Coordination change, lability and hemilability in metal–organic frameworks

Russell E. Morris *a and Lee Brammer *b
aEaStCHEM School of Chemistry, University of St Andrews, Purdie Building, St Andrews, KY16 9ST, UK. E-mail:
bDepartment of Chemistry, University of Sheffield, Brook Hill, Sheffield S3 7HF, UK. E-mail:

Received 13th March 2017

First published on 10th August 2017

Metal–organic frameworks (MOFs) are some of the most exciting materials in current science. Their utility and diversity of applications depends on a combination of their chemistry, their framework topology and the spatial dimensions of their pores. In this review we concentrate on the chemistry of MOFs. Specifically we bring together many aspects of MOFs that underpin their stability, reactivity and dynamic behaviour within a common theme related to (changes in) metal–ligand bonding. In each area we provide examples to illustrate the behaviour and discuss it in the context of metal lability and coordination changes. Starting with flexible behaviour in which metal–linker bonds undergo deformation rather than cleavage, we then consider coordination changes that lead to open metal sites, changes in framework topology, framework dimensionality or degree of network interpenetration. We show how these changes are linked to development of new properties, including changes in magnetic behaviour, gas adsorption characteristics, construction of composite MOFs and amorphous MOFs, as well as providing new synthetic routes for MOF preparation. We discuss how the lability of the species that make up the MOFs can affect aspects from their synthesis to the possibility of metal and linker exchange reactions that may lead to defects and disorder. The final section reviews hemilability in MOFs, where regions of different chemical behaviour within MOFs can lead to unusual properties, such as self-accelerating and ultraselective adsorption.

image file: c7cs00187h-p1.tif

Russell E. Morris

Russell Morris was born and raised in north Wales, and completed his education at the University of Oxford where he gained BA and DPhil degrees. He is currently Bishop Wardlaw Professor of Chemistry at the University of St Andrews. His research interests lie in the synthesis, characterisation and application of porous and layered materials including zeolites and metal–organic frameworks. He is an elected Fellow of the Royal Society (FRS), a Fellow of the Royal Society of Edinburgh, and a Fellow of the Learned Society of Wales. He is currently an associate editor for Dalton Transactions.

image file: c7cs00187h-p2.tif

Lee Brammer

Lee Brammer obtained his PhD in Inorganic Chemistry from the University of Bristol. Following a NATO fellowship at University of New Orleans and Brookhaven National Laboratory, he began his independent career at University of Missouri – St. Louis, before moving to University of Sheffield, where he is Professor of Inorganic and Solid State Chemistry. His research interests include MOFs for gas separation and catalysis, fundamentals/applications of intermolecular interactions, particularly hydrogen bonds and halogen bonds, and dynamic aspects of the solid state, including flexibility and solid–gas reactions. He has strong interests in the hybridisation of diffraction methods with spectroscopic and computational methods for improving insight into the chemistry of the solid state. He is currently president of the British Crystallographic Association.

1 Introduction

Metal–organic frameworks (MOFs) are constructed from metal ions or small metal clusters linked by multitopic organic ligands into periodic microporous materials. The metal–ligand (M–L) coordinate bonds that connect the metal and organic components are generally weaker than most covalent bonds that are found in other types of porous materials, but sufficiently strong to impart the stability necessary to evacuate pores and enable the wide range of applications that are at the centre of research involving MOFs, including gas adsorption and separation, catalysis, sensing and drug delivery.1–5 Nevertheless, these M–L bonds are both thermodynamically the weakest and kinetically the most labile links in most MOFs and it is cleavage of these bonds that provides the limits to thermal stability of this class of materials and often determines the chemical stability of MOFs.

This apparent weakness, however, has its advantages. It is no accident that most MOFs contain labile metal ions. This is important in the self-assembly process by which MOFs are formed, usually from the solution phase, in which reversible metal–ligand bond formation enables correction of ‘mistakes’ in the propagation of the network. Indeed, in situ diffraction studies undertaken during MOF syntheses, either under solvothermal conditions6–8 or using mechanochemical methods,9 have demonstrated the role of reversible M–L bond formation in structural evolution of MOFs during their assembly, and highlighted opportunities for preparing kinetic as well as thermodynamic MOF products. The importance of lability was inherent in the development of supramolecular coordination chemistry using a ‘node and linker’ approach in the design of coordination polymers, first detailed in the early 1990s by Robson and his group.10 The enhancement of thermodynamic stability that led to the establishment of open-framework materials with extremely high permanent porosity can be attributed to the use of anionic ligands with multiple binding sites (carboxylates, azolates) often in combination with metal-oxo/hydoxo clusters or columnar building blocks.11–14 Strengths of M–L bonds have been investigated in the context of MOFs in a recent study by Öhrström and colleagues, who sensibly concluded that although anionic ligands generally form shorter (and stronger) bonds to metal ions than their neutral analogues, there is no need for a difference in classification of these M–L bonds.15

The emergence of ‘soft porous’ crystals,16–19 defined by their relatively large compressibility, especially with respect to traditional porous materials, can be linked to thermodynamic and kinetic properties of M–L bonds. More generally, flexible and dynamic frameworks, labelled 3rd generation materials by Kitagawa,18 often exhibit their response to external stimuli such as heat, light or guest molecules as a consequence of being able to undergo changes in metal–linker bonding. A particularly striking development in this area was the discovery of ‘breathing’ MOFs,20 which show remarkable flexibility in their structure, exemplified by the archetypal flexible MOF MIL-53.14 Such breathing MOFs can transform from a large-pore to a small-pore framework typically via deformation of the metal–linker coordination, but without the need for any bond breaking in the solid.

In this review we will explore the opportunities that the relative weakness of the metal–linker coordinate bond affords21 – namely the possibility of doing chemistry that involves the breaking/making of the metal–linker coordinate bonds. This introduces the possibility of another type of flexibility for MOFs based on changes in coordination, which we will place in the context of flexibility that arises from deformations rather than cleavage of bonds. The review will describe examples of changes in coordination around the metal centre and how this leads to changes in structure and introduces changes in the properties of MOFs. We will discuss how the lability of the metal–linker bonds can be used for post-synthetic modification of materials and the impact of such lability on the stability of MOFs. We will also explain the meaning of hemilability when applied to MOFs, and review how such a concept can be used to design new properties, such as cooperative or ultraselective adsorption, and potentially lead to novel applications. The review is not intended to be comprehensive but rather will highlight a number of examples that illustrate the behaviour and potential of the concepts that we discuss. Our aim is rather to gather together this range of behaviours under a common theme related to (changes in) metal–ligand coordination.

2 Coordination changes in MOFs and related materials

2.1 Changes in coordination geometry – deformations in metal–ligand bonding

The ease of deformation of metal–ligand bonding compared with that of typical covalent bonding, such as C–C bonds, is evident from narrow distribution of bond lengths found for the latter22 and the broader distribution observed for the former.23 More pertinent to the properties of MOFs is the more facile deformation of bond angles associated with metal–ligand bonding arising from the soft potential associated with such changes. Such deformations are inherent in the “breathing” behaviour of many of the known flexible MOFs.14,24–30 They can take the form of a change in coordination geometry at the metal centre (i.e. L–M–L angles), for example the tetrahedral to square-planar coordination change at the Co(II) centres in Co(bdp) (bdp = 1,4-benzenedipyrazolate) shown in Fig. 1.26,27 In Co(bdp) this is accompanied by an hinge-type motion (Fig. 1) which results in a change in bond angle at the coordinating atom(s) of the ligand. The related [Cu(bdt)(μ-solv)] MOF (bdt = 1,4-benzeneditriazolate; solv. = DMF or DEF) shows similar flexibility, involving the hinge motion and some ligand bending, but with significant coordination changes at the metal centres. The hinge motion provides the mechanism for pore opening and closing in breathing MOFs that involve carboxylate coordination, such as MIL-53,14 DMOF28 and MIL-88,29 and is also accompanied by a change in coordination angle at the metal centre in the recently identified continuous breathing MOF SHF-61 (Fig. 2).30
image file: c7cs00187h-f1.tif
Fig. 1 Breathing motion of Co(bdp) showing change from tetrahedral to square-planar coordination at Co(II) centres from the most open-pore “filled” form to other more closed forms. (Reproduced from ref. 26 with permission from ACS, Copyright 2010.)

image file: c7cs00187h-f2.tif
Fig. 2 The principal contributions to the pore (angle) deformation that result from breathing behaviour are shown for carboxylate MOFs DMOF, SHF-61, MIL-53 and MIL-88. Pore angle change (vertical axis) is approximately the sum of the change in coordination angle at the metal centre(s) (γ) and the hinge motion associated with rotation of the ligand about the carboxylate O⋯O vector (θ). The theoretical line defines the pore angle change associated only with the hinge motion contribution. (Reproduced from ref. 30 with permission from Macmillan, Copyright 2017.)

2.2 Non-topological changes – the formation of open metal sites

A number of MOFs are constructed from secondary building units (SBUs) comprising small metal clusters in which two or more metal ions are linked by bridging of carboxylate or azolate moieties of the linker ligands. Many of these SBUs also contain solvent molecules bound to the metal sites, but these solvent molecules (L) can often be removed by heating and/or under vacuum leaving open metal sites within the MOF pores, which can be used as sites of interaction with new guest molecules.31 The lability of the M–L bonds is enhanced by the fact the solvent molecules are coordinated in a monodentate manner. By contrast, the linker ligands typically have two or more coordination sites, imparting greater stability (and inertness), although linker ligands in MOFs can also be labile in some circumstances as described in later sections of this review.

The simplest SBU with coordinated solvent molecules is the M2(O2CR)4 paddlewheel SBU, exemplified by M = Cu(II), which in as-synthesised MOFs typically contains axially bound solvent molecules. Removal of these bound solvent molecules generates open coordination sites that have been used for reversible binding of a variety of gases, such as H2(D2),32 CH433 or C2H234 in [Cu3(BTC)2] (HKUST-1; BTC = 1,3,5-benzenetricarboxylate), and H2(D2) in NOTT-10135 and NOTT-112.36 The [M33-O)(O2CR)6]+ SBU (M = Cr(III), Fe(III) or other trivalent metal ion) found in MOFs such as MIL-10137 and the MIL-88 series29 presents one accessible coordination site at each of the three metal centres. The as-synthesised MOF typically has two (neutral) removable solvent molecules bound in addition to one anionic ligand (often OH or F), which usually remains bound upon MOF activation, providing two coordination sites for reversible binding of guests. These sites involving higher oxidation-state metals than Cu(II) will typically be somewhat less labile in their guest binding.

