuir surface area of 2911 m2/g. During high-pressure CH4 adsorption, the gate-opening step for the co-analogue occurs at 16 bar, with minimal CH4 adsorption was observed before this pressure. During desorption, hysteresis was observed but the loop closed at 7 bar, with less than 0.2 mmol g1 uptake below that pressure. For the Fe analogue, the total CH4 uptake is comparable but the adsorption–desorption occur at considerably higher pressures of 24 bar and 10 bar respectively. The higher gate-opening pressure indicates a higher energy requirement for phase transition. As it is estimated that a commercial on-board natural gas tank will operate between 5 and 65 bar, adsorbents that have significant CH4 uptake below 5 bar is bound to have a lower useable CH4 capacity than the ones with minimal CH4 capacity below 5 bar. For example, the usable CH4 capacity of Co(bdp) at 25 °C is 155 cm3 STP cm (v/v) for adsorption at 35 bar and 197 v/v for adsorption at 65 bar: this is the highest values of usable CH4 capacity reported so far for any adsorbent under these conditions. More importantly, the energy requirement of the phase transition potentially offset the heat generated during CH4 adsorption–desorption: A fraction or all of the heat (energy) generated during the exothermic CH4 adsorption process can be utilized to transform the thermodynamically stable ‘closed’ phase to porous ‘open’ phase in an endothermic process, thus eliminating any need for additional thermal management system. The efficient management of heats, generated during adsorption–desorption process for a gas-storage application is crucial, as excess heat reduce the usable capacity of a given adsorbent. In-situ synchrotron based powder XRD coupled with ab initio structure solution study on Co(bdp) showed the transition of the ‘closed’ phase to the ‘open’ phase in presence of CH4. It was revealed that the although the metal center holds similar coordination geometry upon phase transition, during the phase transition, the angle between the planes of the pyrazolate moieties increase and CoAN bond length decrease as the framework expanded in presence of CH4 gas to form a three-dimensional framework with uniform pore channel, similar to the as-synthesized solvatedframework. In the activated ‘closed’ framework the central benzene ring of the linker molecule twists out of plane of the two pyrazolate rings by 25°, resulting in edge-to-face p–p interactions with four neighboring benzene rings. Such close contact between neighboring linkers leads to no accessible porosity. Disrupting these p–p interactions requires chemical modification of the linker molecule: such as addition of side chains. Indeed, Long and co-workers [149] further tune the adsorption-induced phase change in Co(bdp) by introducing functionalized bdp linker (X-bdp, X = -F, p-F2, -o-F2, D, p-Me2, o: ortho, p: para, Me: methyl) to form isostructural series of Co(X-bdp) framework. These frameworks exhibit similar structural flexibility, transitioning from a low-porosity, collapsed phase to high-porosity, expanded phases with increasing gas pressure. Powder X-ray diffraction studies show that fluorination of the benzene ring disrupts edge-to-face p–p interactions, which work to stabilize the collapsed phase at low gas pressures, while methylation strengthens the interaction. High-pressure CH4 adsorption isotherms exhibit that the gate-opening pressure of the framework can be systematically controlled by linker functionalization: frameworks formed by linkers that disrupt the edge to face interaction ‘open’ at a lower CH4 pressure, while frameworks with linkers that strengthen the edge- to -face interactions need a high 138 S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 Fig. 13. Structural transformations between trans-[Sm(HL)(DMA)2]
DMA
2H2O and cis-[Sm(HL)(DMA)2]. Reproduced with permission from Ref. [147]. Copyright 2015 American Chemical Society. gate-opening pressure (see Fig. 14). Overall these two papers by Long and co-workers show that first of all, flexible frameworks with interesting gate-opening properties can be useful for ‘reallife’ application such as on-board natural gas storage for passenger cars, secondly the gate-opening pressure and thus the useable CH4 storage can be controlled by means of ligand functionalization. 4.2. Phase change There are two types of phase transitions depending on factors such as whether any bond breaking or bond formation is involved during breathing motion. They are termed as (a) displacive phase transitions and (b) reconstructive phase transitions respectively [150]. 4.2.1. Displacive phase transitions This is obvious for the structures that have weak interactions between their layers. In this case, while the framework connectivity remains the same during the phase change, the intraframework interaction changes considerably. Several examples illustrate this case such as [Cu(4,40 -bipy)(dhbc)2]
2H2O [94], (ZnI2)3(tpt)2 [56], [Cu2(bdc)2(bipy)]n [151], [Zn(ip)(bipy)]n [152] and [Ni(bpe)2(N(CN2))][(N(CN2))(H2O)5]n [153] (tpt = 2,4,6-tris(4pyridyl)triazine, ip = isophthalate), where a slipping motion of the interpenetrated or interdigitated layers takes place under the S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 139 Fig. 14. (a) Powder X-ray diffraction structures of Co(p-F2-bdp) under vacuum (top) and under 20 bar CH4 (bottom). Gray, blue, white, purple, and green spheres represent C, N, H, Co, and F atoms, respectively. (b) Low pressure N2 adsorption for Co(bdp) and derivatives at 77 K. Co(bdp) adsorption (black) largely underlays Co(D4-bdp) adsorption (green). (c) High-pressure CH4 adsorption isotherms of Co(bdp), Co(F-bdp),] Co(p-F2-bdp), Co(o-F2-bdp), and Co(D4-bdp) at 25 °C. Reproduced with permission from Ref. [149]. Copyright 2016 American Chemical Society. effect of stimulus such as applied pressure [94], incorporation of nitrobenzene [56], solvent and guest (e.g. methanol). Such displacive phase change is noticed in [M2(bdc)2(dabco)]n (see Section 2.1. for detailed structural description) where the incorporation of three molecules of propan-2-ol per unit cell transform the original tetragonal unit cell to monoclinic with a 21% decrease in cell volume (Fig. 15). Such decrease might be attributed to the shift of the square grid sheets from one to the other. Interestingly, the incorporation of 4.5 molecules of 2-propanol per unit cell leads to regeneration of the original unit cell [71,154]. 4.2.2. Reconstructive phase transitions This behavior is reported in MOFs where a reversible cleavage (or formation) of metal-functional groups takes place [49,50]. There are few reported examples that undergo reconstructive phase change such as [Ni2(4,40 -bipy)3(NO3)4] [117,155,156], and [Cu2(pzdc)2(dpyg)]
8H2O (dpyg = 1,2-Di(4-pyridyl)glycol) [157]. The copper-based pillared framework, [Cu2(pzdc)2(dpyg)]
8H2O, undergoes a cleavage/formation of CuAO (carboxylate) bonds which causes the transformation upon the adsorption process. With the adsorption of methanol and water, the structural connectivity is restored again. The desorption process leads to a 27% contraction of the unit cell that expand again after adsorption of guest solvent molecules. The expansion is attributed to the strong interaction of the guest molecules with the organic linkers through Hbonding. It was found that it remains non-porous to CH4 due to very weak CH4-framework interaction, originating due to the hydrophobic nature of CH4 molecules [61,157]. Another exquisite example reported by Rosseinsky and his co-workers, showing a reversible structural adjustment upon desolvation [158]. The 140 S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 Fig. 15. The adsorption of propan-2-ol in [Zn2(1,4-bdc)2(dabco)] 1. The parameters k, l, m, and n are estimated from the unit-cell parameters obtained from the X-ray analysis. Reproduced with permission from Ref. [154]. Copyright 2007 Wiley-VCH. as-synthesized crystals comprised of square-pyramidal Zn paddlewheel and pyrene-based ligand exhibits a porous 2D layered network. Upon desolvation, the structure undergoes a significant change, leading to the formation of a 3D porous structure with a tetrahedral Zn nodes. The transition was confirmed by 13C crosspolarization (CP) magic-angle-spinning (MAS) NMR spectroscopy which can clearly reveal the coordination modes of the benzoate of the pryrene-based ligand. The 3D desolvated structure exhibits a BET surface area of 523 m2/g calculated from CO2 adsorption isotherm collected at 195 K. The structure also responds differently to the different isomers of xylene: a conventional isotherm was displayed upon adsorption of the p-xylene, while the material showed a higher m-xylene adsorption with hysteresis which