Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2025 Jul 23;147(31):27586–27598. doi: 10.1021/jacs.5c05854

Ordering Bent and Straight Dicarboxylate Linkers in an fcu Zirconium Metal–Organic Framework

Grace S G Farmer †,‡, Daniel J Cheney †, Khai-Nghi Truong §, Nusik Gedikoglu †,‡, Bhupendra P Mali †, Datta Markad †, Dmytro Antypov †,‡, Frédéric Blanc †,‡,∥, Alexandros P Katsoulidis †, Matthew J Rosseinsky †,‡,*
PMCID: PMC12333367  PMID: 40699918

Abstract

The ordering of multiple organic linkers with different sizes and geometries on metal–organic frameworks (MOFs) offers structures with advanced complexity and functionality for enhanced properties. The selection of the components and the number of synthesis steps are commonly rationalized to match existing structural motifs in order to increase the probability of producing the designed new frameworks. Alternatively, combining multiple components, without a preconceived structural model, allows for the exploration of unprecedented structural characteristics. However, this approach requires a large number of experiments to identify both the composition and conditions for a successful synthesis. Here, we report a new Zr-MOF containing ordered straight and bent dicarboxylate linkers prepared by a one-pot synthesis. The discovery was made by high-throughput exploration of the chemical space composed of two linkers, the source of Zr, and the modulator. The linkers and the Zr6 cluster are ordered on an fcu framework that is tetragonally distorted from its typical cubic symmetry. The arrangement of linkers with different sizes and geometries affords cages and windows with shapes that have not been reported previously, despite the plethora of known fcu-based MOFs. The new MOF exhibits chromatographic separation of the n-hexane/benzene/cyclohexane mixture and demonstrates reversible unbinding and rebinding of the bent linkers upon the addition and removal of protic solvents. The unusual structural properties of the new MOF arise from ordering linkers in a non-predetermined manner following high-throughput synthesis exploration.

graphic file with name ja5c05854_0012.jpg

graphic file with name ja5c05854_0010.jpg

Introduction

Metal–organic frameworks (MOFs) consist of metal-based nodes bridged by organic linkers to form crystalline and porous frameworks. Zirconium MOFs (Zr-MOFs) are well-known for their high structural and thermal stability, with a large variety of reported structures and applications. , UiO-66 is the most well-known Zr-MOF, comprising [Zr6(μ3-O)4(μ3–OH)4]12+ clusters and 1,4-benzene dicarboxylate (BDC) linkers. , Each Zr6 cluster is surrounded by 12 BDC linkers connected to further Zr6 clusters, forming the high-symmetry, 12-connected (12-c) framework with the fcu topology. The use of ditopic linkers with differing geometry and lower symmetry compared to BDC provides access to a range of frameworks with different symmetries and connectivities. The available ditopic linkers typically exhibit straight, zigzag, and bent geometries. , Zigzag linkers, such as fumaric acid and 2,6-naphthalenedicarboxylic acid (NDC), can form the fcu framework topology, , as well as 8-c frameworks with the bcu topology. However, the framework connectivity for bent linkers is considerably more complex. Several fcu Zr-MOFs, such as Zr-UiO-66-PZDC MOF, and BUT-10, with bent linkers with an angle of 160° between carboxylates in the framework have been reported. Within these structures, the Zr6 clusters rotate around multiple axes to accommodate the bent linker shape in three dimensions. However, bent linkers with a more acute angle between carboxylates result in further rotation of the clusters, which reduces the favorability of a 12-c framework. 2,5-Thiophenedicarboxylic acid (TDC) is an extensively explored bent linker in Zr-MOF chemistry. The Kaskel group has discovered four polymorphs of single-linker Zr-TDC MOFs, highlighting the different ways in which TDC can be arranged around the Zr6 cluster. In contrast, other known bent linkers, such as isophthalic acid and 2,5-furandicarboxylic acid, have not demonstrated a comparable number of reported structures or the same level of structural diversity as TDC in Zr-MOF chemistry. Each Zr-TDC polymorph displays a distinct topology, with three of them being 8-connected frameworks (DUT-67, -68, and -126) with reo, bon, and hbr topologies, respectively, while one is a 10-c Zr-MOF (DUT-69) with a bct topology. , An fcu zirconium framework with TDC linkers has not been reported, and this can be understood as a mismatch between linker geometry and framework topology.

Multicomponent MOFs , display ordered arrangements of multiple linkers and differ from multivariate MOFs, which have the linkers disordered over the framework. The ordering of multiple linkers can produce new cages and windows that are not achievable with a single linker, resulting in enhanced performance in applications such as adsorption and separation. , Multicomponent Zr-MOFs are of interest due to their high stability and structural diversity, but they are predominantly synthesized by multistep procedures. , Therefore, their structures are restricted to those that can be derived by functionalizing the structure formed in the first synthetic step. The direct combination of multiple linkers in one synthetic step can deliver new structural features due to the assembly of linkers in a non-predetermined manner; however, this approach requires screening of many possible combinations of the composition parameters. A high-throughput synthesis (HTS) workflow facilitates the systematic exploration of large chemical spaces defined by variables such as the ratio of two or more linkers, the ratio of the linkers to the metal source, and the amount of modulator. HTS implementation in materials synthesis is growing, with notable examples in the porous materials community for zeolites, organic porous cages, covalent organic frameworks (COFs), and MOFs. Our HTS workflow, involving robotic liquid dispensing, was previously utilized to discover the first multicomponent Zr-MOF with equal amounts of straight (BDC) and zigzag (fumarate) linkers via a one-step synthesis. This framework is described by the fcu topology and rhombohedral symmetry due to the ordered decoration of the net by the two linkers of different lengths and geometries.

In this work, we implement an HTS workflow to explore efficiently the combination of bent TDC and straight BDC ditopic linkers within Zr-MOF chemistry. The goal is to target the assembly of Zr-MOFs with distinct structural features arising from component linkers that individually afford Zr-MOFs with different topologies. We present a new fcu Zr-MOF containing both BDC and TDC in tetragonal symmetry, discovered in a facile one-step synthesis using the ZrOCl2.8H2O/BDC/TDC/formic acid (FA)/dimethylformamide (DMF) system. TDC forms part of a 12-connected framework in this chemistry for the first time. The effect of geometry mismatch between the bent TDC geometry and the fcu topology is illustrated by the behavior of the framework after guest exchange with methanol, wherein the TDC linker reversibly disconnects from the framework in a structural transition. The distinct cage geometry of this ordered two-linker fcu structure enables separation of small molecules in gas chromatography.

Results and Discussion

The combination of BDC and TDC was explored with formic acid (FA) as the modulator and DMF as the solvent. This ZrOCl2·8H2O/BDC/TDC/FA/DMF system was selected from the available literature data for the synthesis of existing single-linker MOFs. UiO-66 can be formed under many conditions using different solvents, modulators, reaction times, and reaction temperatures. , DUT-67, the most explored TDC Zr-MOF, has been reported under conditions that overlap with those used for UiO-66. , The solvent, DMF, and modulator, FA, successfully synthesized both MOFs and were chosen for the two-linker exploration. The explored chemical space spans a broad range of compositions, covered by a set of 60 points. Five molar ratios of the two linkers were explored, corresponding to BDC:TDC = 0.25:0.75, 0.33:0.67, 0.50:0.50, 0.67:0.33, and 0.75:0.25. Four molar ratios of ZrOCl2·8H2O to linker, Zr:(BDC + TDC), were investigated: 0.67:0.33, 0.50:0.50, 0.40:0.60, and 0.33:0.67. Three molar ratios of formic acid to ZrOCl2·8H2O (FA:Zr) were explored: 120, 310, and 500 (Figure a). The amount of modulator is of particular importance for exploratory Zr-MOF synthesis, as varying modulator amounts can result in UiO-66 formation with different defect densities or induce the formation of different phases, as reported for the TDC Zr-MOF polymorphs. Finally, the total liquid amount in the synthesis vial was fixed to 10 mL, and the reaction time and temperature were fixed to 48 h and 120 °C, respectively.