Probably the most extensively studied MOF with accessible open metal sites is CPO-27 (MOF-74/M-DOBDC) which is a 1D channel MOF of composition [M2(dobdc)] (M = Mg, Mn, Fe, Co, Ni, Cu, Zn; dobdc = 2,5-dioxido-1,4-benzenedicarboxylate). In the as-synthesised material the metal centres, which line the walls of the hexagonal channels, bind one solvent molecule. Activation of the MOF leads to channel walls with open metal sites, the interaction of which with a variety of guests, predominantly gases, has been examined in numerous experimental and computational studies. These studies include binding of small hydrocarbons (Fig. 3),38 separation of O2 from N2,39 binding of NO,40 CO,41 H2S42 and N2O.43 The dynamics of reversible H2, CO2 and water binding to the metal sites has been particularly intensively studied.44 In a quite different application, Yaghi and coworkers have used displaceable formate groups to bind small molecule guests to the resultant vacant Al(III) sites in the [Al8(μ-OH)8(O2CH)4(BTB)4] (MOF-520; BTB = 1,3,5-benzenetribenzoate).45 Their approach has been used to anchor these small molecules to enable accurate molecular structure determination in an adaptation of the ‘crystalline sponge’ approach46 for structure determination of molecules for which crystals cannot be grown, often due to insufficient quantities of materials being available.

image file: c7cs00187h-f3.tif
Fig. 3 Top: View down one hexagonal channel of a portion of the solid-state structure of CPO-27(Fe) showing binding of C2D4 to Fe(II) sites in channel walls, as determined by neutron powder diffraction (NPD); orange, red, grey, and blue spheres represent Fe, O, C, and D atoms, respectively. Bottom: Views of H4(dobdc) and the first coordination spheres for the Fe(II) centres in the solid-state structures obtained upon dosing CPO-27(Fe) with acetylene, ethylene, ethane, propylene, and propane. (Reproduced from ref. 38 with permission from AAAS, Copyright 2012.)

In addition to direct guest binding at the metal sites, open metal sites have been used to bind diamines such as ethylene diamine (en) and its N,N′-dimethyl derivative (mmen) in a monodentate manner, leaving an uncoordinated amino group within the pores that can be used to enhance binding of CO2via R2HN⋯CO2 interactions. Studies in mmen-Mg2(dobpdc) (dobpdc = 4,4′-dioxido-3,3′-biphenyldicarboxylate),47 a larger-pore analogue of CPO-27, and the azolate MOF HCu[Cu4Cl]3(BTTri)8(en)5 (BTTri = benzene-1,3,5-tritriazolate)48 (Fig. 4) have shown promise in selective CO2 binding, with performance of the former being comparable to existing technologies such as aqueous amine solutions under simulated flue gas conditions.

image file: c7cs00187h-f4.tif
Fig. 4 View of HCu[Cu4Cl]3(BTTri)8(en)5 (BTTri = benzene-1,3,5-tritriazolate) showing monodentate binding of en ligands to open Cu(II) sites, enabling enhanced CO2 adsorption capability. (Reproduced from ref. 48 with permission from ACS, Copyright 2009.)

Changes in metal–linker bonding associated with a change of spin-state in the framework metal ions have been observed in a number of Fe(II)-containing MOFs.49,50 In a recent report, this behaviour has been exploited to enable selective, reversible binding of CO, which triggers the high-spin to low-spin transition of Fe(II) ions in the MOF Fe3[(Fe4Cl)3(BTTri)8]2 (H3BTTri = 1,3,5-tris(1H-1,2,3-triazol-5-yl)benzene).51 The spin-transition enhances CO binding, imparting very high selectivity for CO over other gases, which do not trigger the transition, and enabling removal of trace CO from mixtures with H2, N2, and CH4.

2.3 Topological transformations due to coordination changes

Breaking and formation of metal–ligand bonds in MOFs has the potential to change the topology of the framework in a number of ways.52 Here we will consider separately transformations that involve loss or gain of linker ligands and transformations that do not. Transformations that involve exchange of metal ions or exchange of linker ligands, respectively, but with no change in topology are considered in Sections 3.2 and 3.3, respectively.
2.3.1 Transformations without linker loss or gain. Potential instability of secondary building units can be exploited in enabling framework flexibility or rearrangement. Such changes are often linked to removal of coordinated solvent molecules, which lead to changes in coordination number and geometry of the metal centres in the SBU and can prompt rearrangement to a more stable MOF structure rather than decomposition. An example that clearly illustrates this concept is the rearrangement of the 3D diamondoid framework [Zn4(butIso)2(DMF)3(OH2)3]·4H2O to 2D (4,4)-net framework [Zn2(butIso)(OH2)2]·2H2O (H4butIso = 1,4-bis(5-isophthalic acid)but-2-ene).53 The transformation is prompted by desolvation of the diamondoid MOF which leads to inevitable instability of the tetrahedral 4-connected dizinc SBUs (Fig. 5). Rearrangement to planar 4-connected dizinc SBUs is further facilitated by the torsional flexibility of the central butane unit of the linker butIso4− linkers, which rearrange from projecting the four carboxylate groups in a tetrahedral manner in the initial MOF to enabling the arrangement of the carboxylate groups in a planar manner in the resultant MOF. The fragility of zinc-carboxylate SBUs upon solvent removal has been noted in a number of other MOFs.54,55
image file: c7cs00187h-f5.tif
Fig. 5 Zn2(O2CR)4L3 SBU (left) which provides a tetrahedral network node, but transforms on desolvation into Zn2(O2CR)4L2 SBU (right) that enables square-planar network connectivity. (Reproduced from ref. 53 with permission from RSC, Copyright 2007.)

The carboxylate paddlewheel SBU is one of the most commonly used motifs and found in many different MOFs, often with axial ligands, which may serve as pillars to connect the typical 2D carboxylate-propagated network into a 3D framework. Stability of the paddlewheel SBU is central to the integrity of such MOFs. However, it has been noted in a few reports that the paddlewheel SBU can undergo rearrangements resulting from M–O bond cleavage (and possibly reformation). Kitagawa has reported that cleavage of two of the eight M–O bonds, upon MOF desolvation, can lead to slippage of the metal-carboxylate layers relative to each other in an interpenetrated pillared MOF [Zn2(BDC)2(F2-dpb)]·2.5DMF·0.5H2O (F2-dpb = 2,3-difluoro-1,4-bis(4-pyridyl)benzene).56 This leads to a collapsed form of the MOF, which re-inflates during a gated gas adsorption process.

Brammer and Rosseinsky and their coworkers have separately reported more dramatic reversible rearrangements of paddlewheel SBUs that lead to different connectivity and coordination environments of the Zn(II) dimers and also to changes in dimensionality of the framework. In the 2D layered MOF [Zn2(camph)2(py)2]·2EtOH (H2camph = (1R,3S)-(+)-camphoric acid; py = pyridine), loss of EtOH from pores followed by loss of axially bound pyridine ligands results in slippage of pairs of carboxylate groups from each paddlewheel unit. This leads to bridging of the carboxylates to the Zn(II) centres in the adjacent layer, thereby “zipping” the paddlewheels into a linear assembly comprising tetrahedrally coordinated Zn(II) centres.57 This process is reversible on exposure of the resulting condensed-phase MOF to a solution containing pyridine. Alternatively the lability of the metal centres can be exploited by exposure of the condensed phase to solutions containing other monodentate or bridging pillar ligands to generate new MOFs, some of which could not be obtained by more direct synthetic routes (Fig. 6). In the 2D framework [Zn2(TBAPy)(OH2)2x(guests) (H4TBAPy = 1,3,6,8-tetrakis(p-benzoic acid)pyrene), the paddlewheel SBUs are linked by the tetracarboxylate TBAPy4− ligand and carry axially bound water molecules.58 Removal of these axial water molecules induces reversible transformation of the paddlewheel units by migration of two of the four carboxylate moieties to give a series of dizinc-dicarboxylate units further cross-linked by additional pairs of carboxylates (Fig. 7). The overall effect on the structure is to convert the 2D framework into a 3D framework, [Zn2(TBAPy)], which retains its porosity. The mechanism for the transformation was established by a combination of powder X-ray diffraction, SS-NMR spectroscopy and computer simulations. The versatility of the Zn2(O2CR)4 moiety is further demonstrated in a study by Vittal in which another arrangement of this motif is involved in single-crystal-to-single-crystal transformations of a threefold-interpenetrated pillared-layer MOF with pcu topology, [Zn2(bpeb)(obc)2]·5H2O (bpeb = 1,4-bis[2-(4-pyridyl)ethenyl]benzene; H2obc = 4,4′-oxybisbenzoic acid).59

image file: c7cs00187h-f6.tif
Fig. 6 Proposed mechanism for MOF transformations involving paddlewheel SBUs in [Zn2(camph)2(py)2]. The reactions involve slippage of carboxylate groups upon pyridine (py) ligand removal to enable zipping of Zn(II) centres into a continuous chain, followed by unzipping upon addition of a new ligand (here pyrazine, pyz). (Reproduced from ref. 57 with permission from Wiley, Copyright 2013.)

image file: c7cs00187h-f7.tif
Fig. 7 Connectivity changes in transformation of [Zn2(TBAPy)(OH2)2x(guests) to [Zn2(TBAPy)]. (a) Paddlewheel Zn2 units in [Zn2(TBAPy)(OH2)2]. (b). Removal of axial water ligands leading to paddlewheel rearrangement and cross-linking. (c) Resultant cross-linked dizinc-dicarboxylate units in [Zn2(TBAPy)]. Crystal structures of [Zn2(TBAPy)(OH2)2] and [Zn2(TBAPy)] are also shown. (Reproduced from ref. 58 with permission from ACS, Copyright 2012.)

Solvent-induced changes in changes in metal–linker coordination have been observed in a number of coordination polymers, which are not necessarily formally porous. In such cases, removal of coordinated solvent creates a vacant coordination site and enables formation of a new metal–linker bond, increasing the dimensionality of the network. The transformation may also be reversible allowing a reduction in network dimensionality. A number of examples are described in an extensive review by Kole and Vittal.60

2.3.2 Transformations involving loss or addition of linkers – changes in framework dimensionality. Addition or removal of linker ligands can, at a local level, result in defect formation in MOFs, as discussed in Section 3.4. A systematic approach to introduction or removal of a class of linkers, however, can be used as a synthetic strategy which typically results in a change in framework dimensionality.61 Such a strategy for stepwise construction of MOFs was reported by Kitagawa and co-workers over 10 years ago and involved the insertion of pillar ligands into 2D layered MOFs to connect the layers into a 3D framework, as shown schematically in Fig. 8.62 The approach conceptually resembles that used in forming pillared clays and has been most well-developed for use in layered MOFs based on paddlewheel SBUs, in which replacement of axially bound solvent molecules with ditopic ligands enables pillaring. The approach has been used by a number of groups.63–65 It can be used to enhance framework rigidity64 and can even be applied in a multi-step manner66 to build more complex architectures that may not otherwise be synthetically accessible (Fig. 9). Linker insertion has also been used to synthesise MOFs by linking together polyhedral coordination cages, i.e. a 0D to 3D framework transformation.67,68
image file: c7cs00187h-f8.tif
Fig. 8 Schematic representation of construction of 3-D porous framework by pillar insertion. (Reproduced from ref. 62 with permission from ACS, Copyright 2004.)

image file: c7cs00187h-f9.tif
Fig. 9 Construction of 3-D porous framework by stepwise pillar insertion. (PPF = porphyrin paddlewheel framework). (Reproduced from ref. 66 with permission from RSC, Copyright 2012.)