suggests a further structure transformation that allows the packing of more gas in the pores. Kitagawa and co-workers were also observed a structural transformation in nanoporous coordination polymer upon desolvation or admission of guest molecule [159]. The 2D Kagome-type layered structure encompassed Cu-paddlewheel nodes and 5-azidoisophthalate undergoes structural transformation upon removal of water molecules from the axial positions of Cu-paddlewheels which triggers the coordination of one of the adjacent carboxylate oxygen to this empty site. The resulting 3D structure has a narrower pore due to the inclination of the aromatic planes of the organic ligands by virtue of the carboxylate groups’ twisting. This transformation process was found to be reversible upon removal or introduction of water molecules where the 3D dried structure can be changed back to the original 2D structure after exposing to the air containing moisture. The material was examined for the N2 and CO sorption and separation coincidentally with the study of structural dynamics upon the sorption of gases. The gas sorption measurements revealed an unusual step CO sorption isotherm at 120 K, with CO uptake of 63 ml (STP)/g at the first step (5 kPa) and an additional 112 ml (STP)/g for the second step for up to 80 kPa. This behavior was not observed in the adsorption of N2, but it was clearly shown that the material can adsorb CO (175 ml (STP)/g) much higher than N2 (71 ml (STP)/g) at 80 kPa. In-situ powder X-ray diffraction (PXRD) was used to monitor the structural changes upon on the gas adsorption. The structure showed no changes of PXRD patterns upon the adsorption of N2 at all pressure and CO at low pressures blow 5 kPa. By increasing the loading pressure of CO gas the peaks of the original PXRD pattern started to disappear and another peaks showed up, indicating the structural change. It was found that the resulting PXRD pattern in the second step adsorption of CO gas (after 5 kPa) was matching to that of the as-synthesized structure before desolvation, which indicated that the desolvated structure returned to a similar structure as the as-synthesized material upon increasing the pressure of CO to accommodate more gas in the channel. The authors performed a Rietveld analysis using the synchrotron PXRD data on the material under a pressure of CO gas (50 kPa) at 100 K. The analysis revealed the formation of Cu2+–CO complex in the structure at 50 kPa of CO (corresponding to the second step adsorption) which lead to the structural transformation to boost the adsorption of additional CO molecules. The same observation was seen in a flexible PCP known as SNU-9 of formula Zn2(BPnDC)2(bpy) (BPnDC = benzophenone 4,40 -dicarboxylic acid, bpy = 4,40 -bipyridine) [160]. The structural transformation takes place during the activation due to the change of the coordination environment of the Zn nodes prompted by the rearrangement of ZnAO bonds which leads to the decrease in the volume of unit cell by almost 28%. The structure also behaved differently upon adsorption of CO2 and N2 gases, while it showed a quasi-one-step in N2 adsorption isotherm at 77 K and a conventional isotherm in case of CO2 adsorption at 195 K. S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 4.3. Negative gas adsorption (NGA) Negative gas adsorption (NGA) is a unique extremely rare phenomenon in flexible MOFs while there is only one report showed this behavior recently [161]. Primarily, DUT-49 (Dresden University of Technology) was found to exhibit remarkable methane adsorption performance for natural gas vehicles as exemplified by its exceptionally high methane storage capacity of 308 g/kg at 298 K and 110 bar [162]. It was common that when the gas uptake increases by increasing the pressure. However, when DUT-49 material fills with gas (methane and n-butane) and the pressure increases, the structure releases a significant amount of gas at certain temperature and pressure range [161]. The gas sorption measurements combined with in-situ powder X-ray diffraction study and the computational simulations revealed that after 10 kPa all the pores were occupied by the guest molecules triggering the sudden structural deformation along with pore contraction upon increasing the pressure that lead to the release of guest molecules [161,163]. The drop in the methane uptake ‘‘DnNGA” was 8.62 mmol/g reflecting the negative gas adsorption at 111 K and 10 kPa/g. The amount of released gas can cause a pressure amplification ‘‘DPNGA” in the measuring cell of 2.27 kPa. The NGA step was also noticed during the adsorption of n-butane at 30 kPa and 111 K. In contrast, N2 gas collected at 77 K showed type I isotherm without any step which revealed the subtle impact of the guest molecule on the flexibility of the framework. The behavior is unique but we expect to be investigated and understood more deeply in the future. Indeed, this massive response can be beneficial for many technological systems that can respond selectively to variations in the environment. 5. Methods for controlling the flexibility of MOFs 5.1. Metal ions The utilization of certain metal centers can block the breathing of the structure [51]. In comparison of MIL-53(Cr) and MIL-53(Fe); MIL-53(Cr) has a better ability for pore opening upon dehydration. The interaction of the unpaired electrons of the chains of metal centers in MOFs might participate in the stability of the framework which is stronger in case of Fe (d5) than Cr (d3) [74]. In this regard, Elsaidi et al. recently reported the role of ligand flexibility on adsorption of Xe in SIFSIX-3-Ni, a nickel based ultra-microporous MOF that belong to SIFSIX family [164]. SIFSIX-3-M networks are 2 formed by M(pz)2+ anions 2 type square grids, connected by SiF6 and form primitive cubic, pcu, topology frameworks with pore diameters of 3.5–3.8 Å. The pure component adsorption–desorption curve of Xe between 278 K–298 K and 0–1 bar show an inflection point, while no such inflection point was observed for other adsorbates such as Kr, CO2, O2, N2, Ar under similar experimental conditions. The kinetic diameters of these adsorbates are smaller than that of Xe and it was postulated that the movement of the pyrazine ring in presence of similar sized Xe atom leads to the inflection point which was further confirmed by in-situ synchrotron based powder XRD and molecular simulation. It was found that the pyrazine ring can adopt a number of configuration between +16° to 16° upon its axis and the inflection of the Xe adsorption isotherm arises due to a disordered to ordered transition of these configurations of the pyrazine rings in presence of Xe adsorbate as opposed to other phenomena such as guest- guest interactions and breathing, resulting in a non-Langmuirian isotherm with an inflection point. During the transition, the rings organize their rotational configurations to achieve a more efficient adsorbate-adsorbent interaction. In contrary, changing the metal center impacts the dynamics of the framework as other SIFSIX-3- 141 M analogues (M = Zn, Co, Cu and Fe) show no inflection in the Xe adsorption isotherm. This suggests that the inflection and the dynamics are highly sensitive to any subtle change in the crystal structure [164]. 5.2. .Functional groups Among different types of flexible MOFs, MIL-88 is known for inherent flexibility and several attempts to control over the breathing of this class of materials have been reported in literature [51,82]. MIL-88 represents the default net (acs) for 6-c trigonal prismatic nodes: it is formed by linking the vertices of [M3(m3-O) (RCO2)6] (M = Fe) trigonal prismatic moieties by linear linkers [70]. It has the same type of metal clusters such as MIL-101 [90] and MIL-100 [165] shown to exhibit fine-tunable mesopores (25– 34 Å) and ultra-high surface areas (3100–5900 m2/g). However in case of MIL-88, the linear connection of the [Fe3(m3-O)(CO2)6] facilitates the shrinkage upon removal of guest. It can undergo large structural changes upon inclusion of different guests depending on the nature and/or the length of organic guest molecules with large variations of unit cell volume from 70% to 230% upon solvent inclusion [17,70,166–168]. Prevention or reduction of these breathing phenomena for MIL88 has attracted a significant scientific interest [169]. MIL-88B (ACH3) and MIL-88B(Sc) have shown higher permanent porosity compared with other acs nets but they only exhibit BET surface areas of 1220 and 634 m2/g, respectively, considerably lower than other high surface area MOFs such as NU-1000 (NU = Northwes tern University). 