1.

1

Open in a new tab

(a) 3D plot depicting the chemical space explored for the first batch of 60 reactions in the ZrOCl2·8H2O/BDC/TDC/FA/DMF system. The colored points (green, blue, black, purple, and red) indicate that sufficient solids are produced for their PXRD patterns to be collected (Cu Kα1, λ = 1.5406 Å). The dark gray points yielded only a small amount of the solid product; thus, no PXRD measurements were performed. Light gray points indicate that no solid products are produced. The reaction compositions that produced the pure new phase are circled. (b–e) Sets of low-angle PXRD patterns from the points corresponding to the colors used in (a). The remainder of the patterns is provided in Figure S1. The patterns corresponding to the single-linker phases (DUT-67 and UiO-66) are noted in (b). Stars in (b–d) indicate patterns of pure new phase produced from the green, blue, and purple circled points, respectively, in (a). The PXRD pattern corresponding to the black circled point is provided in Figure S1c.

Visual inspection of the vials after the reaction showed that 55 out of the 60 batch samples had produced solid products, and the remaining five reaction mixtures produced clear solutions. Of the 55 samples with solid products, 48 produced enough solid for powder X-ray diffraction (PXRD) measurements. PXRD measurements showed that the majority of these yielded crystalline materials (Figures b–e and S1). The set of five PXRD patterns in Figure b is prepared from reaction mixtures that have a Zr:(BDC + TDC) ratio of 0.33:0.67. The remaining three sets (Figure c–e) are prepared from reaction mixtures that have the same Zr:(BDC + TDC) ratio of 0.40:0.60. Each set has a different FA:Zr ratio: 500 (Figure b,c), 310 (Figure d), and 120 (Figure e). Within each set, the linker ratio varies from BDC:TDC = 0.25:0.75 to BDC:TDC = 0.75:0.25. The formation of the known single-linker MOF phases UiO-66 and DUT-67 is observed in each set at high and low BDC:TDC ratios, respectively, and the appearance of a new phase is evident at intermediate ratios. As shown in Figure b, a cubic phase is present at BDC:TDC = 0.75:0.25, which corresponds to the UiO-66 structure. As BDC:TDC decreases to 0.67:0.33, a phase with low-intensity peaks emerges alongside the UiO-66-like phase, and at BDC:TDC = 0.50:0.50, the unknown phase appears as the only crystalline phase in the sample. These new peaks do not correspond to any of the known phases derived from the components of this system, indicating the discovery of a new phase. A further increase in the amount of TDC results in the appearance of low intensity peaks in the BDC:TDC = 0.33:0.67 pattern, which are identified as arising from a phase with the DUT-67 structure. At BDC:TDC = 0.25:0.75, the new phase disappears, and only the DUT-67 phase is present. In the panels of Figure c,d, the pure new phase is also observed at the two points where BDC:TDC = 0.50:0.50. Comparing the relative intensities of the peaks of the new phase to those of DUT-67 and UiO-66, with the same BDC:TDC ratio for these two sets, it is observed that the phase composition of the solid product differs depending on the Zr:(BDC + TDC) and FA:Zr ratios. For example, ratio BDC:TDC = 0.33:0.67 in Figure c shows more intense peaks of DUT-67 compared to the new phase, whereas in Figure d and for the same BDC:TDC ratio, the peaks for the new phase are more intense compared to those for DUT-67. In the set with the lowest FA:Zr = 120 ratio, the new phase is observed at BDC:TDC = 0.33:0.67, but the peaks are broad, and it is difficult to conclude about the phase purity of this sample. All of the samples produced at a FA:Zr ratio of 120 display relatively broad peaks compared to the higher ratios, which demonstrates the effect of the modulator amount in the present system to produce crystalline materials with sharp peaks that help distinguish those samples that correspond to phase-pure products.

The new phase appears broadly across the chemical space in a mixture with phases of the known single-linker MOF structures in 23 out of the 60 compositions explored and as a pure phase product in 4 out of 60 compositions. In addition to the three compositions delivering the pure phase shown in Figure b–d, the new phase was also identified as pure at BDC:TDC = 0.50:0.50, Zr:(BDC + TDC) = 0.67:0.33, and FA:Zr = 310 (Figure S1c).

A second iteration of 42 reactions in the same system was performed to understand the limits of the region of chemical space that produces the new phase pure material (Figure a). The composition points were selected to cover the space around those that yielded the new phase pure in the first batch, excluding points that yielded the known phases or lower-crystallinity products (Tables S3 and S4). Solid products were produced for all 42 reaction mixtures of the second batch, with 37 of 42 points identified as the pure target phase via PXRD (Figure S2). The four points replicated from batch 1 provided patterns identical to those of the prior reactions (Figure S3). The remaining five points contain low intensity peaks attributed to the DUT-67-like phase in addition to the new phase. 1H nuclear magnetic resonance (NMR) analysis on the digested solids of phase-pure samples demonstrates the presence of both linkers, with variance in their ratio (BDC:TDC from 0.64:0.36 to 0.53:0.47) depending on the reaction mixture compositions (Table S5, Figures S4 and S16). The HTS exploration demonstrated that a new phase can be formed over a large chemical space. All of the hits from the first batch were at equimolar BDC:TDC ratios (0.50:0.50); however, the exploration within the narrower space of batch 2 demonstrated that pure phase is produced at reaction mixture linker ratios varying from BDC:TDC = 0.43:0.57 to BDC:TDC = 0.57:0.43. Interestingly, the new phase is synthesized over a broader range of reaction compositions compared to the previously reported two-linker ordered MOF, Zr6(BDC)3(Fum)3, which is formed over a restricted and narrow region of the ZrOCl2·8H2O/BDC/fumarate/DMF/FA system.

2.

2

Open in a new tab

(a) 3D plot to represent the chemical space explored in the second batch by 42 reactions. The area bound by the gray trapezoidal prism represents the chemical space explored in the first batch of reactions, and the area outlined in blue represents the space explored in the second batch. The green points represent the reactions that yielded the pure new phase, and the red points represent those reactions that yielded the new phase alongside a small impurity of the DUT-67 phase. The PXRD patterns of the second batch are provided in Figure S2. (b) Rietveld refinement of Zr6(BDC)4(TDC)2-DMF from the PXRD pattern of the solid obtained from the point highlighted by a square in (a) with the following composition: BDC:TDC = 0.57:0.43, Zr:(BDC+TDC) = 0.67:0.33, and FA:Zr = 440.