Other examples of linker insertion and removal are less common. An unusual example involving linker loss leading to dimensionality increase is observed in the nonporous 1D coordination polymer [Ag4(O2C(CF2)2CF3)4(TMP)3] (TMP = tetramethylpyrazine) which, under mild heating, loses one-third of TMP ligands via cleavage of Ag–N bonds leading to collapse to form a new 2D coordination polymer [Ag4(O2C(CF2)2CF3)4(TMP)2] through formation of new bridging Ag–O bonds.69 The transformation is facilitated by the lability of the Ag(I) centres and the volatility of TMP which is extruded from the crystals into the vapour phase. The transformation has been followed in situ by PXRD and mechanistic evidence gathered from multi-technique approach to the experimental study.70

2.3.3 Transformation involving change of framework interpenetration. A number of MOFs and coordination polymers are interpenetrated, meaning that the crystal structure comprises more than one copy of the same network and these networks thread through one another.71 In a small number of cases more than one degree of interpenetration is accessible for materials with a given composition and framework. Very few examples of solid-state transformations that change the level of interpenetration have been reported. The 3D MOF [Zn7O2(nbd)5(DMF)2] (MOF-123; H2nbd = 2-nitrobenzene-1,4-dicarboxylic acid), which adopts a non-interpenetrated pcu topology is converted, upon heating to remove DMF solvent molecules, to a 2-fold interpenetrated nonporous analogue [Zn7O2(nbd)5] (MOF-246).72 The conversion occurs in single crystals, resulting in a halving of one of the macroscopic crystal dimensions. In situ PXRD studies suggest that there are a series of intermediate structures formed during the overall transformation. Barbour and coworkers have demonstrated conversion from 2-fold interpenetrated to 3-fold interpenetrated forms of the pillared paddlewheel framework [Zn2(ndc)2(bpy)] (ndc = naphthalene-2,6-dicarboxylate; bpy = 4,4′-bipyridine).73 The transformation occurs in a single-crystal-to-single-crystal manner upon heating, which is accompanied by desolvation. The proposed transformation mechanism (Fig. 10) is supported by computational modelling studies. In a follow-up study the same mechanism was investigated in the Co(II) analogue, the rationale being that with a less labile metal ion, due to its larger crystal field stabilisation energy, the transformation in [Co2(ndc)2(bpy)] should be slower than in its Zn(II) counterpart.74 This proved to be the case and a 2-fold interpenetrated intermediate could be isolated and structurally characterised. Gas adsorption studies for the intermediate and the 3-fold interpenetrated form demonstrate their different absorption behaviour. Such studies illustrate that careful attention should be paid to activation conditions as structural changes, including changes in interpenetration, can be induced depending upon the activation conditions applied.
image file: c7cs00187h-f10.tif
Fig. 10 Schematic representation of the suggested mechanism for transformation from 2-fold to 3-fold interpenetrated frameworks of [Zn2(ndc)2(bpy)]. Independent frameworks are coloured red, green, and blue. The Zn2-paddlewheel SBUs are represented by circles, the ndc linkages by solid lines, and the bpy linkages by double-ended arrows. In the projections shown, the ndc-linked layers form horizontal planes that are pillared in the vertical direction by means of bpy ligands. (Reproduced from ref. 73 with permission from ACS, Copyright 2014.)

Perhaps one of the most interesting and surprising examples of lability in MOFs is the use of pressure to form a porous material from a non-porous, denser solid. At first sight it seems counterintuitive that a less dense phase could be prepared using high pressure. Chapman and co-workers75 prepared an interpenetrated zinc cyanide based MOF, consisting of two independent networks that are entangled. However, the space filling of the interpenetration is not optimal and application of hydrostatic pressure, using pressure-transmitting fluids, leads to a situation where the interactions between fluid and framework outweigh the framework–framework interactions and it eventually becomes favourable for one of the two entangled networks to breakdown to leave an open framework with the pores filled by the fluid. Releasing the pressure removes the occluded fluid, leaving the porous material, whose exact nature depends on the pressure-transmitting fluid used (Fig. 11). This is a remarkable example of how lability of the metal–linker bonds can be rather subtle in controlling how the framework breaks down.

image file: c7cs00187h-f11.tif
Fig. 11 Hydrostatic pressure applied to an interpenetrated MOF can result in the material rearranging to form a single framework of a porous MOF. The overall result is the formation of a porous MOF from a non-porous one by application of high pressure. (Reproduced from ref. 75 with permission from ACS, Copyright 2013.)

Other reported transformations include the solid-state transformation of the complex [Cd(H-Me-trz-pba)2(OH2)4] to 5-fold interpenetrated diamondoid coordination polymer[Cd(H-Me-trz-pba)2] (H-Me-trz-pba = 3-methyl-4-(p-benzoate)-1,2,4-triazine) by thermal dehydration (310 °C). Similarly the 1-D coordination polymer[Cd(Me-3py-trz-pba)2(OH2)2]·H2O was converted into 4-fold interpenetrated diamondoid coordination polymer [Cd(Me-3py-trz-pba)2] (Me-3py-trz-pba = 3-methyl-4-(p-benzoate)-5-(3-pyridyl)-1,2,4-triazine). Both transformations could be followed by PXRD.76

2.3.4 Single-crystal-to-single-crystal transformations. Although some examples of transformations that occur in a single-crystal-to-single-crystal manner are noted in the previous sections, such transformations are far fewer in number than ones that occur in microcrystalline powders. Retention of single crystals throughout the transformation requires that the stresses and strains on the crystal exerted by the changes in coordination can be accommodated without fracturing the crystal. Many reported examples of single-crystal transformations may also need to be treated with caution if the transformation has not been carefully studied during its progress. This is particularly pertinent to cases in which a single crystal is in contact with the solution phase as dissolution-recrystallisation may allow the transformation to start with a single crystal and end with a single crystal without retention of single-crystal form during the transformation. This possibility is illustrated in a cautionary study by Khlobystov, Schröder and coworkers, who used in situ microscopy (AFM and SEM) to investigate a series of metal-ion, ligand or counter-ion exchange reactions in crystals of coordination polymers in contact with the solution phase.77

2.4 New properties resulting from coordination changes

The motivation behind studies of the solid-state transformations described in the previous sections, although related to developing an improved understanding of dynamic behaviour in the solid state, is principally driven by the desire to find new synthetic routes to materials with useful properties. Mendes and Almeida Paz have reviewed some examples of transformations in MOFs that involve changes in metal–ligand coordination and result in materials with new or improved properties.78 Their review provides examples of improved gas adsorption, luminescence, application as a sensor and magnetic behaviour. Further examples of MOFs and coordination polymers in which magnetic behaviour is changed by changes in coordination bonds in the framework include studies by Kitagawa79 and by Mínguez Espallargas.80 In the former example, the 2D framework material [Mn(NNdmenH)(OH2)Cr(CN)6]·H2O (NNdemen = N,N-dimethylethylenediamine) is converted to a 3D framework [Mn(NNdmenH)Cr(CN)6] by loss of water and formation of new Mn–N bonds by bridging of formerly terminal cyanide ligands. The transformation occurs in a single-crystal-to-single-crystal manner upon heating to 343 K and is reversible by exposure of the crystal(s) to air at 298 K. The change in framework structure prompts a change in ferrimagnetic ordering temperature, Tc, which is repeatably switchable between 35.2 K and 60.4 K (Fig. 12). The study by Mínguez Espallargas involves stepwise chemisorption of HCl gas by a 1D coordination polymer [Cu(pyim)(Cl)(MeOH)] (pyimH = 2-(imidazol-2-yl)pyridine). Molecules of HCl insert into and cleave Cu–N bonds and form new Cu–Cl and N–H bonds, converting the coordination polymer firstly into weakly associated molecular crystals and ultimately ionic crystals, which exhibit different magnetic properties: strong antiferromagnetic Cu–Cu interactions in the coordination polymer, weak ferromagnetic interactions in the molecular crystals and weak ferromagnetic interactions coexisting with a dominant anti-ferromagnetic coupling in the ionic crystals. The transformation enables a repeatedly switchable behaviour upon HCl uptake and release to be monitored by change in EPR signal.80
image file: c7cs00187h-f12.tif
Fig. 12 Temperature dependence of molar magnetic susceptibility, χM, for 2D-framework [Mn(NNdmenH)(OH2)Cr(CN)6]·H2O (labelled as 1 (red)), dehydrated 3D framework [Mn(NNdmenH)Cr(CN)6] (labelled as 1a (blue)), and rehydrated 2D-framework (labelled as 1b (green)) in an applied dc field of 500 Oe. Insert shows reversible Tc switching by dehydration/hydration treatments (red/blue arrows, respectively) that result in framework coordination changes. (Reproduced from ref. 79 with permission from ACS, Copyright 2007.)

A number of studies have taken advantage of labile metal–ligand bonds on the surface of crystals to change the surface properties or construct composite MOFs. Matzger and coworkers investigated the fact that, unlike the highly porous HKUST-1, its Zn counterpart Zn-HKUST-1, ([Zn2(BTC)3]) shows no evidence of porosity to a variety of gases despite efforts to activate the MOF with different activation solvents, sc-CO2 or by flow of N2 gas, and despite the fact that it retains crystallinity upon activation (desolvation). They were able to show using positron annihilation lifetime spectroscopy (PALS) that the MOF does in fact contain pore space within the material, consistent with its retention of crystallinity, but these pores are sealed at the surface due to collapse of the MOF in this region.55 This presumably arises from the instability of the Zn2(O2CR)4 paddlewheel SBU on removal of axial (solvent) ligands, as described in Section 2.3.1. Kitagawa, Furakawa and colleagues have constructed hybridised MOFs by epitaxial growth of one MOF on the surface of another after changing one of the linker ligands (or alternatively the metal ion used).81–85 These composite materials provide a foundation for new materials based on MOFs in which different components of each crystal have different functionality (Fig. 13). Construction of these materials is dependent on labile metal–ligand bonds at the surface of the crystal. The approach has been extended to prepare core–shell MOFs and binary Janus MOFs,84 for applications as sensors, and to sequential functionalisation of MOF crystals towards multifunctional materials (Fig. 14).85

image file: c7cs00187h-f13.tif
Fig. 13 (a) Schematic representation of paddlewheel MOFs [M2(dicarboxylate)2(N-ligand)]. (b) Schematic representation of anisotropic epitaxial growth of a second framework, on the core framework. (Reproduced from ref. 81 with permission from RSC, Copyright 2009.)

image file: c7cs00187h-f14.tif
Fig. 14 Sequential functionalisation of MOF crystals to prepare multifunctional materials. (Reproduced from ref. 85 with permission from Wiley, Copyright 2011.)

The lability of metal–ligand bonds is also instrumental in the development in recent years of amorphous MOFs (aMOFs) and MOF glasses.86–88 The transformations from crystalline MOF to aMOF occurs via the introduction of disorder into the crystalline frameworks through metal–ligand bond cleavage. Small-scale disorder in the form of defects, in which long-range order is retained, is discussed in more detail in Section 3.4. Complete disruption of the long-range order leads to amorphisation. This can be achieved through heating, pressure (both hydrostatic and nonhydrostatic), and ball-milling. Potential applications of aMOFs include guest immobilisation, drug delivery, and development of superstrong metal–organic glasses.86

3 Lability in MOFs

3.1 Lability in MOF synthesis

Lability is, of course, an important concept in chemistry. It refers to the ability of the system to change its bonding, and leads to complex dynamics that belies the time- and space-averaged nature of the picture we often have for framework materials such as MOFs. As in many chemical systems, when a MOF is in the right environment, such as in contact with a solution similar to that required for its synthesis, there is always the possibility of dynamics that are not at first apparent to the usual structural characterisation techniques. This could be in the form of opening and closing of bonds to produce defects, or could be the wholescale interchange of metals or linkers. When in contact with the original synthesis solution the lability of the species that make up the MOF obviously has an impact on the kinetics of crystal formation and crystal growth, as reversibility plays an important part in the overall crystallisation process. Crystallisation kinetics clearly depend on the lability characteristics of the metals in the system, and there are now several examples of in situ crystallisation studies of MOFs that demonstrate this.6–8,89,90 Of course, in completing such studies one also has to take into account the thermodynamics of the system. The lability characteristics are of particular importance because they provide the pathway by which endpoints of different thermodynamic stability can be reached.