28 The improvement in permanent porosity in comparison to its nonporous parent MOF-235/236 was attributed to restricted flexibility of the latter [170]. It is therefore anticipated that 6,6-c acs nets which are completely rigid should exhibit a much higher surface area as their hexagonal pore opening would remain unaltered upon solvent removal. In this regard, Férey and co-workers reported a useful strategy that could lead to control over the breathing amplitude of the flexible MOFs: the introduction of bulky side chains at the flexible edge of the MOFs, which reduce or completely prevents the breathing upon removal of solvent molecule [169]. Indeed, using this approach, researchers were able to tune the flexibility of MIL-88 by introducing functional groups with different bulkiness onto the phenyl rings of the bdc and 4,4-bpdc linker (Fig. 16). For example, the presence of four ACH3 groups on the aromatic linker in MIL-88B significantly enhance the permanent porosity, while the modification of the linker of MIL-88-D with 2 or 4 methyl groups facilitate the pore opening in the presence of solvent. Another route for controlling over the flexibility in MOFs using the functional groups was recently reported by Kaskel and coworkers [171]. In contrary to the functionalization of organic linkers, the functionalization of the SBU takes place for [Zn3(bpydc)2(HCOO)2] (bpydc = 2,20 -bipyridine-5,50 -dicarboxylate. Different types of monocarboxylic acids such as acetic, benzoic and cinnamic acids were used for the functionalization. The parent structure has two formic acid anions directly coordinated to the metal cluster. The structural dynamic of [Zn3(bpydc)2(HCOO)2] and its adsorption behavior changes completely upon replacing the formic acid in the synthesis with monocarboxylic acids of different carbon backbone. The data is verified by in-situ adsorption/PXRD that show that the different functional groups tethered to the SBU impact the mechanism of structural transformation. The structural transformation changes from ‘‘gate opening” in the case of formic acid to ‘‘breathing” for benzoic acid. Thus, as the number of the carbon atoms backbone coordinated to SBU increases, the gate pressure needed for the structural transformation decreases which is evident from N2 adsorption isotherm (see Fig. 17). 142 S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 Fig. 16. Schematic representation of the impact of linker modification on the flexibility of the MIL-88B and D modified solids as a function of the number of functional groups (X) per spacer and its effect on the Amplitudes of swelling. Reproduced with permission from Ref. [169]. Copyright 2011 American Chemical Society. 5.3. Host–host interaction If the host–host interactions within the pores/cages of MOFs are stronger than guest–guest or guest–host interaction, the breathing or the dynamic movement of the structure will be forbidden [61,82]. There are several host–host interactions reported in the literature that play important role in the control over the flexibility of MOFs. 5.3.1. p–p interaction The interaction of the p electron clouds of the aromatic rings facing each other might lead to the prohibition of the breathing or opening of the channels. The p–p interaction can be considered as strong if the distance between the rings lies in the range of 2.9 Å–3.5 Å. MIL-53(Fe) has a stronger p–p interaction than MIL-53 (Cr), as the distance between phenyl rings for the MIL-53(Fe) is between 3.41 Å (activated form) and 3.73 Å (hydrated form). The distance between phenyl rings for MIL-53(Cr) was measured to be 4.1 Å [74]. Another example in which the breathing of structure is forbidden due to the strong p–p interaction is MIL-69, where the distance between aromatic rings are recorded as 3.79 Å (Fig. 18) [172]. 5.3.2. Hydrogen bonding Elsaidi et al. recently reported a series of non-breathing, permanently porous materials by a two-step crystal engineering method [173]. These material possess [M3(m3-O)(CO2)6] [M = Cr] based trigonal prismatic clusters as primary building blocks. The combination of decorated trimeric MBBs together with [Cu3(m3-Cl) (NH2)6Cl6] forms non-breathing frameworks tp-PMBB-6-acs-1 (PMBB = Trigonal Prismatic Primary Molecular Building Block) and tp-PMBB-6-stp-1 with acs and stp topology respectively. The tp-PMBB-6-acs-1 represents the first example of an acs net formed from two different 6-connected MBBs. The stability of tp-PMBB-6acs-1 was evaluated by means of variable temperature powder XRD and in different solvents. In contrast to MIL-88 series, tpPMBB-6-acs-1 shows no shifts in peak positions upon heating under vacuum or exposure to different solvents. The rigidity of tp-PMBB-6-acs-1 is attributed to the [Cu3(m3-Cl)(NH2)6Cl6] MBBs where each maba (maba = m-aminobenzoate) linkers are locked through additional NH


Cl hydrogen bonds (Fig. 19). The material exhibit a permanent porosity (710 m2/g based on CO2 adsorption at 195 K) upon removal of guest molecules from the hexagonal channel. MOFs with acs nets in terms of the surface area and swelling amplitude are shown Table 1. 5.4. Post-synthetic modification Sumby and co-workers [177] reported an intriguing case of MOF flexibility, where a flexible manganese based MOF, namely, [Mn3(L)2(L0 )] (L and L0 are crystallographically unique forms of the deprotonated ligand bis(4-(4-carboxyphenyl)-1H-3,5-dimethyl pyrazolyl)methane, LH2] were shown to undergo cooperative flexibility induced post-synthetic metallation. It has a charge-neutral three-dimensional network with two-dimensional layers composed of trinuclear Mn3(L)2 nodes that are ‘pillared by the L0 form of the linker. L0 bridges the two-dimensional layers through the carboxylate donors, giving rise to channels measuring 8.5  10. 5 Å (excluding van der Waals radii) along the crystallographic c axis. As the formula suggests, the MOF has two types of S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 143 Fig. 17. Right: (a) View of [Zn3(bpydc)2(HCOO)2] structure along the c axis. (b–e) SBUs in the isomorphous frameworks with different monocarboxylic acids. Left: N2 sorption isotherms collected at 77 K in a linear (a) and semilogarithmic (b) scale for the isomorphous frameworks with different monocarboxylic acids: Formic acid (black circles), acetic acid (blue triangles), benzoic acid (red squares) and cinnamic acid (green diamonds). Reproduced with permission from Ref. [171]. Copyright 2016 Royal Society of Chemistry. Fig. 18. The structure of MIL-69 showing (a) its channel having water molecules along c axis. (b) the coordination around the AlO4(OH)2 cluster. Reproduced with permission from Ref. [172]. Copyright 2005 Elsevier. crystallographically unique linkers: in the L form carboxylate and pyrazole donors coordinate the Mn atoms of the metal node, while in the L0 form only the carboxylate donors contribute to the coordination environment of the metal node, thus leaving the dipyrazole moieties vacant within the MOF. As a result, L0 provides a di-pyrazole type chelating unit for post-synthetic metal binding. The MOF itself possess significant structural flexibility, as evidenced by gas-adsorption and X-ray diffraction. The activated structure undergoes a structural transformation upon desolvation as depicted by single crystal XRD: the pillaring linker undergoes a significant torsional rotation, resulting in a noticeably more acute bridging angle (97.5° compared to 105.1°). Such rotation leads to a contraction of the two-dimensional planes, attributed to the flexibility and rotational freedom of the non-coordinated L0 linker within the MOF structure (see Fig. 20). The open coordination sites of the L0 linker further undergoes quantitative metallation with a number of first- and second-row transition-metal ions [Co(II), Cu(II), Zn(II), Rh(I), Cd(II)], forming coordination complexes. The coordination sphere of these metal complexes was verified by analytical technique including ICP-MS, EDX and single crystal XRD. In most post-synthetic modification (PSM) scenario, such identification of the post-synthetic coordination complexes using single crystal XRD is rare, as the additional post-synthetic step generally leads to lower crystallinity even for chemically robust MOFs. 144 S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 Fig. 19. Top: Schematic representation of tp-PMBB-acs-1 and tp-PMBB-stp-1, Bottom: (a) hydrogen bonding in [Cu3(m3-Cl)(RNH2)6Cl6] MBBs; (b) hydrogen bonds in tpPMBB-5-acs-1; (c) hydrogen bonds in tp-PMBB-stp-1 (CH hydrogen atoms have been omitted for clarity). Reproduced with permission from Ref. [173]. Copyright 2013 Royal Society of Chemistry. Upon metallation, the metallated MOF undergoes a flexible to rigid transformation, as verified by the N2 BET data at 77 K: the N2 adsorption isotherm of the cobalt-metallated MOF exhibit a BET surface area of 1045 m2/g without any low-pressure step, in contrast to the non-metallated activated MOF. Metallation reaction was carried out by heating the activated MOF in presence of the metal salt in an appropriate solvent and temperature. For example, Co-metallation was carried out using a CoCl2