Structural Description

The new phase, Zr6(BDC)4(TDC)2-DMF, crystallizes in tetragonal space group I41/acd with unit cell parameters a = 19.88173(5) Å and c = 42.49860(13) Å and with a unit cell volume V = 16798.99(10) Å3 (Figure a). The size of the crystallites ranges from 500 to 600 nm (Figure S5), which is too small for single-crystal X-ray diffraction measurements. Therefore, the crystal structure was solved using synchrotron PXRD data. Simulated annealing was implemented to construct the structural model (Figure S6), and the refinement was carried out with the Rietveld method (see Structure Determination in the Supporting Information) on the data collected for the as-made sample with formula Zr6O4(OH)4(BDC)3.56(TDC)1.96(HCOO)0.96·6.99DMF (Figure b). This sample, synthesized from the reaction mixture with BDC:TDC = 0.57:0.43, Zr:(BDC + TDC) = 0.67:0.33, and FA:Zr = 440, exhibits high crystallinity, and its formula close approximates the theoretical formula of the framework Zr6O4(OH)4(BDC)4(TDC)2. The center of the [Zr6O4(OH)4]12+ clusters corresponds to Wyckoff positions 8a of the I41/acd space group. The clusters adopt the fcu net with 12 neighbors, which is most clearly described in terms of their distribution over successive 00l planes (Figure a). Each cluster has four adjacent coplanar clusters at a distance of 14.06 Å and to eight other clustersfour located on the plane above and four on the plane below, at 14.55 Å (Figure S6). These distances are similar to the cluster–cluster distances in DUT-67 (13.83 Å) and UiO-66 (14.66 Å), suggesting the location of four TDC and eight BDC linkers around each [Zr6O4(OH)4]12+ (Figure b). The linkers were introduced at these positions and refined to sensible geometries, with their occupancies fixed to values obtained from 1H NMR analysis of the digested sample, in combination with thermogravimetric analysis (TGA) results (see the Chemical Formula of Bulk Samples in the Supporting Information). The structural model of Zr6(BDC)4(TDC)2-DMF is described as an fcu net featuring an ordered arrangement of eight straight (BDC) and four bent (TDC) linkers (Figure a). The reduced tetragonal I41/acd space group symmetry of Zr6(BDC)4(TDC)2-DMF compared to cubic Fm3̅m of single-linker UiO-66 arises from the ordering of linkers that differ in size and geometry. The difference in linker length induces tetragonal distortion, resulting in the doubling of the c-axis (Figure a) to accommodate the bent TDC around Zr6 clusters in two orientations: one with the sulfur of the thiophene rings pointing clockwise and the other pointing anticlockwise (Figure b–e). Thus, through the 8 + 4 ordering with BDC, TDC gains access to the high-symmetry fcu net, which is inaccessible in single-linker TDC Zr-MOF analogues, like DUT-67. The other known fcu Zr-MOF with 8 + 4 linkers, ZRN-XB, was accessed by the postsynthetic linker installation of X-BDC (X = OH, −NO2, and −NH2) into the 8-c ZRN-bcu framework described by eight NDC zigzag linkers. The addition of the straight BDC linker in that case does not reduce further the I4/mmm symmetry of the starting 8-c framework because the point group symmetry of BDC (mmm) matches the symmetry of the site where the BDC is inserted. Zr6(BDC)4(TDC)2-DMF is the first reported Zr-MOF with the I41/acd space group and the first fcu Zr-MOF containing ordered straight and bent dicarboxylate linkers.

3.

3

Open in a new tab

(a) Tetragonal unit cell of the Zr6(BDC)4(TDC)2-DMF crystal structure. [Zr6O4(OH)4]12+ clusters are represented by teal polyhedra, with Zr atoms coordinated by O atoms in red. BDC linkers are shown in blue, and TDC linkers are shown in orange. Yellow and purple spheres represent the guest-accessible tetrahedral and octahedral cages, respectively. (b) Zr6 octahedron surrounded by eight BDC and four TDC linkers. (c) Zr6 octahedron coordination by the carboxylate groups of the BDC and TDC linkers. The crystallographically distinct Zr atoms are labeled as Zr1 and Zr2. H atoms and solvent molecules are omitted for clarity from all depictions, as are the hydroxides and oxides on the [Zr6O4(OH)4]12+ cluster.

4.

4

Open in a new tab

(a) Stacking of layers of Zr6 octahedra and TDC in the unit cell of Zr6(BDC)4(TDC)2-DMF corresponding to the doubling of the c-axis compared to the UiO-66 crystal structure (Figure S13). (b–e) Four Zr6(TDC)2 layers stacked from top to bottom in the unit cell of Zr6(BDC)4(TDC)2-DMF. Comparison of panel (b) with (d) and panel (c) with (e) revealed that TDC adopts two orientations, with the sulfur atoms of the thiophene ring pointing in opposite directions, connecting Zr6 octahedra that rotate in opposite senses. The sulfur atoms of the TDC linker are highlighted in yellow to emphasize the directionality of the thiophene ring. Zr is shown in teal, and O is shown in red; the C atoms of the TDC linkers are shown in orange, and the S atoms of TDC are shown in yellow. The hydroxides and oxides on the Zr6 cluster, the BDC linkers, H atoms, and solvent molecules are omitted for clarity.

The six Zr atoms of the [Zr6O4(OH)4]12+ cluster in Zr6(BDC)4(TDC)2-DMF form a slightly compressed octahedron, Zr6, along the c-axis resulting from the tetragonal distortion (Figure S12). There are two independent crystallographic Zr sites: Zr1 occupying the two axial sites and Zr2 occupying the four planar sites (Figure c). The Zr1–Zr2 edge is occupied by the carboxylates of BDC, and the Zr2–Zr2 edge is occupied by the carboxylates of TDC.

Each TDC linker coordinates solely to Zr2 and connects two Zr6 clusters, which are rotated in opposite senses by 12° about the c-axis to accommodate the bent TDC geometry (Figure b–e). The assembly of TDC linkers and Zr6 clusters forms Zr6(TDC)2 layers parallel to the ab plane, composed of rectangular rings consisting of four bent TDC and four Zr6 clusters (Figure a); these rings are elongated compared to the regular squares observed in UiO-66 (Figure S13). The long axes of the rectangles alternate in their orientation between edge-sharing rings within the plane. The structural motif with coplanar TDC linkers and Zr6 clusters, produced by rotation around only one axis, is not observed in DUT-67 or any other single-linker Zr-TDC MOFs. DUT-67 contains a Zr6(TDC)2 layer with two kinds of squares, where the S of TDC points above or below the layer (Figure S14). However, in the single-linker Zr-TDC MOF structures, the Zr6 clusters rotate around multiple axes, bringing the TDC linkers out of plane in order to accommodate their bent shape in three dimensions, yielding nets with connectivity lower than 12.

5.

5

Open in a new tab

(a) Zr6(TDC)2 layer of rectangular rings formed by Zr6 octahedra and TDC. The rings are arranged in alternating orientations. (b) Octahedral cage of Zr6(BDC)4(TDC)2-DMF showing the clusters of a Zr6(TDC)2 layer connected to the above and below layers by two types of BDC linkers. (c) and (d) Zr6 octahedra connected by BDC linkers in two ways. (c) BDC on site 1 connects two clusters that rotate in the same sense and intersects the plane defined by Zr1 atoms (dashed line). The rotation of the clusters around the c-axis is highlighted with arrows. (d) BDC on site 2 connects two clusters that rotate in opposite senses and is displaced from the plane defined by the Zr1 atoms (dashed line). (e) Unit cell of Zr6(BDC)4(TDC)2-DMF highlighting the positions of BDC sites alternating down the unit cell, emphasized with a dashed line as shown in (c) and (d). Zr is shown in teal, O is shown in red, the C atoms of BDC site 1 are in blue, the C atoms of BDC site 2 are in cyan, and TDC is shown in orange. The hydroxides and oxides on the Zr6 cluster, H atoms, and solvent molecules are omitted for clarity.

Each Zr6(TDC)2 layer is connected to the neighboring layers of clusters by BDC linkers (Figure b). The pattern of alternating rotation of the clusters creates two crystallographic sites for BDC. Reflecting the effect of differing rotational pairs on the locations of Zr2 atoms in successive layers, BDC site 1 connects two Zr6 clusters that rotate in the same sense (Figure c) and thus is inclined such that it intersects the plane defined by their Zr1 sites. In contrast, BDC site 2 connects oppositely rotated clusters and therefore is displaced from this plane (Figure d). Both BDC sites are present in each Zr6(BDC)2 layer, which lie on the ac and bc planes, in alternate arrangement along the c-axis (Figure e). The BDC linkers in each layer are not coplanar, thus differing from the layers in UiO-66, which have a single BDC site connecting coplanar Zr atoms. The synergistic effect of the bent TDC and straight BDC linkers can accommodate the rotation of the Zr6 cluster about one axis and hence form the Zr6(BDC)4(TDC)2-DMF framework in the high-symmetry 12-c fcu net (Figure e).