An interesting family of MOFs to study are those where the same framework topology can be made using different metals. There are now several different in situ studies that show the dependence of the metals in the synthesis of MOFs.89 Many synthetic pathways are actually relatively complex and can go through several intermediate phases. A beautiful in situ X-ray diffraction study was completed by O’Hare and co-workers7 on mixed-ligand MOFs that can contain different metal ions such as cobalt(II), zinc(II) and nickel(II). The crystallisation pathways were complex, involving several different possible intermediate materials (Fig. 15), but the relatively slow crystallisation of the nickel(II) material (Fig. 16) can be traced back to lability and chemical stability of its complexes. There are now many such studies in the literature that demonstrate the importance of the lability characteristics of the metals.90

image file: c7cs00187h-f15.tif
Fig. 15 Mixed DABCO/carboxylate MOFs with different topologies can be prepared with different metal ions including Zn(II), Co(II) and Ni(II). The formation of these materials can be studied using in situ X-ray diffraction methods and their crystallisation kinetics related to the lability of the different metal ions. (Reproduced from ref. 7 with permission from Wiley, Copyright 2016.)

image file: c7cs00187h-f16.tif
Fig. 16 The complex kinetic studies on zinc and nickel DABCO/carboxylate MOFs (from Fig. 15; orange = kgm; green = sql; pink = total). The much slower crystallisation of the nickel material can be traced back to the differences in lability of the metal ions. (Reproduced from ref. 7 with permission from Wiley, Copyright 2016.)

Of course, changing the linker in the system can also affect the lability of the metal–linker bonds. The development of reticular chemistry means that there are now series of similar ligands that can be used to study the formation of the same or similar network topologies but with different organic linkers. Horcajada and co-workers91 developed a particularly nice study on the isoreticular series formed by zirconium(IV) ions in combination with different terephthalates or other linear polyaromatic dicarboxylates in the formation of UiO-66 MOF analogues (Fig. 17). These materials are particularly intensely studied MOFs at the present time because of their high stability and interesting defect properties.92,93 Changing the organic linker has a significant effect on the crystallisation kinetics, some of which can be traced back to the solubility of the linker in the particular solution. There are now several such studies in the literature that show similar effects.94,95

image file: c7cs00187h-f17.tif
Fig. 17 (a) The UiO-66 structure and (b) examples of organic linkers that can be used to replace the terephthalate linker. (Reproduced from ref. 91 with permission from Wiley, Copyright 2015.)

A different example of the reversibility in bonding and how it affects crystal growth is the use of modulators96,97 – ligands added to the synthesis solution that compete for coordination to the metals with the linker ligands. The kinetics of MOF formation can depend markedly on the use of these ligands that can bind to, for example, particular faces of a growing crystallite selectively. Such a technique is now a relatively common method by which aspects such as crystal size and morphology are controlled by changing the nucleation and crystal growth kinetics. Much information on the mechanism can be deduced from careful studies of the kinetics of the process, as has been shown in recent work by the Behrens group, where the interplay between the modulator (a monocarboxylic acid) and the solvent can be seen to either increase or decrease the rate of formation of a zirconium fumarate MOF depending on the conditions.98

3.2 Lability and metal exchange

Perhaps the most obvious potential use of lability in MOFs is the possibility for post-synthetic modification of frameworks through either exchange of metals or exchange of linkers.99,100 Post-synthetic modification of MOFs is now a major research topic, and this whole field depends on the solids themselves being reactive to some degree. One option for post-synthetic modification occurs because of the inherent lability of the metal ions in most MOFs, which offers the possibility of metal-ion metathesis. A good example of this is in the preparation and post-synthetic modification of POST-65.101 The as-synthesised form of this material is a Mn(II)-based MOF, but when exposed to solutions containing various different metal ions (Fe(II), Co(II), Ni(II) or Cu(II)) the lability of the metal ions present, in both the starting structure and the solutions, allowed the metals to exchange (Fig. 18). A particularly striking feature of this work is the claim that the exchange happens in a single-crystal-to-single-crystal process (see bottom panel of Fig. 18) and is not a dissolution–recrystallization effect. However, as noted in Section 2.3.4, one should be wary that apparent single-crystal-to-single-crystal transformations involving crystal immersion in solvent may in fact go via dissolution–recrystallisation.77 The interesting feature of this work that supports a single crystal metal ion exchange mechanism is that preparation only of the Mn-POST-65 material has been accessible by solvothermal synthesis – the variants containing the other metals have hitherto only been prepared by the metal-ion metathesis process.
image file: c7cs00187h-f18.tif
Fig. 18 The concept of metal-ion metathesis in MOFs, where a framework with one metal is transformed into the same MOF but with a different metal simply though metal exchange reactions. The bottom panel shows a single crystal of Mn-POST-65 that is placed in contact with a solution containing Cu(II) ions. The replacement of the Mn(II) ions by Cu(II) is shown to be a single-crystal-to-single-crystal transformation in the bottom panel. (Reproduced from ref. 101 with permission from Wiley, Copyright 2012.)

Post-synthetic metal exchange reactions can also have significant effects on the physical properties of MOFs. A particularly striking example of this is metal exchange reactions in the CPO-27 (MOF-74/M-DOBDC) type structures. Wright and co-workers102 showed that certain samples of the Mg2+ version of these materials showed little or no nitrogen uptake at 77 K. However, nickel(II) cations were introduced into framework sites of the magnesium dioxidoterephthalate MOF, CPO-27(Mg), via a procedure involving post-synthetic treatment with an aqueous solution of Ni2+ cations and a weak acid. The final solids have Ni concentrations of up to 70 mol% of the total cation content. Selected area EDX analysis, synchrotron X-ray powder diffraction and XPS surface analysis indicate that the nickel is distributed throughout the crystals with highest concentration at the external surface. The modified solids have much enhanced adsorption capacities following simple thermal evacuation under vacuum compared to unmodified CPO-27(Mg), particularly for N2 at 77 K. This is attributed to surface modification that makes the surface more stable to heating under evacuation. A further inference from work on post synthetic metal metathesis is that this may provide an alternative method of preparing core–shell type materials with different concentrations of metal ions at the surface compared to those in the centre of the crystallites.103

Although the occurrence of post-synthetic metal metathesis is now well established99–102,104,105 the precise mechanism of the process is yet to be elucidated completely. Recent work by Dincă and co-workers has, however, used computational methods to predict two possible mechanisms for the process and showed how the degree of metal metathesis in MOF-5 is controlled by the strength of the metal–solvent interaction.106

3.3 Lability and linker exchange

In the same way that metal ion metathesis can take place when a MOF is placed in contact with a solution of metal ions that are different to those in the framework, when a MOF is placed in a solution of a potential linker then there are possibilities for linker exchange.107 There are now several examples of this type of exchange process, referred to as solvent-assisted linker exchange by Hupp and coworkers.107–111 The framework bMOF-100 is cubic, and so replacement of linkers with longer ones leads to pore expansion in all three dimensions. The work by Rosi and coworkers112 shows that this can be achieved sequentially, successively replacing one linker with another and leading to a series of pore-expanded materials. As with the metal ion metathesis described in Section 3.2, it is very likely that this exchange occurs in an ‘outside to inside’ direction within crystals, and proper control over the linker exchange process may lead to core–shell type materials in which the outer layers have larger pore openings than the interior ones (Fig. 19). This concept was demonstrated for the bMOF series by selectively adsorbing large gold–sulfide clusters only in the large pores at the surface of the crystallites.112
image file: c7cs00187h-f19.tif
Fig. 19 An example of linker exchange reactions that successively leads to expanded pore sizes based on replacing the shortest linker in bMOF-100, with longer linkers to eventually end up with bMOF-107. (Reproduced from ref. 112 with permission from ACS, Copyright 2016.)

3.4 Defects and disorder

The concept of disorder in MOFs has, over recent years, gained much traction in the research community.113–115 It is now generally recognised that many MOF materials contain significant defect concentrations. Furthermore, it is established that the defects and disorder in MOFs can play an important part in the properties of the materials.116,117 An archetypal defective MOF is the zirconium-based UiO-66 shown in Fig. 20, and its hafnium analogues. These materials often have a small proportion of the dicarboxylate linkers missing,118,119 leaving zirconium hydroxide defects or even defects where entire metal clusters are missing.120 These defects are equally prone to dynamics caused by the lability of bonds in the structure.121 Of course, characterisation of such situations poses significant experimental challenges, but computational work122–124 plays an important part in our understanding of the structural chemistry and the dynamics of the defects themselves, and how they can affect material properties. Such studies offer great insight into the materials themselves and are set to play an ever more important role in deciphering the subtle details of these important materials.
image file: c7cs00187h-f20.tif
Fig. 20 The structure of defective UiO-66 with one missing linker per unit cell (shown as a transparent linkers). Computational studies show that the hydroxyl/water (O0, O1 and O3) units are dynamic in nature showing how lability in the metal–oxygen bonds affects the structure of the material, and how it can have effects on the acidity and catalytic properties of the material. (Reproduced from ref. 122 with permission from RSC, Copyright 2016.)

3.5 Stability to or degradation by exposure to water

For many applications of MOFs, stability to water, either liquid or vapour, is important, although for MOFs designed for in vivo applications such as drug delivery, controlled dissolution in aqueous medium is essential.125 The adsorption of water and its effect on MOFs has been reviewed126–128 and will be highlighted more briefly here in the context of coordination changes and lability in MOFs. Degradation of carboxylate MOFs, which occurs by cleavage of metal–carboxylate bonds, has been studied extensively and stability is related to M–O bond strength and lability. MOFs constructed from more labile metal ions, predominantly divalent first-row transition metals, tend to be less stable to degradation by water than those containing higher-charged metals. A combined experimental and computational investigation by Low, Willis and coworkers of a series of well-established MOFs highlights these points by discerning the likely mechanism of MOF degradation (Fig. 21).129 The study further compares the calculated stability of the MOF SBUs to ligand displacement by water with the experimentally determined maximum stability of the MOF to exposure to steam (Fig. 22). These studies reveal that one of the most widely studied MOFs, [Zn44-O)(BDC)3] (MOF-5), is particularly susceptible to degradation by water. Indeed reported gas adsorption capacities vary markedly for this reason where such degradation is not controlled; careful synthesis under anhydrous conditions, however, leads to much improved adsorption capacities, notably for hydrogen.130
image file: c7cs00187h-f21.tif
Fig. 21 Computationally modelled ligand displacement reaction for (A) MOF-5, (B) HKUST-1, (C) MIL-101, and (D) ZIF-8. Structures represent reactant, transition state, and hydrolysis product clusters for each MOF, respectively. Colours: C, grey; O, red; H, white; Zn, light blue; Cu, purple; Cr, dark green; F, pink; Cl, light green; N, dark blue. (Reproduced from ref. 129 with permission from ACS, Copyright 2009.)

image file: c7cs00187h-f22.tif
Fig. 22 Steam stability map for a series of MOFs. The position of the structure for a given MOF represents its maximum structural stability as probed by PXRD measurements, whereas the energy of activation for ligand displacement by a water molecule as determined by DFT computational modelling is represented by the magenta numbers (in kcal mol−1). (Reproduced from ref. 129 with permission from ACS, Copyright 2009.)