6H2O solution at 65 °C. ICP-MS data show that that quantitative metallation was achieved in approximately 3 h. During the metallation process, the crystals of activated MOF change color to deep blue, but changes to pink upon cooling to room temperature. Such color change was also observed in acetonitrile solvent. The color change is attributed to the temperature-dependent transformation of the metal coordination. The pink form, as revealed by single crystal XRD has a composition of [Co(H2O)4Cl2] with octahedrally coordinated metal center (see Fig. 20). Four water molecules are associated with the cobalt metal center, along with two N-group of the L0 linker. The metallation is accompanied by an anti- to synconformation switch to the pyrazine rings of the L0 linker, allowing chelating of the cobalt ion. Moving forward, the structure of the blue crystal was elucidated using single crystal XRD, showing the expected octahedral (pink) to tetrahedral (blue) coordination change of the cobalt center. The new coordination sphere has two Cl group coordinated with the meal center along with two AN group of the L0 linker. The transformation happens by means S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 Table 1 Summary of acs nets with respect to the observed swelling amplitude and BET surface area. a Net %Swelling amplitude BET (m2/g) Refs. Sc-bdc Sc-dobdc (dobdc = 2,5-dihydroxyterephatate) MOF-235 MOF-236 PCN-19 MIL-88-B(4H) MIL-88-B(Cl) MIL-88-B(Br) MIL-88-B(NH2) MIL-88-B(NO2) MIL-88-B(CH3) MIL-88-B(2CH3) MIL-88-B(2OH) MIL-88-B(4F) MIL-88-B(2CF3) MIL-88-B(4CH3) MIL-88-D(4CH3) MIL-88-D(2CH3) MIL-88-D(4H) MCF-18(L,M) tp-PMBB-5-acs-1 N/A N/A 0 0 N/A 136 117 94 132 63 97 97 92 54 49 25 105 228 237 75–125 0 634 0 0 0 723 8 19 <1 14 15 80 60 3 35 330 1220 210 29 N/A 224a 710 [174] [175] [170] [170] [176] [169] [169] [169] [169] [169] [169] [169] [169] [169] [169] [169] [169] [169] [169] [168] [173] Langmuir S.A. of Cl migration from the pores to the metal center. The conversion also leads to the rotation of the pyrazole moiety of L0 , exhibiting the inherent flexibility of the MOF backbone. The pink crystals turn blue upon exposure to air within several minutes, converting from tetrahedral to octahedral coordination by adsorbing moisture from atmosphere. The exceptional crystallinity of the MOF-metallated adduct species was used to characterize the final reaction product from the oxidative addition of CH3I to MOF-[Rh(CO)2][RhCl2(CO)2] species to form and subsequent CO insertion into the RhACH3 bond. The key intermediate of the reaction was characterized by the single crystal XRD and DFT calculation. The material reported 145 by Sumby and co-workers indeed represents a rare example of a flexible MOF with excellent stability and intriguing properties. 5.5. Crystal size Particle size has shown recently a great impact on the properties of certain MOF structures [178], however, the particle size effect on the flexibility of MOFs has not yet been fully understood. A recent research by Kitagawa and co-worker demonstrated that the crystal downsizing of a twofold interpenetrated flexible MOF led to the development of third metastable phase exhibiting a ‘‘shape-memory” effect not seen in the larger sized crystals [151]. This interpenetrated MOF of formula [Cu2(dicarboxylate)2(amine)]n can undergo structure change from nonporous closed phase to guest-induced open phase. The downsizing of the crystals to the mesoscale restrained the structural mobility and stabilized the open dried phase without inclusion of guest molecules that can be converted to the original closed phase by thermal treatment ‘‘shape-memory effect” (see Fig. 21) [151]. The dried open phase did not show any gate-opening upon the adsorption of methanol with type I isotherm which obviously demonstrated the rigidity of the structure impacted by the crystal downsizing. Another research reported by Sumby and co-workers addressed the impact of the particle size of [Cu(bcppm)H2O]
S (H2bcppm = bis(4-(4-car boxyphenyl)-1H-pyrazolyl)methane, S = solvent) on the flexibility of its framework [179]. The structure changes from the 2D open form to 3D locked form upon activation because of the structural reorganization from five coordinate square pyramidal Cu(II) centers to six coordinate Jahn–Teller distorted octahedral centers. However, the crystals could be changed back to the open form by resolvation of the activated 3D crystals in methanol for 7 days. Interestingly, the rate of the structural reorganization of this framework to form open material upon resolvation was found to correlate with the particle size of MOF. The control over the crystal size of this MOF was performed by conducting the reaction at room temperature with the utilization of NaOH as a base and systematically varying ligand: base ratio and the reactant concentration that Fig. 20. Crystal structures of [Mn3(L)2(L0 )] and its different conformational forms upon desolvation and metalation. Reproduced with permission from Ref. [177]. Copyright 2014 Nature. 146 S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 Fig. 21. Schematic representation of the impact of crystal downsizing in the induction of shape-memory effect in the twofold interpenetrated MOF; [Cu2(bdc)2(bpy)]n. Reproduced with permission from Ref. [151]. Copyright 2013 Science. afford crystals with different sizes (85–620 nm). Reducing the crystal size results in diminishing of the resolvation time needed for the structure reorganization from 7 days for the bulk materials (larger crystal size) to 3 h for the downsized-crystal samples. The fast rate of the structural reorganization presumably attributed to the fast diffusion of the solvent (methanol) into the smaller size crystals which afford larger external surface area compared to the bulk crystals. Thus, crystal size of MOFs has been shown to dramatically impact the framework flexibility and structural reorganization. Kitagawa and co-workers [180] reported an interesting case where the flexibility towards external guest molecules such as ethanol depends on the particle size of the 2D interdigitated coordination polymer of formula [181]. In the bulk crystal state (135 nm), no significant guest adsorption and associated structural change was observed, but when the crystal is decreased down to nanometer size (16 nm) by means of thin film deposition on a solid substrate, a dynamic structural transformation as a response to ethanol vapor was observed by means of in situ surface XRD analysis. Specifically, the thin film shows an anisotropic expansion along the vertical direction to the substrate surface, completely different from that of the bulk state. Interestingly, with increase in thin film thickness, the flexibility starts to disappear. It is postulated that the ethanol penetration within the coordination polymer is limited in the bulk state because of the strong p–p stacking between pyridine linkers. However, as the surface area increases significantly at the nanoscale, the linker p–p interaction is not as effective as in the bulk because of the reduced number of linker at the surface. As a result, the potential barrier for gate-opening decreases, resulting in specific response to the guests and adsorption kinetics. The guest-induced structural transition impacted by crystal downsizing was also observed in other MOF structures such as [Cu3(btc)2]n nanocrystals on gold-coated QCM substrates and ZIF-8 [182,183]. We believe that such a crystal-downsizing effect should be investigated with other coordination polymers, and may help in the development of exquisite materials for future applications such as gas separation, chemo-switching, sensing, catalysis and other not-yet-explored applications. 5.6. Inclusion of molecular rotors An exquisite way to control the molecular dynamics of the porous materials is the introduction of molecular rotors through the insertion of mobile elements operating as rotors to the molecular structure of robust porous framework. The rotor dynamic can be switched on/off in response to specific external stimuli affording fascinating switchable magnetic, dielectric, and optoelectronic properties that can mimic certain functions in the macroscopic level. The use of MOFs is very motivating since their low density, large surface area and free accessible volume will not only provide the enough space for rotors to freely rotate but also allow regular arrangement of rotors in their periodic structures and faster dynamic by the rapid diffusion of guest molecules (chemical stimuli) in MOF pores. Most importantly, the structure tunability of MOFs in terms of pore size, pore shape and structure functionality allow the systematic modification of the local environment of the rotors which thereby impact their dynamic. As we previously mentioned, the phenyl or pyridyl rings of the organic linkers can act as rotors where the rings organize their rotational configurations to achieve a more efficient adsorbate-adsorbent interaction [71,141,143,164,184,185]. For example, Kitagawa and co-workers presented the first study of the control over the rotational frequencies of rotors in flexible Porous Coordination Polymers without thermal modulation [184]. The rotational frequencies of flexible PCPs [{Zn(5-NO2-ip) (bpy-d8)}(0.5DMF