Zr6(BDC)4(TDC)2-DMF was treated with methanol to replace DMF from the pores and produce a material, Zr6(BDC)4(TDC)2-MeOH, with a volatile solvent in the pores that could be activated at moderate temperatures. The completion of guest exchange was confirmed by 1H NMR (Figure S16), and the PXRD pattern of Zr6(BDC)4(TDC)2-MeOH presents the same diffraction peaks as Zr6(BDC)4(TDC)2-DMF at low angles; however, there are noticeable differences at higher angles (Figure a). Zr6(BDC)4(TDC)2-MeOH has the same space group as Zr6(BDC)4(TDC)2-DMF, I41/acd, with similar unit cell parameters (a = 19.91546(12) Å, c = 42.6593(4) Å). The main structural difference between Zr6(BDC)4(TDC)2-MeOH and Zr6(BDC)4(TDC)2-DMF is the location and orientation of the bent TDC linker, which is no longer connected to the Zr6 clusters through Zr–O bonds (see Structure Determination in the Supporting Information). Instead, TDC is located in the space between the ab planes of Zr6 clusters (Figure a) with the carboxylate groups oriented toward Zr6 clusters that are also linked by the BDC of site 2 (Figure b). The Zr6 clusters and BDC linkers retain their sites and form an 8-c net with a bcu topology. The rotation of the Zr6 clusters has been reduced to 7°, from 12° in Zr6(BDC)4(TDC)2-DMF, resulting in BDC site 1 being less inclined and BDC site 2 being less displaced compared to Zr6(BDC)4(TDC)2-DMF. However, the displaced position of BDC site 2 creates space for the TDC linker to approach and form H-bonds with the cluster. The sites previously occupied by TDC carboxylates in the ab plane are filled by capping MeOH molecules. The TDC linker is stabilized by six hydrogen bonds: two with the μ3–OH groups of the Zr6 clusters and four with the capping MeOH molecules (Figure c). Its position is further stabilized by a favorable interaction between the TDC and BDC site 2 linker. The aromatic ring of the TDC is perpendicular to the aromatic ring of BDC site 2, with a distance of 4.718(3) Å between the thiophene ring and the benzene ring (Figure d). , The structure of Zr6(BDC)4(TDC)2-MeOH demonstrates that methanol can disconnect and replace TDC from the Zr6 cluster, while TDC remains bound to the framework through hydrogen bonding (Figure e). This activity of methanol is similar to its role in postsynthetic linker exchange in UiO-66, where it facilitates the breaking of metal–linker Zr–O bonds and promotes the formation and stabilization of dangling linkers.

7.

7

Open in a new tab

(a) Synchrotron PXRD patterns and (b) 13C CP MAS NMR spectra (shown for ranges 180–130 ppm and 100–25 ppm) of (i) Zr6(BDC)4(TDC)2-DMF, (ii) Zr6(BDC)4(TDC)2-MeOH, (iii) Zr6(BDC)4(TDC)2-H2O, and (iv) activated Zr6(BDC)4(TDC)2. NMR signal assignments are indicated by colored numbers correlating to the carbon environments of BDC (blue) and TDC (orange) linkers. In spectra ii of Zr6(BDC)4(TDC)2-MeOH, the broad signals at 171 and 130.5 ppm are attributed to carbon environments in both BDC and TDC and are labeled with two colored numbers. Signals attributed to solvents are indicated on the spectra, and spinning sidebands are marked with asterisks (*).

6.

6

Open in a new tab

(a) In the Zr6(BDC)4(TDC)2-MeOH crystal structure, TDC is located between the layers of Zr6 octahedra (BDC is omitted). (b) Unit cell of the Zr6(BDC)4(TDC)2-MeOH crystal structure. (c) TDC linker is connected through hydrogen bonds to hydroxides and coordinated methanol molecules on two Zr6 clusters that are linked at BDC site 2. (d) Position of the TDC relative to BDC site 1 and BDC site 2. A gray dashed line is included to show the interaction between the TDC linker and the center of BDC site 2. This is further highlighted with light gray dashed lines within the aromatic rings. (e) Zr6 octahedron surrounded by eight BDC linkers and four TDC linkers in Zr6(BDC)4(TDC)2-MeOH. This can be compared to the analogous representation of Zr6(BDC)4(TDC)2-DMF (Figure b). For (c, d), only the methanol molecules involved in the TDC H-bonding are shown. For all depictions, the Zr6 nodes are shown as teal octahedra with the oxides and hydroxides included to show their interaction with the TDC. The methanol molecule coordinated to the Zr6 clusters is shown. The C atoms of BDC site 1 are in blue, the C atoms of BDC site 2 are in cyan, and the TDC linkers have orange C and S atoms. The oxygen atoms are shown in red. The methyl carbon of methanol is shown in gray, and all hydrogen atoms are shown in light gray. The red dotted lines represent hydrogen bonds between the TDC, the solvent coordinated to the cluster, and the cluster hydroxides. The solvent in the pores and the H atoms of both the linkers’ carbon atoms are omitted for clarity.

When Zr6(BDC)4(TDC)2-MeOH is exposed to air overnight, the methanol is exchanged for atmospheric water. This exchange produces Zr6(BDC)4(TDC)2-H2O and is confirmed by the disappearance of methanol signals in digestion 1H NMR over a 24 h time period (see Solvent exchange and activation in the Supporting Information, Table S10 and Figure S20). This is also supported by the disappearance of methanol signals in 1H magic angle spinning (MAS) NMR (Figure S22ii,iii) at 2.9 and 4.7 ppm and the emergence of a signal at 4.2 ppm (which is assigned to H2O bound to the Zr6 cluster). The PXRD pattern of Zr6(BDC)4(TDC)2-H2O is very similar to that of Zr6(BDC)4(TDC)2-MeOH (Figure a.ii and iii, respectively) and is indexed in the same space group, I41/acd, with unit cell parameters a = 19.8623(1) Å and c = 42.7463(3) Å. The structure of Zr6(BDC)4(TDC)2-H2O was elucidated by continuous rotation three-dimensional electron diffraction (3D ED) measurements, , which allowed the study of submicron-sized crystals; the results demonstrated that TDC occupies a similar position to that in Zr6(BDC)4(TDC)2-MeOH, with H2O molecules serving as capping ligands on the Zr6 clusters (Figure S15).

The observation of 8-c net Zr6(BDC)4 in Zr6(BDC)4(TDC)2-MeOH and Zr6(BDC)4(TDC)2-H2O structures prompts the investigation of any intermediate phase during the reaction synthesis of Zr6(BDC)4(TDC)2-DMF. A set of synthesis experiments at shorter reaction times (see Shorter reaction synthesis of Zr6(BDC)4(TDC)2-DMF in the Supporting Information) demonstrated that a small amount of UiO-66 is formed at the initial stage of the reaction but gradually disappears as Zr6(BDC)4(TDC)2-DMF forms and becomes the only phase present in the reaction mixture (Table S11 and Figure S21).