4 Hemilabile MOFs

4.1 What is a hemilabile MOF?

In solution-state coordination complexes, a hemilabile ligand is a multidentate species that contains donor groups that have different binding properties.131,132 The defining feature of these ligands is that one donor group is more easily displaced from the metal centre than is the other. The concept of hemilability can also be applied to MOFs. In this case the linker should have two types of donor group, one of which binds strongly to the metals, while interesting chemistry (i.e. the breaking and making of bonds) can take place at the more weakly binding site. The general concept behind the idea of hemilabile MOFs is shown in Fig. 23.
image file: c7cs00187h-f23.tif
Fig. 23 Schematic of a hemilabile MOF, with a mixture of strong metal–linker bonds (green) and weaker metal–linker bonds (pink). In this situation strong and weak really refer to less-labile and more-labile bonds, respectively. The strong bonds hold the material intact while chemistry involving the changes in metal coordination can take place at the weakly bonded site. There needs to be enough strong bonds to hold the framework together.

A good example of a set of organic linkers that can be used to make hemilabile MOFs are those containing a mixture of carboxylate and sulfonate groups, such as the 5-sulfoisophthalate (SIP). This linker is commercially available in its acid form (H3SIP) and has been used to make a relatively large number of MOFs with a variety of different metals.133–135 A common feature of many of these MOFs is that they undergo structural transformations on dehydration (or other chemical stimuli). Several of the materials show good properties for gas adsorption, for example the mixed-metal sodium-nickel SIP MOF first reported by Cheetham and coworkers,136,137 which shows excellent hydrogen storage properties.

A well-studied member of this family formed from the reaction of copper(II) ions with the SIP linker is the MOF material called Cu-SIP-3.138–140 The structure of this material is shown in Fig. 24 In its as-synthesised, hydrated form the material consists of tetrameric copper clusters linked by the organic SIP ligand into a three-dimensional structure. Importantly, both oxygen atoms from each of the carboxylate groups are linked to the metal ions. However, only two of the three possible sulfonate oxygen atoms are bound to copper ions. There is also a water molecule attached to the copper ions. As the temperature is raised the water begins to be removed. If nothing else happened this would leave a coordinatively unsaturated copper centre and an open metal site. However, as the water is removed there is a structural change in the material, shown in Fig. 24, where the sulfonate group reorients so the previously uncoordinated oxygen is in a position to coordinate. During the whole process the coordination of the carboxylate groups remains unchanged. The process is perfectly reversible, so that on rehydration the metal–sulfonate bond is broken and is replaced by the adsorbed water molecule to reform the hydrated material. In this way the SIP linker is clearly acting as a hemilabile linker.

image file: c7cs00187h-f24.tif
Fig. 24 The PDF-derived mechanism for hemilability in Cu-SIP3. The change in structure occurs because the sulfonate group coordination changes (for example at point A in the figure). This process can happen at any point in the material, leading to a situation where the full transformation happens at random points, leading to a loss of order, shown in the right-hand diagram as a change from the ordered low-temperature structure (shown as blue) through a half-transformed region (A, shown in green) through to the fully transformed region shown in orange. (Reproduced from ref. 140 with permission from RSC, Copyright 2012.)

There are several interesting features of this particular material, but perhaps the most fascinating is the behaviour of the single crystals. At low temperatures (below 370 K) the hydrated structure can be determined by standard single crystal X-ray diffraction methods. Above 405 K the high temperature structure can also be determined successfully. However, during the transformation process between 370 and 405 K the Bragg diffraction from the single crystals is essentially lost completely. In other words the hemilabile intermediate goes through a disordered ‘crystal’ state. X-ray diffraction is therefore not the best technique to study this particular process, but a total scattering method like PDF analysis can give information on the actual mechanism of the process. The PDF-derived mechanism can be seen in Fig. 24.

There are several other examples of hemilability using the SIP ligand, but such a property is not limited to this particular ligand. Other organic linkers, such as 2-sulfoterephthalate produce MOFs that show similar coordination changes around the metals.141

Another class of material that can be considered hemilable MOFs those based on the well-known copper paddle-wheel unit in combination with functionalised benzene tricarboxylate ligands (such as the MOF STAM-1142–144) or other substituted isophthalate ligands (such the azo-isophthalate (aip) based MOFs prepared by Kitagawa and coworkers145). The structural behaviour, and the consequences in terms of properties (see below) are subtly dependent on the chemical composition of the solids. However, the common feature in all the materials seems to be a change in coordination around the metals (which is usually copper). In the as-synthesised (hydrated) structure the paddlewheel units are isolated from each other as they are in many MOFs, such as HKUST-1. On dehydration, however, the water molecules that are connected to the paddlewheels (one to each copper centre) are lost. Again, as for the SIP materials described above, in most cases this dehydration would lead to an open metal site. However, in the hemilabile materials there is enough flexibility in the solid to move the paddlewheels relative to each other so that the copper centres are coordinated through relatively long interactions to oxygens in another paddlewheel unit (Fig. 25).

image file: c7cs00187h-f25.tif
Fig. 25 (A and B) The low temperature structure of Cu(aip) showing isolated Cu-paddlewheel units compared with the high temperature structure (C and D) where there has been a change of coordination around the copper(II) centres. The changes in coordination change the shape of the pores running through the structure considerably. Key: Cu = green, C = grey, O = red. (Adapted from ref. 145 with permission from AAAS, Copyright 2014.)

4.2 Properties of hemilabile MOFs

It is clear from the discussion above that there are actually significant numbers of hemilabile MOFs already available, but the question always remains as to what interesting and potential useful properties hemilability might introduce into the final materials. In the next section we will describe some of the interesting features, especially in the area of unusual adsorption.
4.2.1 Ultraselective adsorption on Cu-SIP-3. The structural changes described above for Cu-SIP-3 lead directly to an unusually high selectivity in gas adsorption.138 The dehydrated structure of Cu-SIP-3 is actually non-porous, and there is negligible uptake of any of the common gases used to probe adsorption such as N2, H2, CO2, CH4, etc. under the normal experimental conditions used for the particular gas. The only form of this material that is potentially porous is the as-synthesised/hydrated one, which contains non-coordinated water occluded into the pores. For any gas to be adsorbed it must interact strongly enough with the framework to stimulate the transformation from the dehydrated structure to the porous structure. The only gases likely to do this are those that coordinate fairly strongly to the metal sites. Nitric oxide (NO) is one such gas that is known to coordinate to metals in MOFs relatively strongly,146 so that it can replace the weak metal–sulfonate bonds. Indeed, because of this NO is the only gas so far measured that can be adsorbed by Cu-SIP-3 precisely because it reverses the phase transformation. NO in itself is an interesting adsorbate to study, both for the removal of NO from exhaust gases (as NO is toxic in large amounts) and for delivery of smaller, biologically beneficial amounts for medical applications.147
4.2.2 Self-accelerating adsorption. The Cu(aip) MOF prepared by Kitagawa (Fig. 26) shows the very interesting feature of self-accelerating adsorption.145 As for Cu-SIP-3, the hemilabilty in this material closes some of the pore space in the solid, which is only opened up by interaction upon adsorption of the first molecule of CO. Further adsorption is then made easier by the presence of this initial molecule, and leads to the adsorption ‘self-accelerating’ (Fig. 26). The Kitagawa group completed beautiful characterisation work that shows how this property is derived. Of particular significance is how, once again, the hemilability of the material leads to high selectivity in gas adsorption. In this case the self-accelerating nature of the CO adsorption offers great possibilities in, for example, the separation of CO and N2.
image file: c7cs00187h-f26.tif
Fig. 26 (A) CO sorption isotherms at 120 K for Cu(aip). Adsorption and desorption profiles are shown in solid and open circles, respectively. (B) XRPD patterns measured at each point (a–j) shown in the CO sorption isotherms. The simulated patterns for the dried MOF and CO-adsorbed MOF are shown at the bottom and top, respectively. (Adapted from ref. 145 with permission from AAAS, Copyright 2014.)
4.2.3 Switchable adsorption. The STAM-1 type structure is very closely related to the Cu-azo-isophthalate MOF discussed in Section 4.2.2. However, the difference is that the material is prepared using benzene tricarboxylic acid that undergoes a methyl esterification reaction at one of the carboxyl groups, meaning that only two carboxylate groups per linker molecule are coordinated to the Cu(II) ions in the solid.143 Despite their similar topologies, the properties of the two MOFs are subtly different. An interesting feature of the STAM-1 material is the possibly of switching the adsorption between the two different channels in the material. There are two different types of channel in this material, one of which lined by organic ester groups and one that is lined by the metals. They therefore have quite different hydrophobic/hydrophilic properties. Once again, dehydration leads to a change in structure (but one different to that seen in Kitagawa's azo-isophthalate Cu(aip) MOF145) that again leads to the closing of one of the pores. This means that depending on the exact activation treatment one can control which pores are open for adsorption, allowing one to control or switch the adsorption between the different channels.
4.2.4 Dielectric properties of hemilabile MOFs. In recent years there has been a surge of interest in the dielectric, optical, magnetic and other physical properties of hybrid materials and MOFs.148–151 Much of this work has been driven by the important development of hybrid perovskite-type frameworks.152 The properties of these solids can be switched through phase transitions. However, switchable optical, magnetic and electric/dielectric properties of MOFs are actually very rarely seen. Hemilabile materials can play an important part in such switching as their potential for undergoing structural change when challenged with the appropriate stimulus offers the opportunity to have large changes in such properties. For example, recent work on zinc-SIP MOFs153 prepared using the ionothermal method showed multiple transitions. One example, exhibits intriguing dielectric properties. There are two-step dielectric anomalies (at 280 K and >353 K) together with two-step dielectric relaxations (Fig. 27). The structure of the material shows the same partial coordination of the sulfonate groups as seen in the other SIP materials described above. However, one of the uncoordinated sulfonate oxygens interacts with imidazolium-based cations occluded within the channels of the MOF. The dielectric anomaly occurring at approx. 280 K is associated with an isostructural phase transition, but the one at higher temperature is strongly dependent on the ac frequency, which probably arises because at higher temperature the interaction of the imidazolium cation with the sulfonate group is broken and allows dynamic disorder of the cations in the channels.
image file: c7cs00187h-f27.tif
Fig. 27 (Top) Portion of the structure of (EMIM)[Zn(SIP)] (SIP = 5-sulfonatoisophthalate, EMIM = 1-ethyl-3-methylimidazolium) showing the interaction of the SIP sulfonate oxygens with the EMIM cations. (bottom) Dielectric measurements on the material showing a frequency-independent anomaly at approx. 280 K and a frequency-dependent anomaly above 353 K. (Reproduced from ref. 153 with permission from ACS, Copyright 2016.)

5 Conclusions

Chemical features of metal–organic frameworks that are a direct consequence of the lability of the metal centres and resultant changes in metal coordination have quite widespread consequences on the properties and applications of MOFs. This review has sought to collectively examine these behaviours by emphasising their common origin.

As MOFs have developed over the last two decades, it can be argued that greatest emphasis has been placed on framework topology, pore shape and pore size as the defining features of these solids. Linker chemistry has been explored through post-synthetic modification involving a range of organic reactions to introduce new chemical functionality to enhance applications including gas separation and catalysis. The framework chemistry, namely that involving changes in metal–linker bonding has often been viewed as secondary, or in the worst cases even a nuisance (notably when chemical stability of the materials is important for their applications). It is, however, inherently of comparable importance to the framework topology, and as uses of MOFs are developed one must match both framework chemistry and topology to the needs of the proposed applications. As this review points out there is much information in the literature regarding the lability of the metal–linker bonds in MOFs. The major question to be answered in the future of MOFs is how we can take advantage of the framework chemistry to make sure the topological properties of the materials can be fully utilized in applications. There are several possible strategies. First, there is always the option of limiting the reactivity of the framework – this is akin to making the MOF simply a box that can be used to encapsulate or adsorb other entities. To do this requires making MOFs with strong and nonlabile metal–organic linkages. Of course this places limitations on potential chemical compositions and operating conditions for applications of MOFs, but nonetheless this is still a major interest in the community. At the opposite end of the spectrum we could attempt to design the chemical properties of the MOFs to be advantageous by imparting dynamic behaviour inherent in labile metal–linker or, more generally, metal–ligand coordination. This is by far a much more interesting chemical challenge both to enable and to control.