0.5MeOH)]n (CID-5  G; S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 147 Fig. 22. Structures of (a) CID-5  G and (b) (CID-6  G). Deuterium atoms and hydrogen atoms are omitted for clarity. Schematic representation of the local environments of the pyridyl ring rotors in (c) CID-5  G and (d) CID-6  G. Green represents the rotating mode of the pyridyl ring rotors while gray represents the static mode. Illustrations of flip modes of PY1 and PY2 in (c) CID-5  G and (d) CID-6  G. Reproduced with permission from Ref. [184]. Copyright 2015 American Chemical Society. 5-NO2-ip = 5-nitroisophthalate, and bpy-d8 = deuterated 4,40 bipyridyl), and [{Zn(5-MeO-ip) (bpy-d8)}(0.5 DMF
0.5 MeOH)]n (C ID-6  G; 5-MeO-ip = 5-methoxyisophthalate)], where G denotes guest molecule (DMF and MeOH) [184], were controlled using the solid-solution approach [186]. In these materials, the pyridyl groups of the pillar linkers (bpy) act as rotors forming 2 types of rotors (PY1 and PY2) which can undergo different flip modes; a 2-site flip of the pyridyl ring rotors between 24.9° and 180° and a 4-site flip between 0°, 24.9°, 180°, and 204.9°. The rotation of the pyridyl ring rotors was investigated by 2H solid-state NMR along with the computational simulations. It was found that the local environment around the pyridyl rotors strongly impact their flip modes and rotation frequencies. As shown in Fig. 22, the steric hindrance between the PY1 and the adjacent 5-NO2-ip group in CID5  G, suggests the presence of PY1 in a static mode, while PY2 endures both 2-site and 4-site flip rotations. Conversely, 2H solid-state NMR spectra of CID6  G suggested that both PY1 and PY2 exist in static modes. To further understand the impact of the local environment around the molecular rotor on their rotation modes and rotational frequencies, the author exploited the solidsolution approach by changing the solid-solution ratio x that affords a variation of CID-5/6 isostructures with variant cell volume and rotor local environment and thereby different frequencies. The control over the dynamic of the molecular rotors in MOF is not only affected by the local environment around the rotor but is also impacted by the guest molecules which can act as a chemical stimuli. Chen and his co-workers presented the dynamics of the pyrimidine groups of the organic linkers in UTSA-76a structure 148 S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 Fig. 23. Schematic representation of the concept of MIM admission in the MOF structure. Reproduced with permission from Ref. [190]. Copyright 2015 Nature. upon the adsorption of methane gas at high pressures [187]. UTSA76 is an isostructural to NOTT-101 in which the central phenyl group of the tetra carboxylate organic linker in NOTT-101 is replaced by pyrimidine group in UTSA-76. However, the activated structure UTSA-76a showed significantly improved volumetric methane storage capacity at 65 bar and 298 K from 237 cm3 (STP) cm3 for NOTT-101a to 257 cm3 (STP) cm3 for UTSA-76a. The X-ray diffraction data revealed no alternation in the pore structure/geometry of both MOFs. The dispersion-corrected densityfunctional theory (DFT-D) calculations proved that both MOFs have the same binding sites (open Cu sites and cage window sites) and rather similar binding energies for CH4 either around the phenyl ring (NOTT-101a) or pyrimidine rings (UTSA-76a). These results motivated the authors to examine the two MOF from the structural dynamic perspective [187]. Interestingly, the DFT calculations revealed that the rotational energy barrier of the pyrimidine rings to rotate in UTSA-76 of 8.2 kJ/mol is much lower than that of the central phenyl ring in NOTT-101 (20.2 kJ/mol). Accordingly, the pyrimidine rings of the organic linkers in UTSA-76 can be rotate upon the adsorption of methane gas under high pressures in order to optimize the methane packing in the MOF pores much more than the central phenyl groups in UTSA-76 which explains the 6% difference in the methane uptake between the two MOFs. The data is experimentally validated by the neutron scattering measurements to monitor the H motion of the central rings of organic linkers in both MOFs. The data showed that the pyrimidine rings can display a rotation motion at much lower temperature than the phenyl rings revealing the higher mobility of pyrimidine rings than the phenyl rings which is in a good agreement with the DFT calculations. Briefly, the rings of organic linkers (phenyl, pyridyl, pyrimidine) can act as molecular rotors in MOF/PCP structures that exhibit fascinating properties in response of specific stimuli which can lead to the development of the next-generation of molecular machine (Fig. 23). More recently, a novel approach for the admission of molecular rotors to the MOF structures and the control of their dynamics were reported [188–191]. Mechanically interlocked molecules (MIMs) such as catenanes, and rotaxanes have received a dramatic interest especially after the pioneering work reported by Stoddard and his co-workers [192–196]. They discovered that MIMs can work as a ‘‘molecular shuttle” in which their macrocyclic rings can move backwards and forwards on a linear track between two recognition sites ended by bulky plugs. This mechanism could be very beneficial for mimicking sophisticated functions of the macroscopic switches and machines. Loeb, Schurko and their co-workers reported the integration a molecular shuttle in MOF structures [190,191,197]. The authors at first synthesized MOF-containing MIM constructed from the Cu nodes and tetracarboxylate organic linker-containing [2]rotaxane which showed an independent motion in the rigid MOF structure [188]. As a further stage, the admission of molecular shuttle in a MOF designated UWDM-4 was accomplished by the utilization of rigid organic linker containing a [24]crown-8 ether (24C8) macrocycle as a molecular shuttle moving between two benzimidazole recognition sites and immobilized by two triphenyldicarboxylic acid struts [190]. Single-crystal X-ray diffraction revealed perfect construction of the molecular shuttle in MOF, demonstrating that 24C8 macrocycle interacts with one of the benzimidazole rings by means of a hydrogen bond between a NAH of benzimidazole and an oxygen atom of the crown ether. The suttling motion in the MOF was tracked as a function of temperature using 13C SSNMR by labeling the carbon atoms at the two positions of benzimidazole rings by 50% 13C. The data showed only single resonance for both empty and occupied recognition sites at room temperature which revealed a fast motion of the molecular shuttle that cannot detect it by the NMR spectra. The rate of the shuttling motion was controlled and slowed by decreasing the temperature which leads to the appearance of two distinct resonances. The approach for the inclusion of interlocked molecules in MOFs were previously reported by Stoddard, Yaghi and co-workers and relied on the synthesis of catenanebased MOF as a porotype for the development of the nextgeneration molecular machines or switches [189]. The threedimensional MIM@MOF designated MOF-1030 was synthesized by the reaction of Cu(I) with a catenated dicarboxylic acid strut containing MIM. The large linker strut of 32.9 Å allows the integration of such bulky MIM and to be precisely anchored in certain locations through the framework. However, this example did not demonstrate the transitional dynamics of MIM in MOF but it S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 provides the ability of MOF to enhance the organization and orientation of MIM and the potential for introducing a mechanical motion in the 3D porous materials. These materials can be considered as blueprints for the future generation of solid-state molecular switches and machines. 