The structural transformation from Zr6(BDC)4(TDC)2-DMF to Zr6(BDC)4(TDC)2-MeOH and Zr6(BDC)4(TDC)2-H2O was corroborated by MAS NMR measurements. The 13C cross-polarization (CP) MAS spectra of Zr6(BDC)4(TDC)2-DMF are depicted in Figure b.i. Spectral assignment of the BDC and TDC signals was aided through 1H MAS NMR spectra (Figure S22i and Table S12) and two-dimensional (2D) 1H–13C HETeronuclear CORrelation (HETCOR) NMR spectra (Figure S23), with the chemical shifts being in good agreement with those previously reported for UiO-66 and DUT-67. , There are also signals assigned to DMF and tetrahydrofuran (THF) (see Solvent exchange and activation in the Supporting Information). The 13C CP MAS spectrum of Zr6(BDC)4(TDC)2-MeOH (Figure b.ii) exhibits no signals arising from DMF, which confirms the removal of the DMF molecules from the pores. Three new signals appear between 50 and 55 ppm, which are associated with the methyl group of MeOH. This indicates that MeOH adopts three different environments in this structure: one as an unbound mobile guest in the pores at 50 ppm (as confirmed by the significantly increased intensity of this peak in the directly excited 13C MAS NMR spectrum versus the CP spectrum, Figure S24) and two as capping ligands on the Zr6 cluster at 52 and 54 ppm. In the HETCOR spectrum (Figure S25), these signals correlate with the aromatic 1H signals, indicating spatial proximity. Further correlations are observed between the carboxylate 13C signals and a 1H signal at 5.7 ppm (which likely corresponds to the OH protons of capping methanol), illustrating that methanol binds to the Zr6 cluster, in agreement with the refined structure of Zr6(BDC)4(TDC)2-MeOH. The peaks of BDC and TDC carbons are broad due to the presence of MeOH in the pores. In Zr6(BDC)4(TDC)2-H2O, the three MeOH 13C signals disappeared, confirming that this solvent has largely evaporated (Figure b.iii). The HETCOR spectrum (Figure S26) shows correlations between the linker 13C signals and a 1H signal at 5.4 ppm, which likely corresponds to H2O bound to the Zr6 cluster, confirming that water has replaced methanol as the capping ligand and is now in close proximity to the linkers. The lines are also significantly narrower than those for Zr6(BDC)4(TDC)2-MeOH, and signals arising from BDC and TDC can be distinguished in the HETCOR spectrum. The carboxylate signal of TDC has been shifted to 171 ppm compared to 167 ppm for Zr6(BDC)4(TDC)2-DMF. This shift indicates an environment change for the TDC carboxylate group in both Zr6(BDC)4(TDC)2-MeOH and Zr6(BDC)4(TDC)2-H2O, which is consistent with the change of the TDC position in the refined crystal structures. In both Zr6(BDC)4(TDC)2-DMF and Zr6(BDC)4(TDC)2-H2O, there are two signals attributed to the BDC carboxylate carbons due to the two crystallographic sites for BDC within the framework, leading to slightly different metal–Ocarboxylate bonding to the Zr6 cluster. There is one 13C MAS NMR signal for the quaternary aromatic carbons (at 138 ppm), representing both BDC sites for Zr6(BDC)4(TDC)2-DMF; however, in Zr6(BDC)4(TDC)2-H2O, these BDC sites are now clearly resolved into two signals. The differences between the BDC sites are increased because BDC site 2 is now interacting with the thiophene ring of the relocated TDC linker, whereas BDC site 1 is not (Figure d).

H2O was removed from Zr6(BDC)4(TDC)2-H2O through activation at 150 °C and 10–3 mbar for 16 h, as confirmed by the absence of the H2O signal in the 1H MAS NMR spectrum (Figure S22.iv), where a signal at 2.2 ppm arising from the μ3–OH sites is now clearly visible. The PXRD pattern of the activated sample, Zr6(BDC)4(TDC)2, confirms that the framework retains its crystallinity and closely resembles the pattern of Zr6(BDC)4(TDC)2-DMF (Figure a.iv). Rietveld refinement of this pattern with the two structural models described above showed that Zr6(BDC)4(TDC)2 has the same arrangement of linkers as Zr6(BDC)4(TDC)2-DMF, with TDC linking clusters through Zr–O coordination bonds (Figure S11). The movement of TDC from the site occupied in Zr6(BDC)4(TDC)2-H2O back to the site of the as-made Zr6(BDC)4(TDC)2-DMF is confirmed by the 13C CP MAS spectra of Zr6(BDC)4(TDC)2, which shows that the signal for the TDC carboxylate C shifts from 171 to 168 ppm, as determined by spectral assignments aided by HETCOR (Figure S27). This movement is triggered by the removal of H2O capping ligands from the Zr6 cluster of Zr6(BDC)4(TDC)2-H2O during activation, which breaks the hydrogen bonds that stabilized the TDC and creates uncoordinated sites on the Zr6 clusters that attract the carboxylates of TDC. The rotation of Zr6 clusters in Zr6(BDC)4(TDC)2 increases to 10° from 7° in Zr6(BDC)4(TDC)2-H2O to accommodate the bent shape of TDC. Thus, the Zr6(TDC)2 layers are reformed, and Zr6(BDC)4(TDC)2 is described as a 12-c fcu net. The reintroduction of atmospheric water to the activated material again disconnects the TDC from the cluster, affording Zr6(BDC)4(TDC)2-H2O (Figure S28 for the variable-temperature PXRD data). Here, water acts in a manner similar to methanol, preferentially binding to the Zr6 clusters and securing the TDC linker in the framework with hydrogen bonding. Similarly to methanol, water has been reported for postsynthetic linker exchange and linker removal. , This reversible structural transformation demonstrates that the rotation of the Zr6 clusters, emerging from the direct combination with the two linkers in one-pot synthesis, can act as a structural degree of freedom triggered by the chemical control of the TDC position.

The N2 adsorption–desorption isotherm of Zr6(BDC)4(TDC)2 exhibits a type I profile, which is typical for microporous materials, with a BET surface area of 1054 m2/g and a pore volume of 0.41 cm3/g (Figures a and S29 for the pore size distribution). Even though these values are smaller than the values reported for UiO-66, 1290 m2/g and 0.49 cm3/g, respectively, they are in line with differences between the two MOFs, as one-third of the BDC linkers in UiO-66 are replaced by the shorter TDC in Zr6(BDC)4(TDC)2. The theoretical values of surface area and pore volume of Zr6(BDC)4(TDC)2 were determined using Zeo++ to be 927 m2/g and 0.30 cm3/g, respectively, and the difference from the experimental measured values is attributed to the missing linkers. The Zr6(BDC)4(TDC)2 sample after N2 adsorption measurements was analyzed by TGA and 1H NMR after base digestion to derive the chemical formula Zr6O4(OH)4(BDC)3.48(TDC)1.93(OH)1.18, where the presence of –OH as the only terminal coordinated ligands on Zr atoms was supported by solid-state 1H MAS NMR (Figure S22.iv). The total number of linkers per formula unit, 5.41, indicates that 10% of the linkers of the theoretical formula are missing, and they predominantly correspond to missing BDC. The proportion of missing linkers from the chemical analysis of Zr6(BDC)4(TDC)2 is similar to that observed for UiO-66 samples. ,

9.

9

Open in a new tab

(a) N2 adsorption–desorption isotherm of Zr6(BDC)4(TDC)2. The closed symbols correspond to the adsorption branch, and the open symbols correspond to the desorption branch. (b) Chromatographic separation of n-hexane, cyclohexane, and benzene by a column packed with Zr6(BDC)4(TDC)2. The column temperature was kept isothermal at 250 °C during the separation.

The porous features of MOFs, often described as cages, are defined by their topology, and their exact size and shape are determined by the geometric characteristics of the linker. The porosity of Zr6(BDC)4(TDC)2 more closely resembles that of UiO-66 than DUT-67, as these two phases share the same underlying fcu topology, characterized by octahedral and tetrahedral cages, which are accessed through triangular windows. In contrast, the pore system of DUT-67, reo topology, consists of octahedral and cuboctahedral cages, and it is accessible through both triangular windows and square-shaped windows. Therefore, there is an alternative diffusion pathway for guest molecules within DUT-67 through only the square-shaped windows, which is not available in the fcu structures (Figure S30). In fcu Zr6(BDC)4(TDC)2, the windows are composed of two BDC and one TDC linker, and they are smaller than those in UiO-66. The bent shape of the TDC results in two different arrangements with respect to BDC and two types of triangular windows. One window has the sulfur atom of the thiophene ring pointing away from the two BDC linkers (Figure a), while the other has the thiophene ring pointing toward the two BDC linkers (Figure b). The two BDC sites also affect the window shape. In particular, the displacement of BDC site 2, from the plane defined by Zr1 atoms of the two connected Zr6 clusters, creates a convex side in the wider window (Figure a) and a concave side in the narrower window (Figure b). Consequently, the tetrahedral cage of Zr6(BDC)4(TDC)2 is distorted from the regular shape adopted in UiO-66, with point symmetry 4̅3m, to an isosceles tetrahedron (Figure c) with point symmetry 2. The arrangement of the two windows on the octahedral cage affords a base built from four TDC linkers, creating an elongated accessible space (Figure d). This cage has a lower point symmetry, 222, compared to the m3̅m point symmetry of the regular octahedral cage in UiO-66. Therefore, the ordering of bent and straight linkers in Zr6(BDC)4(TDC)2 develops a pore system with unique features that cannot be depicted by the simple net representation of the fcu topology.

8.