Opportunities for post-synthetic modification of MOFs as a consequence of metal lability provide opportunities that rival those of PSM of organic linker ligands. These opportunities include grafting on to open metal sites, which has already received significant attention, and post-synthetic exchange of either the framework metal ions or linker ligands. Such exchange processes have begun to receive attention quite recently both to enable synthesis of otherwise inaccessible MOF compositions or to tune chemical properties of the MOF. Perhaps the most intriguing area for the future is further development of hemilabile and other more complex systems. The ability to exert a finer control over bond breaking/bond making processes would really have a large effect on MOF chemistry. For example, the lower stability of MOFs can be traced to the indiscriminate breaking of metal–linker bonds, which eventually leads to framework collapse. If we were to control the bond breaking in a material more precisely it may be possible to direct the process so that only nonstructural bonds could be broken. This would decrease the likelihood of structure collapse. Similarly, there is now evidence that interesting and unique properties are possible by introducing some sort of hemilability. The prime example of this is the cooperative effects that lead to enhanced carbon monoxide in Kitagawa's Cu(aip) material described above. Such effects are dominated by the chemical changes occurring at the metal, and there is every reason to suggest that other interesting properties are available if we develop the strategies associated with these materials further.


REM is grateful for support from the EPSRC (EP/K025112) and the Royal Society. LB is grateful for support from EPSRC, CCDC and University of Sheffield.