6. Conclusion Flexibility is a unique physical phenomenon in MOFs. Extended solid-state materials such as aluminosilicate zeolites show some degree of flexibilities for various types of adsorbates, but not to the extent seen in MOF based materials. Flexible MOFs can have properties not seen in typical rigid MOFs. For example, Long and co-workers recently reported a flexible MOF with recordbreaking CH4 storage properties. The most common way to design and synthesize a flexible MOF is the use of flexible linkers. The flexibility arises in the MOF as the organic linkers flexes as a function of external stimuli such as pressure, temperature and light. However, other parameters such as types of SBU and topology have a role in determining the potential of a framework to be flexible in nature. As a result, a framework composed of flexible organic linker is not necessarily always flexible and several other conditions (including metal ions, SBU, solvent) and proper condition of external stimuli must be satisfied. Further exploratory syntheses, in-situ structural studies and gas-adsorption will likely yield further fundamental and applied results. 7. List of abbreviations AzDC H2AZDC azo-bipy BB-pc bdc H2bdc BET Gly-Ala BHE-bpb bipy bpydc BME-bdc H2BMEbdc bpdc H2bpdc bpee or bpe bpy btc H3btc CBMC CDC H2CDC CID CPs dabco DB-bdc H2DBbdc DFT dhbc H2dhbc azobenzene-4,40 -dicarboxylate azobenzene-4,40 -dicarboxylic acid 3-azo-phenyl-4,40 -bipyridine 1,4-butane bis(phosphonic acid) 1,4-benzenedicarboxylate 1,4-benzenedicarboxylic acid Brunauer–Emmett–Teller glycylalanine 2,5-bis(2-hydroxyethoxy)-1,4-bis(4-pyridyl)ben zene bipyridine 2,20 -bipyridine-5,50 -dicarboxylate 2,5-bis(2-methoxyethoxy)benzenedicarboxylate 2,5-bis(2-methoxyethoxy)benzenedicarboxylic acid biphenyl-4,40 -dicarboxylate biphenyl-4,40 -dicarboxylic acid 1,2-bis(2-pyridyl)ethylene dobdc H2dobdc dpyg DUT fa H2fa F-bdc H2F-bdc fu-bdc H2fu-bdc G Bdp GPa HKUST HMOF ip H2ip IRMOF IUPAC maba H2maba MBB MIL GCMC mIm or Im MOFs MOMs ndc H2ndc tmenH2 NIY-bc NU PBU PCN PXRD pyz pzdc H2pzdc Qst RPM SBUs tp-PMBB tpt ZAG ZIF PSM NGA H2bcppm 5-NO2-ip bpy-d8 MeO-ip MIM 149 2,5-dihydoxyterephthalate (also 2,5-dioxido-1,4benzenedicarboxylate) 2,5-dihydoxyterephthalic acid 1,2-Di(4-pyridyl)glycol Dresden University of Technology fumarate fumaric acid alkoxy functionalized 1,4-benzenedicarboxylate alkoxy functionalized 1,4-benzenedicarboxylic acid functionalized 1,4-benzenedicarboxylate functionalized 1,4-benzenedicarboxylic acid guest 1,4-benzenedipyrazolate gigapascal Hong Kong University of Science and Technology Hinged Metal–Organic Framework isophthalate isophthalic acid Isoreticular Metal–Organic Framework International Union of Pure and Applied Chemistry m-aminobenzoate m-aminobenzoic acid Molecular Building Block Materials of Institut Lavoisier Grand Canonical Monte Carlo 2-methylimidazole Metal–Organic Frameworks Metal–Organic Materials naphthalenedicarboxylate naphthalenedicarboxylic acid N,N,N0 ,N0 -tetramethylethylenediammonium 4-(1H-naphtho[2,3-d]imidazol-1-yl)benzoate Northwestern University Primary Building Unit Porous Coordination Networks powder X-ray diffraction pyrazine 2,3-pyrazinedicarboxylate 2,3-pyrazinedicarboxylic acid Isosteric Heat of Adsorption (or heats of adsorption) Rutgers Recyclable Porous Material secondary building units Trigonal Prismatic Primary Molecular Building Block 2,4,6-tris(4-pyridyl)triazine Zinc Alkyl Gate Zeolitic Imidazolate Framework post-synthetic modification negative gas adsorption bis(4-(4-carboxyphenyl)-1H-pyrazolyl)methane 5-nitroisophthalate deuterated 4,40 -bipyridyl 5-methoxyisophthalate mechanically interlocked molecules 4,4 -bipyridine 1,3,5-benzenetricarboxylate 1,3,5-benzenetricarboxylic acid configurational-bias Monte Carlo 1,4-cyclohexanedicarboxylate 1,4-cyclohexanedicarboxylic acid coordination polymer with interdigitated structures coordination Polymers 1,4-diazabicyclo[2.2.2]-octane 2,5-dibutoxybenzenedicarboxylate 2,5-dibutoxybenzenedicarboxylate Acknowledgements Density Functional Theory 2,5-dihydroxybenzoate 2,5-dihydroxybenzoic acid The authors would like to acknowledge the Department of Energy Office of Nuclear Energy DOE Basic Energy Science and DOE – EERE Office of Geothermal program. 0 150 S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 References [1] G.B. Gardner, D. Venkataraman, J.S. Moore, S. Lee, Nature 374 (1995) 792–795. [2] A.J. Blake, N.R. Champness, P. Hubberstey, W.-S. Li, M.A. Withersby, M. Schröder, Coord. Chem. Rev. 183 (1999) 117–138. [3] F. Nouar, J.F. Eubank, T. Bousquet, L. Wojtas, M.J. Zaworotko, M. Eddaoudi, J. Am. Chem. Soc. 130 (2008) 1833–1835. [4] J.J. Perry Iv, J.A. Perman, M.J. Zaworotko, Chem. Soc. Rev. 38 (2009) 1400. [5] A.J. Cairns, J.A. Perman, L. Wojtas, V.C. Kravtsov, M.H. Alkordi, M. Eddaoudi, M. J. Zaworotko, J. Am. Chem. Soc. 130 (2008) 1560–1561. [6] S.R. Batten, N.R. Champness, X.-M. Chen, J. Garcia-Martinez, S. Kitagawa, L. Öhrström, M. O’Keeffe, M. Paik Suh, J. Reedijk, Pure Appl. Chem. 85 (2013). [7] S. Kitagawa, S.-I. Noro, T. Nakamura, Chem. Commun. (2006) 701. [8] S. Kitagawa, R. Kitaura, S.-I. Noro, Angew. Chem. Int. Ed. 43 (2004) 2334– 2375. [9] H. Abourahma, G.J. McManus, B. Moulton, R.D. Bailey Walsh, M.J. Zaworotko, Macromol. Symp. 196 (2003) 213–227. [10] L. Huang, H. Wang, J. Chen, Z. Wang, J. Sun, D. Zhao, Y. Yan, Microporous Mesoporous Mater. 58 (2003) 105–114. [11] C. Janiak, Dalton Trans. (2003) 2781–2804. [12] L.R. MacGillivray, In: Metal-Organic Frameworks: Design and Application, John Wiley & Sons, Hoboken, 2010. [13] S.S.Y. Chui, S.M.F. Lo, J.P.H. Charmant, A.G. Orpen, I.D. Williams, Science 283 (1999) 1148–1150. [14] H. Li, M. Eddaoudi, M. O’Keeffe, O.M. Yaghi, Nature 402 (1999) 276–279. [15] S.R. Batten, S.M. Neville, D.R. Turner, Coordination Polymers: Design, Analysis and Application, Royal Society of Chemistry, Cambridge, 2009. [16] M. Eddaoudi, J. Kim, N. Rosi, D. Vodak, J. Wachter, M. O’Keeffe, O.M. Yaghi, Science 295 (2002) 469–472. [17] S. Surblé, C. Serre, C. Mellot-Draznieks, F. Millange, G. Férey, Chem. Commun. (2006) 284. [18] S.J. Garibay, S.M. Cohen, Chem. Commun. 46 (2010) 7700–7702. [19] O.K. Farha, C.E. Wilmer, I. Eryazici, B.G. Hauser, P.A. Parilla, K. O’Neill, A.A. Sarjeant, S.T. Nguyen, R.Q. Snurr, J.T. Hupp, J. Am. Chem. Soc. 134 (2012) 9860–9863. [20] K. Koh, J.D. Van Oosterhout, S. Roy, A.G. Wong-Foy, A.J. Matzger, Chem. Sci. 3 (2012) 2429–2432. [21] K. Koh, A.G. Wong-Foy, A.J. Matzger, J. Am. Chem. Soc. 131 (2009) 4184–4185. [22] O.K. Farha, A. Özgür Yazaydın, I. Eryazici, C.D. Malliakas, B.G. Hauser, M.G. Kanatzidis, S.T. Nguyen, R.Q. Snurr, J.T. Hupp, Nat. Chem. 2 (2010) 944–948. [23] O.K. Farha, I. Eryazici, N.C. Jeong, B.G. Hauser, C.E. Wilmer, A.A. Sarjeant, R.Q. Snurr, S.T. Nguyen, A.O. Yazaydin, J.T. Hupp, J. Am. Chem. Soc. 134 (2012) 15016–15021. [24] T.R. Zeitler, M.D. Allendorf, J.A. Greathouse, J. Phys. Chem. C 116 (2012) 3492– 3502. [25] M. Kawano, T. Kawamichi, T. Haneda, T. Kojima, M. Fujita, J. Am. Chem. Soc. 129 (2007) 15418–15419. [26] Q.-G. Zhai, Q. Lin, T. Wu, L. Wang, S.-T. Zheng, X. Bu, P. Feng, Chem. Mater. 24 (2012) 2624–2626. [27] Q. Lin, T. Wu, S.-T. Zheng, X. Bu, P. Feng, J. Am. Chem. Soc. 134 (2012) 784– 787. [28] R. Banerjee, H. Furukawa, D. Britt, C. Knobler, M. O’Keeffe, O.M. Yaghi, J. Am. Chem. Soc. 131 (2009) 3875–3877. [29] J.-R. Li, R.J. Kuppler, H.-C. Zhou, Chem. Soc. Rev. 38 (2009) 1477. [30] M.P. Suh, H.J. Park, T.K. Prasad, D.-W. Lim, Chem. Rev. 112 (2012) 782–835. [31] K. Sumida, D.L. Rogow, J.A. Mason, T.M. McDonald, E.D. Bloch, Z.R. Herm, T.-H. Bae, J.R. Long, Chem. Rev. 112 (2012) 724–781. [32] Z. Zhang, Y. Zhao, Q. Gong, Z. Li, J. Li, Chem. Commun. 49 (2013) 653–661. [33] S. Ma, H.-C. Zhou, Chem. Commun. 46 (2010) 44–53. [34] Z. Zhang, W.-Y. Gao, L. Wojtas, S. Ma, M. Eddaoudi, M.J. Zaworotko, Angew. Chem. Int. Ed. 51 (2012) 9330–9334. [35] J. Lee, O.K. Farha, J. Roberts, K.A. Scheidt, S.T. Nguyen, J.T. Hupp, Chem. Soc. Rev. 38 (2009) 1450. [36] J. Zhang, S. Chen, T. Wu, P. Feng, X. Bu, J. Am. Chem. Soc. 130 (2008) 12882– 12883. [37] K. Lillerud, U. Olsbye, M. Tilset, Top. Catal. 53 (2010) 859–868. [38] F.X. Llabrés i Xamena, A. Abad, A. Corma, H. Garcia, J. Catal. 250 (2007) 294– 298. [39] P. Horcajada, R. Gref, T. Baati, P.K. Allan, G. Maurin, P. Couvreur, G. Férey, R.E. Morris, C. Serre, Chem. Rev. 112 (2012) 1232–1268. [40] A.C. McKinlay, R.E. Morris, P. Horcajada, G. Férey, R. Gref, P. Couvreur, C. Serre, Angew. Chem. Int. Ed. 49 (2010) 6260–6266. [41] G. Givaja, P. Amo-Ochoa, C.J. Gómez-García, F. Zamora, Chem. Soc. Rev. 41 (2012) 115. [42] G. Xu, K. Otsubo, T. Yamada, S. Sakaida, H. Kitagawa, J. Am. Chem. Soc. 135 (2013) 7438–7441. [43] G.R. Desiraju, Crystal Engineering. The Design of Organic Solids, Elsevier, Amsterdam, 1989. [44] H. Li, M. Eddaoudi, T.L. Groy, O.M. Yaghi, J. Am. Chem. Soc. 120 (1998) 8571– 8572. [45] M. Kondo, T. Yoshitomi, H. Matsuzaka, S. Kitagawa, K. Seki, Angew. Chem., Int. Ed. Engl. 36 (1997) 1725–1727. [46] G.R. Desiraju, J. Am. Chem. Soc. 135 (2013) 9952–9967. [47] G.R. Desiraju, J.J. Vittal, A. Ramanan, Crystal Engineering, World Scientific, 2011. [48] G.R. Desiraju, Angew. Chem. Int. Ed. 46 (2007) 8342–8356. [49] K. Uemura, R. Matsuda, S. Kitagawa, J. Solid State Chem. 178 (2005) 2420– 2429. [50] S. Bureekaew, S. Shimomura, S. Kitagawa, Sci. Technol. Adv. Mater. 