8

Open in a new tab

Shape of the triangular windows in Zr6(BDC)4(TDC)2 is determined by the orientation of TDC. (a) Wider triangular window with the thiophene ring of TDC pointing away from the BDC linkers. The shape of the window is illustrated by a dark purple distorted triangle. (b) Narrower triangular window with the thiophene ring pointing toward the BDC linkers. The shape of the window is illustrated by a light purple distorted triangle. The purple triangles are included to aid the visual differentiation between the two types of windows. (c) Tetrahedral cage of Zr6(BDC)4(TDC)2 is assembled by two windows of each type and has a distorted shape. (d) Octahedral cage of Zr6(BDC)4(TDC)2 consisting of four windows of each type, and its base has a rectangular shape. Zr is shown in teal, O is shown in red, the C atoms of BDC site 1 are in blue, the C atoms of BDC site 2 are in cyan, and TDC is shown in orange. The hydroxides and oxides on the Zr6 cluster and H atoms are omitted for clarity.

Zr6(BDC)4(TDC)2 was produced at a 0.5 g scale (see Scaled up synthesis in the Supporting Information) and utilized to pack a chromatography column for testing the separation of n-hexane, cyclohexane, and benzene (see Separation in Gas Chromatography in the Supporting Information). This application explores the effect of Zr6(BDC)4(TDC)2 cages on guests with different shapes and evaluates the material for the important separation of cyclohexane and benzene, which is challenging in typical distillation setups. Zr6(BDC)4(TDC)2 separates the mixture into three clear fractions under optimized conditions, whereby the column temperature is kept isothermal at 250 °C. Zr6(BDC)4(TDC)2 exhibits reverse shape selectivity, with n-hexane eluting first, followed by benzene and cyclohexane (Figure ). The concept of reverse shape selectivity was described for zeolites, contrasting the usual selectivity preference for linear hydrocarbons; this phenomenon has been observed, for example, in SAPO-5 and MCM-22. − This selectivity has been widely reported for UiO-66, where it is attributed to the greater rotational freedom of more compact hydrocarbons within framework cages, resulting in their increased retention time and preferential adsorption compared to longer linear hydrocarbons. − The same rationale can be applied to understand the observed reverse shape selectivity of the Zr6(BDC)4(TDC)2 system. Denayer et al. reported the results from a packed GC column containing UiO-66, and the results of Zr6(BDC)4(TDC)2 followed a similar trend. , The thermodynamic parameters indicate stronger separation performance for Zr6(BDC)4(TDC)2, with slightly larger differences in Gibbs free energy between benzene/cyclohexane and cyclohexane/n-hexane compared to the reported values for UiO-66 (Tables S13–S15 and Figure S26). We attribute these changes in selectivity to the unique cage and window geometries of Zr6(BDC)4(TDC)2, resulting from the ordering of the two linkers with differing geometries. Single-component adsorption isotherms of n-hexane, benzene, and cyclohexane were measured at 25 °C compared to the higher temperatures of the GC column; however, the differences in kinetics among the three guests (n-hexane > benzene > cyclohexane) are consistent with the chromatographic results (Figure S32).

We used molecular dynamics (MD) simulations to explore how the differences in window shapes between Zr6(BDC)4(TDC)2 and UiO-66 affect the separation of n-hexane, benzene, and cyclohexane (see the Molecular Dynamics Simulations section of the Supporting Information). First, we placed guest molecules in the tetrahedral cages of each MOF and monitored their displacement over time. In all simulations, the guest molecules remained within the tetrahedral cages, and no diffusion events were observed over 50 ps. These simulations indicate that, in both MOFs, the three guests preferentially adsorb in the tetrahedral cages and, at T = 523 K, reside there over 95% of the time, consistent with previous reports on UiO-66. In the second set of simulations, we placed guest molecules at the center of each octahedral cage and observed how long it takes for them to enter one of the neighboring tetrahedral cages. Table S16 shows that the time it takes for each guest to pass from octahedral to tetrahedral cages in Zr6(BDC)4(TDC)2 follows the same order as their separation behavior: n-hexane passes first, and cyclohexane migrates last. In contrast, in UiO-66, while n-hexane also migrates first, benzene and cyclohexane exhibit similar migration times. Overall, cyclohexane requires significantly more time to enter the tetrahedral cage in Zr6(BDC)4(TDC)2 than for any other guest molecule in either MOF, indicating an enhanced kinetic-based separation of n-hexane/benzene/cyclohexane due to the smaller triangular windows in Zr6(BDC)4(TDC)2 compared to those of UiO-66. When analyzing the guest trajectories in Zr6(BDC)4(TDC)2, it is observed that the guests preferentially use one of the two windows, the wider one (Figure a) instead of the narrower one (Figure b). Therefore, this selective usage restricts the available migration pathways to the tetrahedral cage compared to UiO-66, which features only one window type (Table S16). These simulation results align well with the experimental findings for Zr6(BDC)4(TDC)2 and demonstrate the effect of ordering two linkers with different sizes and shapes on the separation performance.

Conclusions

A new two-linker ordered Zr-MOF containing both bent and straight linkers was identified by high-throughput experimental screening of the ZrOCl2·8H2O/BDC/TDC/FA/DMF chemical space. Zr6(BDC)4(TDC)2-DMF is formed as a pure phase in a wide region of this space, despite competition from the two MOFs formed by each linker individually. Its structure and reactivity reflect the synergy between the geometries of the linear BDC and bent TDC linkers. TDC is only known in frameworks with less than the complete 12-fold connectivity of the parent UiO-66 because of the requirement for the clusters to rotate about multiple axes in order for TDC to connect them, preventing complete bridging of the clusters by 12 ditopic linkers. Here, the bent TDC is accommodated within the 12-c fcu net by coordinating to the equatorial plane of the Zr6 clusters, which rotate about the axis defined by their two axial vertices to afford a Zr6(TDC)2 layer that matches the geometry of the linker. This single rotation defines two distinct sites for the BDC linkers, one displaced from and the other inclined to the single BDC site in UiO-66, that connect these layers to the 12-c Zr6(BDC)4(TDC)2 framework. The combination of the bent shape of TDC and the straight BDC induces distortions in the tetrahedral and octahedral cages of the fcu net, resulting in unique pore shapes, which are utilized for the separation of a n-hexane/benzene/cyclohexane mixture and reflect the ability of multiple linkers to tailor the porosity of established framework topologies.

The displaced BDC creates a second environment for TDC, which it occupies upon the addition of a protic solvent. The protic solvent displaces TDC from its original position, coordinating to the equatorial Zr sites, to the new location and thus forming an 8-c net. Here, the thiophene ring of TDC interacts favorably with the aromatic ring of the displaced BDC, while its carboxylate groups form hydrogen bonds both the cluster hydroxides and cluster-bound protic ligands, in an environment rich in noncovalent interactions that originally arises from the synergy between the linker geometries that enables the formation of the original 12-c net. This reversible transition is an intrinsic property of the Zr6(BDC)4(TDC)2 structure, which allows the disconnection of one linker while the rest of the framework remains intact, affording an environment that retains the disconnected linker within the structure available for the reconnection step. The ordering of multiple linkers to afford new MOF structures by decorating parent frameworks can be delivered by the efficient exploration of large chemical spaces through high-throughput experimental workflows, yielding structural motifs and reactivity patterns beyond those readily envisaged from the structures of single-linker MOFs.

Supplementary Material

ja5c05854_si_001.pdf (8.9MB, pdf)

Acknowledgments

This project was funded by the Leverhulme Trust through the Leverhulme Research Centre for Functional Materials Design (RC-2015-036). We thank EPSRC for support under EP/W036673/1. DJC was supported by the Leverhulme Trust under Research Project grant RPG-2020-066 awarded to FB. We acknowledge the Diamond Light Source for provision of beamtime on the I11 beamline. We thank Craig M. Robertson and Marco Zanella for useful discussions on electron diffraction. We acknowledge the use of resources provided by the Isambard 3 Tier-2 HPC Facility hosted by the University of Bristol.

The data underlying this study is available in the University of Liverpool Data Repository at 10.17638/datacat.liverpool.ac.uk/2986.

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacs.5c05854.