Notes and references

  1. J. R. Li, J. Sculley and H. C. Zhou, Chem. Rev., 2012, 112, 869–932 CrossRef CAS PubMed .
  2. L. E. Kreno, K. Leong, O. K. Farha, M. Allendorf, R. P. Van Duyne and J. T. Hupp, Chem. Rev., 2012, 112, 1105–1125 CrossRef CAS PubMed .
  3. P. Horcajada, R. Gref, T. Baati, P. K. Allan, G. Maurin, P. Couvreur, G. Férey, R. E. Morris and C. Serre, Chem. Rev., 2012, 112, 1232–1268 CrossRef CAS PubMed .
  4. A. Corma, H. Garcia and F. X. L. I. Xamena, Chem. Rev., 2010, 110, 4606–4655 CrossRef CAS PubMed .
  5. K. Sumida, D. L. Rogow, J. A. Mason, T. M. McDonald, E. D. Bloch, Z. R. Herm, T. H. Bae and J. R. Long, Chem. Rev., 2012, 112, 724–781 CrossRef CAS PubMed .
  6. H. H.-M. Yeung, Y. Wu, S. Henke, A. K. Cheetham, D. O’Hare and R. I. Walton, Angew. Chem., Int. Ed., 2016, 55, 2012–2016 CrossRef CAS PubMed .
  7. Y. Wu, S. Henke, G. Kieslich, I. Schwedler, M. Yang, D. A. X. Fraser and D. O’Hare, Angew. Chem., Int. Ed., 2016, 55, 14081–14084 CrossRef CAS PubMed .
  8. Y. Wu, S. Moorhouse and D. O'Hare, Chem. Mater., 2015, 27, 7236–7239 CrossRef CAS .
  9. T. Friščić, I. Halasz, P. J. Beldon, A. M. Belenguer, F. Adams, S. A. J. Kimber, V. Honkimäki and R. E. Dinnebier, Nat. Chem., 2013, 5, 66–73 CrossRef PubMed .
  10. B. F. Hoskins and R. Robson, J. Am. Chem. Soc., 1990, 112, 1546–1554 CrossRef CAS .
  11. H. Li, M. Eddaoudi, M. O'Keeffe and O. M. Yaghi, Nature, 1999, 402, 276–279 CrossRef CAS .
  12. S. S. Y. Chui, S. M. F. Lo, J. P. H. Charmant, A. G. Orpen and I. D. Williams, Science, 1999, 283, 1148–1150 CrossRef CAS PubMed .
  13. A. Phan, C. J. Doonan, F. J. Uribe-Romo, C. B. Knobler, M. O’Keeffe and O. M. Yaghi, Acc. Chem. Res., 2010, 43, 58–67 CrossRef CAS PubMed .
  14. C. Serre, F. Millange, C. Thouvenot, M. Nogues, G. Marsolier, D. Louer and G. Férey, J. Am. Chem. Soc., 2002, 124, 13519–13526 CrossRef CAS PubMed .
  15. A. Nimmermark, L. Öhrström and J. Reedijk, Z. Kristallogr., 2013, 228, 311–317 CAS .
  16. S. Horike, S. Shimomura and S. Kitagawa, Nat. Chem., 2009, 1, 695–704 CrossRef CAS PubMed .
  17. G. Fraux and F.-X. Coudert, Chem. Commun., 2017, 53, 7211–7221 RSC .
  18. S. Kitagawa and K. Uemura, Chem. Soc. Rev., 2005, 34, 109–119 RSC .
  19. T. D. Bennett, A. K. Cheetham, A. H. Fuchs and F.-X. Coudert, Nat. Chem., 2017, 9, 11–16 CrossRef CAS PubMed .
  20. G. Férey and C. Serre, Chem. Soc. Rev., 2009, 38, 1380–1399 RSC .
  21. R. E. Morris and J. Čejka, Nat. Chem., 2015, 7, 381–388 CrossRef CAS PubMed .
  22. F. H. Allen, O. Kennard, D. G. Watson, L. Brammer, A. G. Orpen and R. Taylor, J. Chem. Soc., Perkin Trans. 2, 1987, S1–S19 RSC .
  23. A. G. Orpen, L. Brammer, F. H. Allen, O. Kennard, D. G. Watson and R. Taylor, J. Chem. Soc., Dalton Trans., 1989, S1–S83 RSC .
  24. A. Schneemann, V. Bon, I. Schwedler, I. Senkovska, S. Kaskel and R. A. Fischer, Chem. Soc. Rev., 2014, 43, 6062–6096 RSC .
  25. C. R. Murdock, B. C. Hughes, Z. Lu and D. M. Jenkins, Coord. Chem. Rev., 2014, 258-259, 119–136 CrossRef CAS .
  26. F. Salles, G. Maurin, C. Serre, P. L. Llewellyn, C. Knöfel, H. J. Choi, Y. Filinchuk, L. Oliviero, A. Vimont, J. R. Long and G. Férey, J. Am. Chem. Soc., 2010, 132, 13782–13788 CrossRef CAS PubMed .
  27. 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 and J. R. Long, Nature, 2015, 527, 357–361 CrossRef CAS PubMed .
  28. D. N. Dybtsev, H. Chun and K. Kim, Angew. Chem., Int. Ed., 2004, 43, 5033–5036 CrossRef CAS PubMed .
  29. C. Mellot-Draznieks, C. Serre, S. Surblé, N. Audebrand and G. Férey, J. Am. Chem. Soc., 2005, 127, 16273–16278 CrossRef CAS PubMed .
  30. E. J. Carrington, C. A. McAnally, A. J. Fletcher, S. P. Thompson, M. Warren and L. Brammer, Nat. Chem., 2017 DOI:10.1038/Nchem.2747 .
  31. T. L. Easun, F. Moreau, Y. Yan, S. Yang and M. Schröder, Chem. Soc. Rev., 2017, 46, 239–274 RSC .
  32. V. K. Peterson, C. M. Brown, Y. Liu and C. J. Kepert, J. Phys. Chem. C, 2011, 115, 8851–8857 CAS .
  33. H. Wu, J. M. Simmons, Y. Liu, C. M. Brown, X. S. Wang, S. Ma, V. K. Peterson, P. D. Southon, C. J. Kepert, H. C. Zhou, T. Yildirim and W. Zhou, Chem. – Eur. J., 2010, 16, 5205–5214 CrossRef CAS PubMed .
  34. S. C. Xiang, W. Zhou, J. M. Gallegos, Y. Liu and B. L. Chen, J. Am. Chem. Soc., 2009, 131, 12415–12419 CrossRef CAS PubMed .
  35. X. Lin, I. Telepeni, A. J. Blake, A. Dailly, C. M. Brown, J. M. Simmons, M. Zoppi, G. S. Walker, K. M. Thomas, T. J. Mays, P. Hubberstey, N. R. Champness and M. Schröder, J. Am. Chem. Soc., 2009, 131, 2159–2171 CrossRef CAS PubMed .
  36. Y. Yan, I. Telepeni, S. Yang, X. Lin, W. Kockelmann, A. Dailly, A. J. Blake, W. Lewis, G. S. Walker, D. R. Allan, S. A. Barnett, N. R. Champness and M. Schröder, J. Am. Chem. Soc., 2010, 132, 4092–4094 CrossRef CAS PubMed .
  37. G. Férey, C. Mellot-Draznieks, C. Serre, F. Millange, J. Dutour, S. Surblé and I. Margiolaki, Science, 2005, 309, 2040–2042 CrossRef PubMed .
  38. E. D. Bloch, W. L. Queen, R. Krishna, J. M. Zadrozny, C. M. Brown and J. R. Long, Science, 2012, 335, 1606–1610 CrossRef CAS PubMed .
  39. E. D. Bloch, L. M. Murray, W. L. Queen, S. Chavan, S. N. Maximoff, J. P. Bigi, R. Krishna, V. K. Peterson, F. Grandjean, G. J. Long, B. Smit, S. Bordiga, C. M. Brown and J. R. Long, J. Am. Chem. Soc., 2011, 133, 14814–14822 CrossRef CAS PubMed .
  40. F. Bonino, S. Chavan, J. G. Vitillo, E. Groppo, G. Agostini, C. Lamberti, P. D. C. Dietzel, C. Prestipino and S. Bordiga, Chem. Mater., 2008, 20, 4957–4968 CrossRef CAS .
  41. E. D. Bloch, M. R. Hudson, J. A. Mason, S. Chavan, V. Crocellà, J. D. Howe, K. Lee, A. L. Dzubak, W. L. Queen, J. M. Zadrozny, S. J. Geier, L.-C. Lin, L. Gagliardi, B. Smit, J. B. Neaton, S. Bordiga, C. M. Brown and J. R. Long, J. Am. Chem. Soc., 2014, 136, 10752–10761 CrossRef CAS PubMed .
  42. P. K. Allan, P. S. Wheatley, D. Aldous, M. I. Mohideen, C. Tang, J. A. Hriljac, I. L. Megson, K. W. Chapman, G. De Weireld, S. Vaesen and R. E. Morris, Dalton Trans., 2012, 41, 4060–4066 RSC .
  43. D. J. Xiao, E. D. Bloch, J. A. Mason, W. L. Queen, M. R. Hudson, N. Planas, J. Borycz, A. L. Dzubak, P. Verma, K. Lee, F. Bonino, V. Crocellà, J. Yano, S. Bordiga, D. G. Truhlar, L. Gagliardi, C. M. Brown and J. R. Long, Nat. Chem., 2014, 6, 590–595 CrossRef CAS PubMed .
  44. See ref. 31 and L. Valenzano, J. G. Vitillo, S. Chavan, B. Civalleri, F. Bonino, S. Bordiga and C. Lamberti, Catal. Today, 2012, 182, 67–79 CrossRef CAS .
  45. S. Lee, E. A. Kapustin and O. M. Yaghi, Science, 2016, 353, 808–811 CrossRef CAS PubMed .
  46. Y. Inokuma, S. Yoshioka, J. Ariyoshi, T. Arai, Y. Hitora, K. Takada, S. Matsunaga, K. Rissanen and M. Fujita, Nature, 2013, 495, 461–466 CrossRef CAS PubMed .
  47. T. M. McDonald, W. R. Lee, J. A. Mason, B. M. Wiers, C. S. Hong and J. R. Long, J. Am. Chem. Soc., 2012, 134, 7056–7065 CrossRef CAS PubMed .
  48. A. Demessence, D. M. D’Alessandro, M. L. Foo and J. R. Long, J. Am. Chem. Soc., 2009, 131, 8784–8786 CrossRef CAS PubMed .
  49. J. E. Clements, J. R. Price, S. M. Neville and C. J. Kepert, Angew. Chem., Int. Ed., 2014, 53, 10164–10168 CrossRef CAS PubMed .
  50. N. F. Sciortino, S. M. Neville, J.-F. Létard, B. Moubaraki, K. S. Murray and C. J. Kepert, Inorg. Chem., 2014, 53, 7886–7893 CrossRef CAS PubMed .
  51. D. A. Reed, D. J. Xiao, M. I. Gonzalez, L. E. Darago, Z. R. Herm, F. Grandjean and J. R. Long, J. Am. Chem. Soc., 2016, 138, 5594–5602 CrossRef CAS PubMed .
  52. J.-P. Zhang, P.-Q. Liao, H.-L. Zhou, R.-B. Lin and X.-M. Chen, Chem. Soc. Rev., 2014, 43, 5789–5814 RSC .
  53. S. M. Hawxwell, G. Mínguez Espallargas, D. Bradshaw, M. J. Rosseinsky, T. J. Prior, A. J. Florence, J. van de Streek and L. Brammer, Chem. Commun., 2007, 1532–1534 RSC .
  54. M. Edgar, R. Mitchell, A. M. Z. Slawin, P. Lightfoot and P. A. Wright, Chem. – Eur. J., 2001, 7, 5168–5175 CrossRef CAS .
  55. J. I. Feldblyum, M. Liu, D. W. Gidley and A. J. Matzger, J. Am. Chem. Soc., 2011, 133, 18257–18263 CrossRef CAS PubMed .
  56. J. Seo, C. Bonneau, R. Matsuda, M. Takata and S. Kitagawa, J. Am. Chem. Soc., 2011, 133, 9005–9013 CrossRef CAS PubMed .
  57. P. Smart, C. A. Mason, J. R. Loader, A. J. H. M. Meijer, A. J. Florence, K. Shankland, A. J. Fletcher, S. P. Thompson, M. Brunelli, A. H. Hill and L. Brammer, Chem. – Eur. J., 2013, 19, 3552–3557 CrossRef CAS PubMed .
  58. 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 and M. J. Rosseinsky, J. Am. Chem. Soc., 2012, 134, 20466–20478 CrossRef CAS PubMed .
  59. I.-H. Park, S. S. Lee and J. J. Vittal, Chem. – Eur. J., 2013, 19, 2695–2702 CrossRef CAS PubMed .
  60. G. K. Kole and J. J. Vittal, Chem. Soc. Rev., 2013, 42, 1755–1775 RSC .
  61. B. J. Burnett and W. Choe, Dalton Trans., 2012, 41, 3889–3894 RSC .
  62. R. Kitaura, F. Iwahori, R. Matsuda, S. Kitagawa, Y. Kubota, M. Takata and T. C. Kobayashi, Inorg. Chem., 2004, 43, 6522–6524 CrossRef CAS PubMed .
  63. Z. Chen, S. Xiang, D. Zhao and B. Chen, Cryst. Growth Des., 2009, 9, 5293–5296 CAS .
  64. Y.-X. Tan, Y.-P. He and J. Zhang, Inorg. Chem., 2012, 51, 9649–9654 CrossRef CAS PubMed .
  65. J. Sun, Y. Zhou, Q. Fang, Z. Chen, L. Weng, G. Zhu, S. Qiu and D. Zhao, Inorg. Chem., 2006, 45, 8677–8684 CrossRef CAS PubMed .
  66. B. J. Burnett and W. Choe, CrystEngComm, 2012, 14, 6129–6131 RSC .
  67. J.-R. Li, D. J. Timmons and H.-C. Zhou, J. Am. Chem. Soc., 2009, 131, 6368–6369 CrossRef CAS PubMed .
  68. H.-N. Wang, X. Meng, G.-S. Yang, X.-L. Wang, K.-Z. Shao, Z.-M. Su and C.-G. Wang, Chem. Commun., 2011, 47, 7128–7130 RSC .
  69. I. J. Vitórica-Yrezábal, G. Mínguez Espallargas, J. Soleimannejad, A. J. Florence, A. J. Fletcher and L. Brammer, Chem. Sci., 2013, 4, 696–708 RSC .
  70. I. J. Vitórica-Yrezábal, S. Libri, J. R. Loader, G. Mínguez Espallargas, M. Hippler, A. J. Fletcher, S. P. Thompson, J. E. Warren, D. Musumeci, M. D. Ward and L. Brammer, Chem. – Eur. J., 2015, 21, 8799–8811 CrossRef PubMed .
  71. S. R. Batten and R. Robson, Angew. Chem., Int. Ed., 1998, 37, 1460–1494 CrossRef .
  72. S. B. Choi, H. Furukawa, H. J. Nam, D.-Y. Jung, Y. H. Jhon, A. Walton, D. Book, M. O’Keeffe, O. M. Yaghi and J. Kim, Angew. Chem., Int. Ed., 2012, 51, 8791–8795 CrossRef CAS PubMed .
  73. H. Aggarwal, P. M. Bhatt, C. X. Bezuidenhout and L. J. Barbour, J. Am. Chem. Soc., 2014, 136, 3776–3779 CrossRef CAS PubMed .
  74. H. Aggarwal, R. K. Das, P. M. Bhatt and L. J. Barbour, Chem. Sci., 2015, 6, 4986–4992 RSC .
  75. S. H. Lapidus, G. J. Halder, P. J. Chupas and K. W. Chapman, J. Am. Chem. Soc., 2013, 135, 7621–7628 CrossRef CAS PubMed .
  76. D. Lässig, J. Lincke, R. Gerhardt and H. Krautscheid, Inorg. Chem., 2012, 51, 6180–6189 CrossRef PubMed .