9 (2008) 014108. [51] A. Schneemann, V. Bon, I. Schwedler, I. Senkovska, S. Kaskel, R.A. Fischer, Chem. Soc. Rev. 43 (2014) 6062–6096. [52] P. Horcajada, C. Serre, G. Maurin, N.A. Ramsahye, F. Balas, M.A. Vallet-Regí, M. Sebban, F. Taulelle, G.R. Férey, J. Am. Chem. Soc. 130 (2008) 6774–6780. [53] S. Horike, S. Shimomura, S. Kitagawa, Nat. Chem. 1 (2009) 695–704. [54] G. Ferey, Nat. Mater. 2 (2003) 136–137. [55] D. Maspoch, D. Ruiz-Molina, K. Wurst, N. Domingo, M. Cavallini, F. Biscarini, J. Tejada, C. Rovira, J. Veciana, Nat. Mater. 2 (2003) 190–195. [56] K. Biradha, M. Fujita, Angew. Chem. Int. Ed. 41 (2002) 3392–3395. [57] K. Uemura, S. Kitagawa, M. Kondo, K. Fukui, R. Kitaura, H.-C. Chang, T. Mizutani, Chem. – A Eur. J. 8 (2002) 3586–3600. [58] K. Barthelet, J. Marrot, D. Riou, G. Férey, Angew. Chem. Int. Ed. 41 (2002) 281– 284. [59] C. Serre, F. Millange, C. Thouvenot, M. Noguès, G. Marsolier, D. Louër, G. Férey, J. Am. Chem. Soc. 124 (2002) 13519–13526. [60] T. Devic, P. Horcajada, C. Serre, F. Salles, G. Maurin, B. Moulin, D. Heurtaux, G. Clet, A. Vimont, J.-M. Grenèche, B.L. Ouay, F. Moreau, E. Magnier, Y. Filinchuk, J. Marrot, J.-C. Lavalley, M. Daturi, G. Férey, J. Am. Chem. Soc. 132 (2010) 1127–1136. [61] G. Férey, C. Serre, Chem. Soc. Rev. 38 (2009) 1380. [62] S. Kitagawa, K. Uemura, Chem. Soc. Rev. 34 (2005) 109. [63] G. Férey, Dalton Trans. (2009) 4400. [64] J. Seo, C. Bonneau, R. Matsuda, M. Takata, S. Kitagawa, J. Am. Chem. Soc. 133 (2011) 9005–9013. [65] M. Witman, S. Ling, S. Jawahery, P.G. Boyd, M. Haranczyk, B. Slater, B. Smit, J. Am. Chem. Soc. 139 (2017) 5547–5557. [66] T.D. Bennett, A.H. Fuchs, A.K. Cheetham, F.-X. Coudert, Dalton Trans. 45 (2016) 4058–4059. [67] C.M. Simon, E. Braun, C. Carraro, B. Smit, Proc. Natl. Acad. Sci. 114 (2017) E287–E296. [68] S. Watanabe, H. Sugiyama, H. Adachi, H. Tanaka, M.T. Miyahara, J. Chem. Phys. 130 (2009) 164707. [69] R. Numaguchi, H. Tanaka, S. Watanabe, M.T. Miyahara, J. Chem. Phys. 138 (2013) 054708. [70] C. Mellot-Draznieks, C. Serre, S. Surblé, N. Audebrand, G. Férey, J. Am. Chem. Soc. 127 (2005) 16273–16278. [71] D.N. Dybtsev, H. Chun, K. Kim, Angew. Chem. Int. Ed. 43 (2004) 5033–5036. [72] F. Millange, C. Serre, G. Férey, Chem. Commun. (2002) 822–823. [73] T. Loiseau, C. Serre, C. Huguenard, G. Fink, F. Taulelle, M. Henry, T. Bataille, G. Férey, Chem. – A Eur. J. 10 (2004) 1373–1382. [74] F. Millange, N. Guillou, R.I. Walton, J.-M. Grenèche, I. Margiolaki, G. Férey, Chem. Commun. (2008) 4732. [75] J.P.S. Mowat, V.R. Seymour, J.M. Griffin, S.P. Thompson, A.M.Z. Slawin, D. Fairen-Jimenez, T. Düren, S.E. Ashbrook, P.A. Wright, Dalton Trans. 41 (2012) 3937–3941. [76] E.V. Anokhina, M. Vougo-Zanda, X. Wang, A.J. Jacobson, J. Am. Chem. Soc. 127 (2005) 15000–15001. [77] C. Volkringer, T. Loiseau, N. Guillou, G. Férey, E. Elkaïm, A. Vimont, Dalton Trans. (2009) 2241. [78] H. Wang, J. Getzschmann, I. Senkovska, S. Kaskel, Microporous Mesoporous Mater. 116 (2008) 653–657. [79] K. Seki, W. Mori, J. Phys. Chem. B 106 (2002) 1380–1385. [80] N. Klein, C. Herzog, M. Sabo, I. Senkovska, J. Getzschmann, S. Paasch, M.R. Lohe, E. Brunner, S. Kaskel, PCCP 12 (2010) 11778. [81] N. Klein, H.C. Hoffmann, A. Cadiau, J. Getzschmann, M.R. Lohe, S. Paasch, T. Heydenreich, K. Adil, I. Senkovska, E. Brunner, S. Kaskel, J. Mater. Chem. 22 (2012) 10303. [82] C.R. Murdock, B.C. Hughes, Z. Lu, D.M. Jenkins, Coord. Chem. Rev. 258–259 (2014) 119–136. [83] D. Fairen-Jimenez, S.A. Moggach, M.T. Wharmby, P.A. Wright, S. Parsons, T. Düren, J. Am. Chem. Soc. 133 (2011) 8900–8902. [84] D. Fairen-Jimenez, R. Galvelis, A. Torrisi, A.D. Gellan, M.T. Wharmby, P.A. Wright, C. Mellot-Draznieks, T. Düren, Dalton Trans. 41 (2012) 10752. [85] G. Garberoglio, S. Taioli, Microporous Mesoporous Mater. 163 (2012) 215– 220. [86] J. Seo, R. Matsuda, H. Sakamoto, C. Bonneau, S. Kitagawa, J. Am. Chem. Soc. 131 (2009) 12792–12800. [87] S. Henke, A. Schneemann, R.A. Fischer, Adv. Funct. Mater. 23 (2013) 5990– 5996. [88] J. Rabone, Y.-F. Yue, S.Y. Chong, K.C. Stylianou, J. Bacsa, D. Bradshaw, G.R. Darling, N.G. Berry, Y.Z. Khimyak, A.Y. Ganin, P. Wiper, J.B. Claridge, M.J. Rosseinsky, Science 329 (2010) 1053–1057. [89] K. Barthelet, J. Marrot, G. Ferey, D. Riou, Chem. Commun. (2004) 520. [90] G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surblé, I. Margiolaki, Science 309 (2005) 2040–2042. [91] S. Biswas, T. Rémy, S. Couck, D. Denysenko, G. Rampelberg, J.F.M. Denayer, D. Volkmer, C. Detavernier, P. Van Der Voort, PCCP 15 (2013) 3552. [92] R. Kitaura, K. Fujimoto, S.-I. Noro, M. Kondo, S. Kitagawa, Angew. Chem. Int. Ed. 41 (2002) 133–135. [93] L. Carlucci, G. Ciani, M. Moret, D.M. Proserpio, S. Rizzato, Angew. Chem. Int. Ed. 39 (2000) 1506–1510. S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 [94] R. Kitaura, K. Seki, G. Akiyama, S. Kitagawa, Angew. Chem. Int. Ed. 42 (2003) 428–431. [95] E.C. Spencer, M.S.R.N. Kiran, W. Li, U. Ramamurty, N.L. Ross, A.K. Cheetham, Angew. Chem.-Int. Ed. 53 (2014) 5583–5586. [96] M.T. Wharmby, S. Henke, T.D. Bennett, S.R. Bajpe, I. Schwedler, S.P. Thompson, F. Gozzo, P. Simoncic, C. Mellot-Draznieks, H.Z. Tao, Y.Z. Yue, A. K. Cheetham, Angew. Chem.-Int. Ed. 54 (2015) 6447–6451. [97] A.U. Ortiz, A. Boutin, K.J. Gagnon, A. Clearfield, F.X. Coudert, J. Am. Chem. Soc. 136 (2014) 11540–11545. [98] F.X. Coudert, Chem. Mater. 27 (2015) 1905–1916. [99] A. Clearfield, Dalton Trans. 45 (2016) 4100–4112. [100] K.J. Gagnon, C.M. Beavers, A. Clearfield, J. Am. Chem. Soc. 135 (2013) 1252– 1255. [101] S. Cao, T.D. Bennett, D.A. Keen, A.L. Goodwin, A.K. Cheetham, Chem. Commun. 48 (2012) 7805–7807. [102] S.A. Moggach, T.D. Bennett, A.K. Cheetham, Angew. Chem. Int. Ed. 48 (2009) 7087–7089. [103] P.G. Yot, Q. Ma, J. Haines, Q. Yang, A. Ghoufi, T. Devic, C. Serre, V. Dmitriev, G. Férey, C. Zhong, G. Maurin, Chem. Sci. 3 (2012) 1100. [104] J.M. Ogborn, I.E. Collings, S.A. Moggach, A.L. Thompson, A.L. Goodwin, Chem. Sci. 3 (2012) 3011. [105] C. Yang, X. Wang, M.A. Omary, Angew. Chem. Int. Ed. 48 (2009) 2500–2505. [106] S. Henke, A. Schneemann, A. Wütscher, R.A. Fischer, J. Am. Chem. Soc. 134 (2012) 9464–9474. [107] L.D. DeVries, P.M. Barron, E.P. Hurley, C. Hu, W. Choe, J. Am. Chem. Soc. 133 (2011) 14848–14851. [108] I. Grobler, V.J. Smith, P.M. Bhatt, S.A. Herbert, L.J. Barbour, J. Am. Chem. Soc. 135 (2013) 6411–6414. [109] A. Modrow, D. Zargarani, R. Herges, N. Stock, Dalton Trans. 40 (2011) 4217. [110] J. Park, D. Yuan, K.T. Pham, J.-R. Li, A. Yakovenko, H.-C. Zhou, J. Am. Chem. Soc. 134 (2012) 99–102. [111] J.W. Brown, B.L. Henderson, M.D. Kiesz, A.C. Whalley, W. Morris, S. Grunder, H. Deng, H. Furukawa, J.I. Zink, J.F. Stoddart, O.M. Yaghi, Chem. Sci. 4 (2013) 2858. [112] R. Lyndon, K. Konstas, B.P. Ladewig, P.D. Southon, P.C.J. Kepert, M.R. Hill, Angew. Chem. Int. Ed. 52 (2013) 3695–3698. [113] N. Yanai, T. Uemura, M. Inoue, R. Matsuda, T. Fukushima, M. Tsujimoto, S. Isoda, S. Kitagawa, J. Am. Chem. Soc. 134 (2012) 4501–4504. [114] S. Surblé, C. Serre, C. Mellot-Draznieks, F. Millange, G. Férey, Chem. Commun. (2006) 284–286. [115] S. Bourrelly, P.L. Llewellyn, C. Serre, F. Millange, T. Loiseau, G. Férey, J. Am. Chem. Soc. 127 (2005) 13519–13521. [116] S. Shimomura, M. Higuchi, R. Matsuda, K. Yoneda, Y. Hijikata, Y. Kubota, Y. Mita, J. Kim, M. Takata, S. Kitagawa, Nat. Chem. 2 (2010) 633–637. [117] X. Zhao, B. Xiao, A.J. Fletcher, K.M. Thomas, D. Bradshaw, M.J. Rosseinsky, Science 306 (2004) 1012–1015. [118] F.-X. Coudert, C. Mellot-Draznieks, A.H. Fuchs, A. Boutin, J. Am. Chem. Soc. 131 (2009) 11329–11331. [119] S. Aguado, G. Bergeret, M.P. Titus, V. Moizan, C. Nieto-Draghi, N. Bats, D. Farrusseng, New J. Chem. 35 (2011) 546–550. [120] H. Tanaka, S. Ohsaki, S. Hiraide, D. Yamamoto, S. Watanabe, M.T. Miyahara, J. Phys. Chem. C 118 (2014) 8445–8454. [121] P. Zhao, G.I. Lampronti, G.O. Lloyd, E. Suard, S.A.T. Redfern, J. Mater. Chem. A 2 (2014) 620–623. [122] C. Gücüyener, J. van den Bergh, J. Gascon, F. Kapteijn, J. Am. Chem. Soc. 132 (2010) 17704–17706. [123] B. Zheng, Y. Pan, Z. Lai, K.-W. Huang, Langmuir 29 (2013) 8865–8872. [124] B. Russell, J. Villaroel, K. Sapag, A.D. Migone, J. Phys. Chem. C 118 (2014) 28603–28608. [125] D.-L. Chen, N. Wang, F.