  • Experimental procedures and characterization data, materials and general procedures, PXRD patterns, further structural refinement details, additional structure figures, sorption isotherms, and gas chromatography data (PDF)

The authors declare no competing financial interest.

References

  1. Furukawa H., Cordova K. E., O’Keeffe M., Yaghi O. M.. The Chemistry and Applications of Metal-Organic Frameworks. Science. 2013;341(6149):1230444. doi: 10.1126/science.1230444. [DOI] [PubMed] [Google Scholar]
  2. Bai Y., Dou Y., Xie L. H., Rutledge W., Li J. R., Zhou H. C.. Zr-based metal-organic frameworks: Design, synthesis, structure, and applications. Chem. Soc. Rev. 2016;45:2327–2367. doi: 10.1039/C5CS00837A. [DOI] [PubMed] [Google Scholar]
  3. Chen Z., Hanna S. L., Redfern L. R., Alezi D., Islamoglu T., Farha O. K.. Reticular chemistry in the rational synthesis of functional zirconium cluster-based MOFs. Coord. Chem. Rev. 2019;386:32–49. doi: 10.1016/j.ccr.2019.01.017. [DOI] [Google Scholar]
  4. Cavka J. H., Jakobsen S., Olsbye U., Guillou N., Lamberti C., Bordiga S., Lillerud K. P.. A New Zirconium Inorganic Building Brick Forming Metal Organic Frameworks with Exceptional Stability. J. Am. Chem. Soc. 2008;130(42):13850–13851. doi: 10.1021/ja8057953. [DOI] [PubMed] [Google Scholar]
  5. Øien S., Wragg D., Reinsch H., Svelle S., Bordiga S., Lamberti C., Lillerud K. P.. Detailed structure analysis of atomic positions and defects in zirconium metal-organic frameworks. Cryst. Growth Des. 2014;14(11):5370–5372. doi: 10.1021/cg501386j. [DOI] [Google Scholar]
  6. Assen A. H., Adil K., Cordova K. E., Belmabkhout Y.. The chemistry of metal–organic frameworks with face-centered cubic topology. Coord. Chem. Rev. 2022;468:214644. doi: 10.1016/j.ccr.2022.214644. [DOI] [Google Scholar]
  7. Yaghi, O. M. ; Kalmutzki, M. J. ; Diercks, C. S. . Building Units of MOFs. In Introduction to Reticular Chemistry; Wiley-VCH, 2019; pp 57–81. [Google Scholar]
  8. Guillerm V., Maspoch D.. Geometry Mismatch and Reticular Chemistry: Strategies to Assemble Metal-Organic Frameworks with Non-default Topologies. J. Am. Chem. Soc. 2019;141:16517–16538. doi: 10.1021/jacs.9b08754. [DOI] [PubMed] [Google Scholar]
  9. Furukawa H., Gándara F., Zhang Y. B., Jiang J., Queen W. L., Hudson M. R., Yaghi O. M.. Water adsorption in porous metal-organic frameworks and related materials. J. Am. Chem. Soc. 2014;136(11):4369–4381. doi: 10.1021/ja500330a. [DOI] [PubMed] [Google Scholar]
  10. Bon V., Senkovska I., Weiss M. S., Kaskel S.. Tailoring of network dimensionality and porosity adjustment in Zr- and Hf-based MOFs. CrystEngComm. 2013;15(45):9572–9577. doi: 10.1039/c3ce41121d. [DOI] [Google Scholar]
  11. Guillerm V., Grancha T., Imaz I., Juanhuix J., Maspoch D.. Zigzag Ligands for Transversal Design in Reticular Chemistry: Unveiling New Structural Opportunities for Metal-Organic Frameworks. J. Am. Chem. Soc. 2018;140(32):10153–10157. doi: 10.1021/jacs.8b07050. [DOI] [PubMed] [Google Scholar]
  12. Jacobsen J., Reinsch H., Stock N.. Systematic Investigations of the Transition between Framework Topologies in Ce/Zr-MOFs. Inorg. Chem. 2018;57(20):12820–12826. doi: 10.1021/acs.inorgchem.8b02019. [DOI] [PubMed] [Google Scholar]
  13. Wang B., Huang H., Lv X.-L., Xie Y., Li M., Li J.-R.. Tuning CO2 Selective Adsorption over N2 and CH4 in UiO-67 Analogues through Ligand Functionalization. Inorg. Chem. 2014;53(17):9254–9259. doi: 10.1021/ic5013473. [DOI] [PubMed] [Google Scholar]
  14. Panagiotou N., Liatsou I., Pournara A., Angeli G. K., Giappa R. M., Tylianakis E., Manos M. J., Froudakis G. E., Trikalitis P. N., Pashalidis I., Tasiopoulos A. J.. Water-stable 2-D Zr MOFs with exceptional UO2 2+ sorption capability. J. Mater. Chem. A. 2020;8(4):1849–1857. doi: 10.1039/C9TA10701K. [DOI] [Google Scholar]
  15. Wang S., Chen L., Wahiduzzaman M., Tissot A., Zhou L., Ibarra I. A., Gutiérrez-Alejandre A., Lee J. S., Chang J.-S., Liu Z., Marrot J., Shepard W., Maurin G., Xu Q., Serre C.. A Mesoporous Zirconium-Isophthalate Multifunctional Platform. Matter. 2021;4(1):182–194. doi: 10.1016/j.matt.2020.10.009. [DOI] [Google Scholar]
  16. Dreischarf A. C., Lammert M., Stock N., Reinsch H.. Green Synthesis of Zr-CAU-28: Structure and Properties of the First Zr-MOF Based on 2,5-Furandicarboxylic Acid. Inorg. Chem. 2017;56(4):2270–2277. doi: 10.1021/acs.inorgchem.6b02969. [DOI] [PubMed] [Google Scholar]
  17. Bon V., Senkovska I., Baburin I. A., Kaskel S.. Zr- and Hf-based metal-organic frameworks: Tracking down the polymorphism. Cryst. Growth Des. 2013;13(3):1231–1237. doi: 10.1021/cg301691d. [DOI] [Google Scholar]
  18. Drache F., Bon V., Senkovska I., Getzschmann J., Kaskel S.. The modulator driven polymorphism of Zr­(IV) based metal–organic frameworks. Philos. Trans. R. Soc., A. 2017;375(2084):20160027. doi: 10.1098/rsta.2016.0027. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Lee S. J., Telfer S. G.. Multicomponent Metal-Organic Frameworks. Angew. Chem., Int. Ed. 2023;62(44):e202306341. doi: 10.1002/anie.202306341. [DOI] [PubMed] [Google Scholar]
  20. Jiang H., Benzaria S., Alsadun N., Jia J., Czaban-Jóźwiak J., Guillerm V., Shkurenko A., Thiam Z., Bonneau M., Maka V. K., Chen Z., Ameur Z. O., O’Keeffe M., Eddaoudi M.. Merged-nets enumeration for the systematic design of multicomponent reticular structures. Science. 2024;386(6722):659–666. doi: 10.1126/science.ads7866. [DOI] [PubMed] [Google Scholar]
  21. Helal A., Yamani Z. H., Cordova K. E., Yaghi O. M.. Multivariate metal-organic frameworks. Natl. Sci. Rev. 2017;4(3):296–298. doi: 10.1093/nsr/nwx013. [DOI] [Google Scholar]
  22. Kong X., Deng H., Yan F., Kim J., Swisher J. A., Smit B., Yaghi O. M., Reimer J. A.. Mapping of Functional Groups in Metal-Organic Frameworks. Science. 2013;341(6148):882–885. doi: 10.1126/science.1238339. [DOI] [PubMed] [Google Scholar]
  23. Tu B., Diestel L., Shi Z.-L., Bandara W. R. L. N., Chen Y., Lin W., Zhang Y.-B., Telfer S. G., Li Q.. Harnessing Bottom-Up Self-Assembly To Position Five Distinct Components in an Ordered Porous Framework. Angew. Chem., Int. Ed. 2019;58(16):5348–5353. doi: 10.1002/anie.201900863. [DOI] [PubMed] [Google Scholar]