  77. X. Cui, A. N. Khlobystov, X. Chen, D. H. Marsh, A. J. Blake, W. Lewis, N. R. Champness, C. J. Roberts and M. Schröder, Chem. – Eur. J., 2009, 15, 8861–8873 CrossRef CAS PubMed .
  78. R. F. Mendes and F. A. Almeida Paz, Inorg. Chem. Front., 2015, 2, 495–509 RSC .
  79. W. Kaneko, M. Ohba and S. Kitagawa, J. Am. Chem. Soc., 2007, 129, 13706–13712 CrossRef CAS PubMed .
  80. E. Coronado, M. Giménez-Marqués, G. Mínguez Espallargas and L. Brammer, Nat. Commun., 2012, 3, 828 CrossRef PubMed .
  81. S. Furukawa, K. Hirai, Y. Takashima, K. Nakagawa, M. Kondo, T. Tsuruoka, O. Sakata and S. Kitagawa, Chem. Commun., 2009, 5097–5099 RSC .
  82. O. Shekhah, K. Hirai, H. Wang, H. Uehara, M. Kondo, S. Diring, D. Zacher, R. A. Fischer, O. Sakata, S. Kitagawa, S. Furukawa and C. Wöll, Dalton Trans., 2011, 40, 4954–4958 RSC .
  83. K. Hirai, J. Reboul, N. Morone, J. E. Heuser, S. Furukawa and S. Kitagawa, J. Am. Chem. Soc., 2014, 136, 14966–14973 CrossRef CAS PubMed .
  84. M. Meilikhov, S. Furukawa, K. Hirai, R. A. Fischer and S. Kitagawa, Angew. Chem., Int. Ed., 2013, 52, 341–345 CrossRef CAS PubMed .
  85. K. Hirai, S. Furukawa, M. Kondo, H. Uehara, O. Sakata and S. Kitagawa, Angew. Chem., Int. Ed., 2011, 50, 8057–8061 CrossRef CAS PubMed .
  86. T. D. Bennett and A. K. Cheetham, Acc. Chem. Res., 2014, 47, 1555–1562 CrossRef CAS PubMed .
  87. T. D. Bennett, D. A. Keen, J. C. Tan, E. R. Barney, A. L. Goodwin and A. K. Cheetham, Angew. Chem., Int. Ed., 2011, 50, 3067–3071 CrossRef CAS PubMed .
  88. T. D. Bennett, Y. Yue, P. Lim, A. Qiao, H. Tao, G. N. Greaves, T. Richards, G. I. Lampronti, S. A. T. Redfern, F. Blanc, O. K. Farha, J. T. Hupp, A. K. Cheetham and D. A. Keen, J. Am. Chem. Soc., 2016, 138, 3484–3492 CrossRef CAS PubMed .
  89. G. Férey, M. Haouas, T. Loiseau and F. Taulelle, Chem. Mater., 2014, 26, 299–309 CrossRef .
  90. L. Maserati, S. M. Meckler, C. Y. Li and B. A. Helms, Chem. Mater., 2016, 28, 1581–1588 CrossRef CAS .
  91. F. Ragon, H. Chevreau, T. Devic, C. Serre and P. Horcajada, Chem. – Eur. J., 2015, 21, 7135–7143 CrossRef CAS PubMed .
  92. S. Jakobsen, D. Gianolio, D. S. Wragg, M. H. Nilsen, H. Emerich, S. Bordiga, C. Lamberti, U. Olsbye, M. Tilset and K. P. Lillerud, Phys. Rev. B: Condens. Matter Mater. Phys., 2012, 86, 125429 CrossRef .
  93. G. C. Shearer, S. Chavan, S. Bordiga, S. Svelle, U. Olsbye and K. P. Lillerud, Chem. Mater., 2016, 28, 3749–3761 CrossRef CAS .
  94. J. Lippke, B. Brosent, T. von Zons, E. Virmani, S. Lilienthal, T. Preusse, M. Hulsmann, A. M. Schneider, S. Wuttke, P. Behrens and A. Godt, Inorg. Chem., 2017, 56, 748–761 CrossRef CAS PubMed .
  95. M. Taddei, D. Tiana, N. Casati, J. A. van Bokhoven, B. Smit and M. Ranocchiari, Phys. Chem. Chem. Phys., 2017, 19, 1551–1559 RSC .
  96. R. J. Marshall, C. L. Hobday, C. F. Murphie, S. L. Griffin, C. A. Morrison, S. A. Moggach and R. S. Forgan, J. Mater. Chem. A, 2016, 4, 6955–6963 CAS .
  97. B. Seoane, S. Castellanos, A. Dikhtiarenko, F. Kapteijn and J. Gascon, Coord. Chem. Rev., 2016, 307, 147–187 CrossRef CAS .
  98. G. Zahn, P. Zerner, J. Lippke, F. L. Kempf, S. Lilienthal, C. A. Schroder, A. M. Schneider and P. Behrens, CrystEngComm, 2014, 16, 9198–9207 RSC .
  99. J. D. Evans, C. J. Sumby and C. J. Doonan, Chem. Soc. Rev., 2014, 43, 5933–5951 RSC .
  100. Y. Han, J.-R. Li, Y. Xie and G. Guo, Chem. Soc. Rev., 2014, 43, 5952–5981 RSC .
  101. Y. Kim, S. Das, S. Bhattacharya, S. Hong, M. G. Kim, M. Yoon, S. Natarajan and K. Kim, Chem. – Eur. J., 2012, 18, 16642–16648 CrossRef CAS PubMed .
  102. J. Kahr, R. E. Morris and P. A. Wright, CrystEngComm, 2013, 15, 9779–9786 RSC .
  103. For already established approach to core-shell MOF synthesis, see: K. Hirai, S. Furukawa, M. Kondo, H. Uehara, O. Sakata and S. Kitagawa, Angew. Chem., Int. Ed., 2011, 50, 8057–8061 CrossRef CAS PubMed .
  104. C. K. Brozek and M. Dincă, Chem. Soc. Rev., 2014, 43, 5456–5467 RSC .
  105. C. K. Brozek and M. Dincă, Chem. Commun., 2015, 51, 11780–11782 RSC .
  106. L. Bellarosa, C. K. Brozek, M. Garcia-Melchor, M. Dincă and N. Lopez, Chem. Mater., 2015, 27, 3422–3429 CrossRef CAS .
  107. T. Islamoglu, S. Goswami, Z. Li, A. J. Howarth, O. K. Farha and J. T. Hupp, Acc. Chem. Res., 2017, 50, 805–813 CrossRef CAS PubMed .
  108. Y. Xu, A. J. Howarth, T. Islamoglu, C. T. da Silva, J. T. Hupp and O. K. Farha, Inorg. Chem. Commun., 2016, 67, 60–63 CrossRef CAS .
  109. M. B. Lalonde, J. E. Mondloch, P. Deria, A. A. Sarjeant, S. S. Al-Juaid, O. I. Osman, O. K. Farha and J. T. Hupp, Inorg. Chem., 2015, 54, 7142–7144 CrossRef CAS PubMed .
  110. S. T. Madrahimov, T. A. Atesin, O. Karagiaridi, A. A. Sarjeant, O. K. Farha, J. T. Hupp and S. T. Nguyen, Cryst. Growth Des., 2014, 14, 6320–6324 CAS .
  111. O. Karagiaridi, W. Bury, J. E. Mondloch, J. T. Hupp and O. K. Farha, Angew. Chem., Int. Ed., 2014, 53, 4530–4540 CrossRef CAS PubMed .
  112. C. Liu, C. J. Zeng, T. Y. Luo, A. D. Merg, R. C. Jin and N. L. Rosi, J. Am. Chem. Soc., 2016, 138, 12045–12048 CrossRef CAS PubMed .
  113. P. Jain, V. Ramachandran, R. J. Clark, H. D. Zhou, B. H. Toby, N. S. Dalal, H. W. Kroto and A. K. Cheetham, J. Am. Chem. Soc., 2009, 131, 13625–13627 CrossRef CAS PubMed .
  114. T. D. Bennett, A. H. Fuchs, A. K. Cheetham and F.-X. Coudert, Dalton Trans., 2016, 45, 4058–4059 RSC .
  115. A. K. Cheetham, T. D. Bennett, F.-X. Coudert and A. L. Goodwin, Dalton Trans., 2016, 45, 4113–4126 RSC .
  116. M. J. Cliffe, J. A. Hill, C. A. Murray, F.-X. Coudert and A. L. Goodwin, Phys. Chem. Chem. Phys., 2015, 17, 11586–11592 RSC .
  117. H. Wu, Y. S. Chua, V. Krungleviciute, M. Tyagi, P. Chen, T. Yildirim and W. Zhou, J. Am. Chem. Soc., 2013, 135, 10525–10532 CrossRef CAS PubMed .
  118. P. Ghosh, Y. J. Colon and R. Q. Snurr, Chem. Commun., 2014, 50, 11329–11331 RSC .
  119. C. Q. Zhang, Y. J. Zhao, Y. L. Li, X. T. Zhang, L. F. Chi and G. Lu, Chem. – Asian J., 2016, 11, 207–210 CrossRef CAS PubMed .
  120. M. J. Cliffe, W. Wan, X. D. Zou, P. A. Chater, A. K. Kleppe, M. G. Tucker, H. Wilhelm, N. P. Funnell, F.-X. Coudert and A. L. Goodwin, Nat. Commun., 2014, 5, 4176 CAS .
  121. G. C. Shearer, J. G. Vitillo, S. Bordiga, S. Svelle, U. Olsbye and K. P. Lillerud, Chem. Mater., 2016, 28, 7190–7193 CrossRef CAS .
  122. S. L. Ling and B. Slater, Chem. Sci., 2016, 7, 4706–4712 RSC .
  123. T. D. Bennett, T. K. Todorova, E. F. Baxter, D. G. Reid, C. Gervais, B. Bueken, B. Van de Voorde, D. De Vos, D. A. Keen and C. Mellot-Draznieks, Phys. Chem. Chem. Phys., 2016, 18, 2192–2201 RSC .
  124. M. Vandichel, J. Hajek, F. Vermoortele, M. Waroquier, D. E. De Vos and V. Van Speybroeck, CrystEngComm, 2015, 17, 395–406 RSC .
  125. W. J. Rieter, K. M. Pott, K. M. L. Taylor and W. Lin, J. Am. Chem. Soc., 2008, 130, 11584–11585 CrossRef CAS PubMed .
  126. J. Canivet, A. Fateeva, Y. Guo, B. Coasne and D. Farrusseng, Chem. Soc. Rev., 2014, 43, 5594–5617 RSC .
  127. N. C. Burtch, H. Jasuja and K. S. Walton, Chem. Rev., 2014, 114, 10575–10612 CrossRef CAS PubMed .
  128. N. ul Qadir, S. A. M. Said and H. M. Bahaidarah, Microporous Mesoporous Mater., 2015, 201, 61–90 CrossRef .
  129. J. J. Low, A. I. Benin, P. Jakubczak, J. F. Abrahamian, S. A. Faheem and R. R. Willis, J. Am. Chem. Soc., 2009, 131, 15834–15842 CrossRef CAS PubMed .
  130. S. S. Kaye, A. Dailly, O. M. Yaghi and J. R. Long, J. Am. Chem. Soc., 2007, 129, 14176–14177 CrossRef CAS PubMed .
  131. C. S. Slone, D. A. Weinberger and C. A. Mirkin, Prog. Inorg. Chem., 1999, 48, 233–350 CAS .
  132. A. Bader and E. Lindner, Coord. Chem. Rev., 1991, 108, 27–110 CrossRef CAS .
  133. I. Voda, C. Turta, V. Lozan, A. C. Benniston and U. Baisch, Polyhedron, 2014, 67, 301–305 CrossRef CAS .
  134. X. Zhu, S. Zhao, Y. F. Peng, B. L. Li and B. Wu, CrystEngComm, 2013, 15, 9154–9160 RSC .
  135. Z. S. Liu, E. Yang, Y. Kang and J. A. Zhang, Inorg. Chem. Commun., 2011, 14, 355–357 CrossRef CAS .
  136. D. S. Kim, P. M. Forster, R. Le Toquin and A. K. Cheetham, Chem. Commun., 2004, 2148–2149 RSC .
  137. P. M. Forster, J. Eckert, J. S. Chang, S. E. Park, G. Férey and A. K. Cheetham, J. Am. Chem. Soc., 2003, 125, 1309–1312 CrossRef CAS PubMed .
  138. B. Xiao, P. J. Byrne, P. S. Wheatley, D. S. Wragg, X. B. Zhao, A. J. Fletcher, K. M. Thomas, L. Peters, J. S. O. Evans, J. E. Warren, W. Z. Zhou and R. E. Morris, Nat. Chem., 2009, 1, 289–294 CrossRef CAS PubMed .
  139. P. K. Allan, B. Xiao, S. J. Teat, J. W. Knight and R. E. Morris, J. Am. Chem. Soc., 2010, 132, 3605–3611 CrossRef CAS PubMed .
  140. P. K. Allan, K. W. Chapman, P. J. Chupas, J. A. Hriljac, C. L. Renouf, T. C. A. Lucas and R. E. Morris, Chem. Sci., 2012, 3, 2559–2564 RSC .
  141. Y. Y. Tian, P. K. Allan, C. L. Renouf, X. He, L. J. McCormick and R. E. Morris, Dalton Trans., 2014, 43, 1519–1523 RSC .
  142. H. E. L. Mkami, M. I. H. Mohideen, C. Pal, A. McKinlay, O. Scheimann and R. E. Morris, Chem. Phys. Lett., 2012, 544, 17–21 CrossRef .
  143. M. I. H. Mohideen, B. Xiao, P. S. Wheatley, A. C. McKinlay, Y. Li, A. M. Z. Slawin, D. W. Aldous, N. F. Cessford, T. Duren, X. B. Zhao, R. Gill, K. M. Thomas, J. M. Griffin, S. E. Ashbrook and R. E. Morris, Nat. Chem., 2011, 3, 304–310 CrossRef CAS PubMed .
  144. D. M. Dawson, L. E. Jamieson, M. I. H. Mohideen, A. C. McKinlay, I. A. Smellie, R. Cadou, N. S. Keddie, R. E. Morris and S. E. Ashbrook, Phys. Chem. Chem. Phys., 2013, 15, 919–929 RSC .
  145. H. Sato, W. Kosaka, R. Matsuda, A. Hori, Y. Hijikata, R. V. Belosludov, S. Sakaki, M. Takata and S. Kitagawa, Science, 2014, 343, 167–170 CrossRef CAS PubMed .
  146. A. C. McKinlay, J. F. Eubank, S. Wuttke, B. Xiao, P. S. Wheadey, P. Bazin, J. C. Lavalley, M. Daturi, A. Vimont, G. De Weireld, P. Horcajada, C. Serre and R. E. Morris, Chem. Mater., 2013, 25, 1592–1599 CrossRef CAS .
  147. B. Xiao, P. S. Wheatley, X. B. Zhao, A. J. Fletcher, S. Fox, A. G. Rossi, I. L. Megson, S. Bordiga, L. Regli, K. M. Thomas and R. E. Morris, J. Am. Chem. Soc., 2007, 129, 1203–1209 CrossRef CAS PubMed .
  148. E. Coronado and G. Mínguez Espallargas, Chem. Soc. Rev., 2013, 42, 1525–1539 RSC .
  149. M. H. Zeng, Z. Yin, Y. X. Tan, W. X. Zhang, Y. P. He and M. Kurmoo, J. Am. Chem. Soc., 2014, 136, 4680–4688 CrossRef CAS PubMed .
  150. B. Pattengale, S. Z. Yang, J. Ludwig, Z. Q. Huang, X. Y. Zhang and J. Huang, J. Am. Chem. Soc., 2016, 138, 8072–8075 CrossRef CAS PubMed .
  151. K. T. Butler, C. H. Hendon and A. Walsh, J. Am. Chem. Soc., 2014, 136, 2703–2706 CrossRef CAS PubMed .
  152. H. J. Snaith, J. Phys. Chem. Lett., 2013, 4, 3623–3630 CrossRef CAS .
  153. Y. B. Tong, S. X. Liu, Y. Zou, C. Xue, H. B. Duan, J. L. Liu and X. M. Ren, Inorg. Chem., 2016, 55, 11716–11726 CrossRef CAS PubMed .

This journal is © The Royal Society of Chemistry 2017