-F. Wang, J. Xie, Y. Zhong, W. Zhu, J.K. Johnson, R. Krishna, J. Phys. Chem. C 118 (2014) 17831–17837. [126] H. Bux, C. Chmelik, R. Krishna, J. Caro, J. Membr. Sci. 369 (2011) 284–289. [127] C. Chmelik, D. Freude, H. Bux, J. Haase, Microporous Mesoporous Mater. 147 (2012) 135–141. [128] A. Lan, K. Li, H. Wu, L. Kong, N. Nijem, D.H. Olson, T.J. Emge, Y.J. Chabal, D.C. Langreth, M. Hong, J. Li, Inorg. Chem. 48 (2009) 7165–7173. [129] N. Nijem, H. Wu, P. Canepa, A. Marti, K.J. Balkus, T. Thonhauser, J. Li, Y.J. Chabal, J. Am. Chem. Soc. 134 (2012) 15201–15204. [130] H. Wu, R.S. Reali, D.A. Smith, M.C. Trachtenberg, J. Li, Chem. – A Eur. J. 16 (2010) 13951–13954. [131] A.M. Marti, N. Nijem, Y.J. Chabal, K.J. Balkus Jr, Microporous Mesoporous Mater. 174 (2013) 100–107. [132] P.K. Thallapally, J. Tian, M. Radha Kishan, C.A. Fernandez, S.J. Dalgarno, P.B. McGrail, J.E. Warren, J.L. Atwood, J. Am. Chem. Soc. 130 (2008) 16842–16843. [133] P.K. Thallapally, B. Peter McGrail, S.J. Dalgarno, H.T. Schaef, J. Tian, J.L. Atwood, Nat. Mater. 7 (2008) 146–150. [134] M. Radha Kishan, J. Tian, P.K. Thallapally, C.A. Fernandez, S.J. Dalgarno, J.E. Warren, B.P. McGrail, J.L. Atwood, Chem. Commun. 46 (2010) 538–540. [135] C.A. Fernandez, P.K. Thallapally, B.P. McGrail, ChemPhysChem 13 (2012) 3275–3281. [136] C.A. Fernandez, P.K. Thallapally, R.K. Motkuri, S.K. Nune, J.C. Sumrak, J. Tian, J. Liu, Cryst. Growth Des. 10 (2010) 1037–1039. [137] R.K. Motkuri, P.K. Thallapally, H.V.R. Annapureddy, L.X. Dang, R. Krishna, S.K. Nune, C.A. Fernandez, J. Liu, B.P. McGrail, Chem. Commun. 51 (2015) 8421– 8424. 151 [138] R.K. Motkuri, P.K. Thallapally, S.K. Nune, C.A. Fernandez, B.P. McGrail, J.L. Atwood, Chem. Commun. 47 (2011) 7077–7079. [139] D. Banerjee, H. Wang, A.M. Plonka, T.J. Emge, J.B. Parise, J. Li, Chem. – A Eur. J. 22 (2016) 11816–11825. [140] D.I. Kolokolov, H. Jobic, A.G. Stepanov, V. Guillerm, T. Devic, C. Serre, G. Férey, Angew. Chem. Int. Ed. 49 (2010) 4791–4794. [141] W. Morris, R.E. Taylor, C. Dybowski, O.M. Yaghi, M.A. Garcia-Garibay, J. Mol. Struct. 1004 (2011) 94–101. [142] A.E. Khudozhitkov, D.I. Kolokolov, A.G. Stepanov, V.A. Bolotov, D.N. Dybtsev, J. Phys. Chem. C 119 (2015) 28038–28045. [143] S.L. Gould, D. Tranchemontagne, O.M. Yaghi, M.A. Garcia-Garibay, J. Am. Chem. Soc. 130 (2008) 3246–3247. [144] Y.-J. Kim, D.-W. Park, J. Nanosci. Nanotechnol. 13 (2013) 2307–2312. [145] Y. Yue, J.A. Rabone, H. Liu, S.M. Mahurin, M.-R. Li, H. Wang, Z. Lu, B. Chen, J. Wang, Y. Fang, S. Dai, J. Phys. Chem. C 119 (2015) 9442–9449. [146] S. Castellanos, F. Kapteijn, J. Gascon, CrystEngComm 18 (2016) 4006–4012. [147] C. Wang, L.J. Li, J.G. Bell, X.X. Lv, S.F. Tang, X.B. Zhao, K.M. Thomas, Chem. Mater. 27 (2015) 1502–1516. [148] J.A. Mason, J. Oktawiec, M.K. Taylor, M.R. Hudson, J. Rodriguez, J.E. Bachman, M.I. Gonzalez, A. Cervellino, A. Guagliardi, C.M. Brown, P.L. Llewellyn, N. Masciocchi, J.R. Long, Nature 527 (2015), 357-+. [149] M.K. Taylor, T. Runčevski, J. Oktawiec, M.I. Gonzalez, R.L. Siegelman, J.A. Mason, J. Ye, C.M. Brown, J.R. Long, J. Am. Chem. Soc. 138 (2016) 15019– 15026. [150] A.J. Fletcher, K.M. Thomas, M.J. Rosseinsky, J. Solid State Chem. 178 (2005) 2491–2510. [151] Y. Sakata, S. Furukawa, M. Kondo, K. Hirai, N. Horike, Y. Takashima, H. Uehara, N. Louvain, M. Meilikhov, T. Tsuruoka, S. Isoda, W. Kosaka, O. Sakata, S. Kitagawa, Science 339 (2013) 193–196. [152] S. Horike, D. Tanaka, K. Nakagawa, S. Kitagawa, Chem. Commun. (2007) 3395–3397. [153] T.K. Maji, R. Matsuda, S. Kitagawa, Nat. Mater. 6 (2007) 142–148. [154] K. Uemura, Y. Yamasaki, Y. Komagawa, K. Tanaka, H. Kita, Angew. Chem. Int. Ed. 46 (2007) 6662–6665. [155] A.J. Fletcher, E.J. Cussen, T.J. Prior, M.J. Rosseinsky, C.J. Kepert, K.M. Thomas, J. Am. Chem. Soc. 123 (2001) 10001–10011. [156] E.J. Cussen, J.B. Claridge, M.J. Rosseinsky, C.J. Kepert, J. Am. Chem. Soc. 124 (2002) 9574–9581. [157] R. Kitaura, K. Fujimoto, S.-I. Noro, M. Kondo, S. Kitagawa, Angew. Chem. 114 (2002) 141–143. [158] K.C. Stylianou, J. Rabone, S.Y. Chong, R. Heck, J. Armstrong, P.V. Wiper, K.E. Jelfs, S. Zlatogorsky, J. Bacsa, A.G. McLennan, C.P. Ireland, Y.Z. Khimyak, K.M. Thomas, D. Bradshaw, M.J. Rosseinsky, J. Am. Chem. Soc. 134 (2012) 20466– 20478. [159] H. Sato, W. Kosaka, R. Matsuda, A. Hori, Y. Hijikata, R.V. Belosludov, S. Sakaki, M. Takata, S. Kitagawa, Science 343 (2014) 167–170. [160] V. Bon, I. Senkovska, D. Wallacher, D.M. Többens, I. Zizak, R. Feyerherm, U. Mueller, S. Kaskel, Inorg. Chem. 53 (2014) 1513–1520. [161] S. Krause, V. Bon, I. Senkovska, U. Stoeck, D. Wallacher, D.M. Többens, S. Zander, R.S. Pillai, G. Maurin, F.-X. Coudert, S. Kaskel, Nature 532 (2016) 348–352. [162] U. Stoeck, S. Krause, V. Bon, I. Senkovska, S. Kaskel, Chem. Commun. 48 (2012) 10841–10843. [163] J.D. Evans, L. Bocquet, F.-X. Coudert, Chem 1 (2016) 873–886. [164] S.K. Elsaidi, M.H. Mohamed, C.M. Simon, E. Braun, T. Pham, K.A. Forrest, W. Xu, D. Banerjee, B. Space, M.J. Zaworotko, P.K. Thallapally, Chem. Sci. 8 (2017) 2373–2380. [165] G. Férey, C. Serre, C. Mellot-Draznieks, F. Millange, S. Surblé, J. Dutour, I. Margiolaki, Angew. Chem. Int. Ed. 43 (2004) 6296–6301. [166] C. Serre, F. Millange, S. Surblé, G. Férey, Angew. Chem. Int. Ed. 43 (2004) 6285–6289. [167] C. Serre, C. Mellot-Draznieks, S. Surblé, N. Audebrand, Y. Filinchuk, G. Férey, Science 315 (2007) 1828–1831. [168] Y.-S. Wei, K.-J. Chen, P.-Q. Liao, B.-Y. Zhu, R.-B. Lin, H.-L. Zhou, B.-Y. Wang, W. Xue, J.-P. Zhang, X.-M. Chen, Chem. Sci. (2013). [169] P. Horcajada, F. Salles, S. Wuttke, T. Devic, D. Heurtaux, G. Maurin, A. Vimont, M. Daturi, O. David, E. Magnier, N. Stock, Y. Filinchuk, D. Popov, C. Riekel, G. Férey, C. Serre, J. Am. Chem. Soc. 133 (2011) 17839–17847. [170] A.C. Sudik, A.P. Côté, O.M. Yaghi, Inorg. Chem. 44 (2005) 2998–3000. [171] V. Bon, N. Kavoosi, I. Senkovska, P. Müller, J. Schaber, D. Wallacher, D.M. Többens, U. Mueller, S. Kaskel, Dalton Trans. 45 (2016) 4407–4415. [172] T. Loiseau, C. Mellot-Draznieks, H. Muguerra, G. Férey, M. Haouas, F. Taulelle, C. R. Chim. 8 (2005) 765–772. [173] S.K. Elsaidi, M.H. Mohamed, L. Wojtas, A.J. Cairns, M. Eddaoudi, M.J. Zaworotko, Chem. Commun. 49 (2013) 8154–8156. [174] I.A. Ibarra, X. Lin, S. Yang, A.J. Blake, G.S. Walker, S.A. Barnett, D.R. Allan, N.R. Champness, P. Hubberstey, M. Schröder, Chem. – A Eur. J. 16 (2010) 13671– 13679. [175] P.D.C. Dietzel, R. Blom, H. Fjellvåg, Dalton Trans. (2006) 2055. [176] S. Ma, J.M. Simmons, D. Yuan, J.-R. Li, W. Weng, D.-J. Liu, H.-C. Zhou, Chem. Commun. (2009) 4049–4051. [177] W.M. Bloch, A. Burgun, C.J. Coghlan, R. Lee, M.L. Coote, C.J. Doonan, C.J. Sumby, Nat. Chem. 6 (2014) 906–912. [178] N. Stock, S. Biswas, Chem. Rev. 112 (2012) 933–969. [179] O.M. Linder-Patton, W.M. Bloch, C.J. Coghlan, K. Sumida, S. Kitagawa, S. Furukawa, C.J. Doonan, C.J. Sumby, CrystEngComm 18 (2016) 4172–4179. 152 S.K. Elsaidi et al. / Coordination Chemistry Reviews 358 (2018) 125–152 [180] S. Sakaida, K. Otsubo, O. Sakata, C. Song, A. Fujiwara, M. Takata, H. Kitagawa, Nat. Chem. 8 (2016) 377–383. [181] S. Sakaida, K. Otsubo, O. Sakata, C. Song, A. Fujiwara, M. Takata, H. Kitagawa, Nat. Chem. (2016), advance online publication. [182] H. Uehara, S. Diring, S. Furukawa, Z. Kalay, M. Tsotsalas, M. Nakahama, K. Hirai, M. Kondo, O. Sakata, S. Kitagawa, J. Am. Chem. Soc. 133 (2011) 11932– 11935. [183] C. Zhang, J.A. Gee, D.S. Sholl, R.P. Lively, J. Phys. Chem. C 118 (2014) 20727– 20733. [184] M. Inukai, T. Fukushima, Y. Hijikata, N. Ogiwara, S. Horike, S. Kitagawa, J. Am. Chem. Soc. 137 (2015) 12183–12186. [185] W. Huang, J. Jiang, D. Wu, J. Xu, B. Xue, A.M. Kirillov, Inorg. Chem. 54 (2015) 10524–10526. [186] H. Deng, C.J. Doonan, H. Furukawa, R.B. Ferreira, J. Towne, C.B. Knobler, B. Wang, O.M. Yaghi, Science 327 (2010) 846–850. [187] B. Li, H.-M. Wen, H. Wang, H. Wu, M. Tyagi, T. Yildirim, W. Zhou, B. Chen, J. Am. Chem. Soc. 136 (2014) 6207–6210. [188] V.N. Vukotic, K.J. Harris, K. Zhu, R.W. Schurko, S.J. Loeb, Nat. Chem. 4 (2012) 456–460. [189] Q. Li, C.-H. Sue, S. Basu, A.K. Shveyd, W. Zhang, G. Barin, L. Fang, A.A. Sarjeant, J.F. Stoddart, O.M. Yaghi, Angew. Chem. Int. Ed. 49 (2010) 6751–6755. [190] K. Zhu, C.A. O’Keefe, V.N. Vukotic, R.W. Schurko, S.J. Loeb, Nat. Chem. 7 (2015) 514–519. [191] V.N. Vukotic, C.A. O’Keefe, K. Zhu, K.J. Harris, C. To, R.W. Schurko, S.J. Loeb, J. Am. Chem. Soc. 137 (2015) 9643–9651. [192] J.E.M. Lewis, M. Galli, S.M. Goldup, Chem. Commun. 53 (2017) 298–312. [193] P.L. Anelli, N. Spencer, J.F. Stoddart, J. Am. Chem. Soc. 113 (1991) 5131–5133. [194] R.A. Bissell, E. Cordova, A.E. Kaifer, J.F. Stoddart, Nature 369 (1994) 133–137. [195] A. Coskun, M. Banaszak, R.D. Astumian, J.F. Stoddart, B.A. Grzybowski, Chem. Soc. Rev. 41 (2012) 19–30. [196] H. Deng, M.A. Olson, J.F. Stoddart, O.M. Yaghi, Nat. Chem. 2 (2010) 439–443. [197] M.A. Olson, Nat. Chem. 7 (2015) 470–471.