  24. Liu L., Yao Z., Ye Y., Yang Y., Lin Q., Zhang Z., O’Keeffe M., Xiang S.. Integrating the Pillared-Layer Strategy and Pore-Space Partition Method to Construct Multicomponent MOFs for C2H2/CO2 Separation. J. Am. Chem. Soc. 2020;142(20):9258–9266. doi: 10.1021/jacs.0c00612. [DOI] [PubMed] [Google Scholar]
  25. Chen Y., Idrees K. B., Mian M. R., Son F. A., Zhang C., Wang X., Farha O. K.. Reticular Design of Precise Linker Installation into a Zirconium Metal-Organic Framework to Reinforce Hydrolytic Stability. J. Am. Chem. Soc. 2023;145(5):3055–3063. doi: 10.1021/jacs.2c11830. [DOI] [PubMed] [Google Scholar]
  26. Yuan S., Lu W., Chen Y. P., Zhang Q., Liu T. F., Feng D., Wang X., Qin J., Zhou H. C.. Sequential linker installation: Precise placement of functional groups in multivariate metal-organic frameworks. J. Am. Chem. Soc. 2015;137(9):3177–3180. doi: 10.1021/ja512762r. [DOI] [PubMed] [Google Scholar]
  27. Yuan S., Chen Y. P., Qin J. S., Lu W., Zou L., Zhang Q., Wang X., Sun X., Zhou H. C.. Linker Installation: Engineering Pore Environment with Precisely Placed Functionalities in Zirconium MOFs. J. Am. Chem. Soc. 2016;138(28):8912–8919. doi: 10.1021/jacs.6b04501. [DOI] [PubMed] [Google Scholar]
  28. Corma A., Díaz-Cabañas M. J., Jordá J. L., Martínez C., Moliner M.. High-throughput synthesis and catalytic properties of a molecular sieve with 18- and 10-member rings. Nature. 2006;443(7113):842–845. doi: 10.1038/nature05238. [DOI] [PubMed] [Google Scholar]
  29. Greenaway R. L., Santolini V., Bennison M. J., Alston B. M., Pugh C. J., Little M. A., Miklitz M., Eden-Rump E. G. B., Clowes R., Shakil A., Cuthbertson H. J., Armstrong H., Briggs M. E., Jelfs K. E., Cooper A. I.. High-throughput discovery of organic cages and catenanes using computational screening fused with robotic synthesis. Nat. Commun. 2018;9(1):2849. doi: 10.1038/s41467-018-05271-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Zhao W., Yan P., Li B., Bahri M., Liu L., Zhou X., Clowes R., Browning N. D., Wu Y., Ward J. W., Cooper A. I.. Accelerated Synthesis and Discovery of Covalent Organic Framework Photocatalysts for Hydrogen Peroxide Production. J. Am. Chem. Soc. 2022;144(22):9902–9909. doi: 10.1021/jacs.2c02666. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Bauer S., Serre C., Devic T., Horcajada P., Marrot J., Férey G., Stock N.. High-Throughput Assisted Rationalization of the Formation of Metal Organic Frameworks in the Iron­(III) Aminoterephthalate Solvothermal System. Inorg. Chem. 2008;47(17):7568–7576. doi: 10.1021/ic800538r. [DOI] [PubMed] [Google Scholar]
  32. Tollitt A. M., Vismara R., Daniels L. M., Antypov D., Gaultois M. W., Katsoulidis A. P., Rosseinsky M. J.. High-throughput discovery of a rhombohedral twelve-connected zirconium-based metal-organic framework with ordered terephthalate and fumarate linkers. Angew. Chem., Int. Ed. 2021;4:26939–26946. doi: 10.1002/anie.202108150. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Morris W., Wang S., Cho D., Auyeung E., Li P., Farha O. K., Mirkin C. A.. Role of Modulators in Controlling the Colloidal Stability and Polydispersity of the UiO-66 Metal–Organic Framework. ACS Appl. Mater. Interfaces. 2017;9(39):33413–33418. doi: 10.1021/acsami.7b01040. [DOI] [PubMed] [Google Scholar]
  34. Trickett C. A., Gagnon K. J., Lee S., Gándara F., Bürgi H.-B., Yaghi O. M.. Definitive Molecular Level Characterization of Defects in UiO-66 Crystals. Angew. Chem., Int. Ed. 2015;54(38):11162–11167. doi: 10.1002/anie.201505461. [DOI] [PubMed] [Google Scholar]
  35. Precisvalle N., Ho P. H., Cacciaguerra T., Deabate S., Girard L., Toquer G., Nguyen K. D., Le H. V., Di Renzo F., Martucci A., Trens P.. Extra-framework zirconium clusters in metal organic framework DUT-67 controlled by the choice of the metal precursor. Microporous Mesoporous Mater. 2021;324:111293. doi: 10.1016/j.micromeso.2021.111293. [DOI] [Google Scholar]
  36. Valekar A. H., Lee M., Yoon J. W., Kwak J., Hong D.-Y., Oh K.-R., Cha G.-Y., Kwon Y.-U., Jung J., Chang J.-S., Hwang Y. K.. Catalytic Transfer Hydrogenation of Furfural to Furfuryl Alcohol under Mild Conditions over Zr-MOFs: Exploring the Role of Metal Node Coordination and Modification. ACS Catal. 2020;10(6):3720–3732. doi: 10.1021/acscatal.9b05085. [DOI] [Google Scholar]
  37. Wu H., Chua Y. S., Krungleviciute V., Tyagi M., Chen P., Yildirim T., Zhou W.. Unusual and Highly Tunable Missing-Linker Defects in Zirconium Metal–Organic Framework UiO-66 and Their Important Effects on Gas Adsorption. J. Am. Chem. Soc. 2013;135(28):10525–10532. doi: 10.1021/ja404514r. [DOI] [PubMed] [Google Scholar]
  38. Kim H., Kim D., Moon D., Choi Y. N., Baek S. B., Lah M. S.. Symmetry-guided syntheses of mixed-linker Zr metal-organic frameworks with precise linker locations. Chem. Sci. 2019;10(22):5801–5806. doi: 10.1039/C9SC01301F. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Tsuzuki S., Honda K., Azumi R.. Model Chemistry Calculations of Thiophene Dimer Interactions: Origin of π-Stacking. J. Am. Chem. Soc. 2002;124(41):12200–12209. doi: 10.1021/ja0204877. [DOI] [PubMed] [Google Scholar]
  40. Castellano O., Gimon R., Soscun H.. Theoretical Study of the σ–π and π–π Interactions in Heteroaromatic Monocyclic Molecular Complexes of Benzene, Pyridine, and Thiophene Dimers: Implications on the Resin–Asphaltene Stability in Crude Oil. Energy Fuels. 2011;25(6):2526–2541. doi: 10.1021/ef101471t. [DOI] [Google Scholar]
  41. Marreiros J., Caratelli C., Hajek J., Krajnc A., Fleury G., Bueken B., De Vos D. E., Mali G., Roeffaers M. B. J., Van Speybroeck V., Ameloot R.. Active Role of Methanol in Post-Synthetic Linker Exchange in the Metal–Organic Framework UiO-66. Chem. Mater. 2019;31(4):1359–1369. doi: 10.1021/acs.chemmater.8b04734. [DOI] [Google Scholar]
  42. Ito S., White F. J., Okunishi E., Aoyama Y., Yamano A., Sato H., Ferrara J. D., Jasnowski M., Meyer M.. Structure determination of small molecule compounds by an electron diffractometer for 3D ED/MicroED. CrystEngComm. 2021;23(48):8622–8630. doi: 10.1039/D1CE01172C. [DOI] [Google Scholar]
  43. Truong K.-N., Ito S., Wojciechowski J. M., Göb C. R., Schürmann C. J., Yamano A., Del Campo M., Okunishi E., Aoyama Y., Mihira T., Hosogi N., Benet-Buchholz J., Escudero-Adán E. C., White F. J., Ferrara J. D., Bücker R.. Making the Most of 3D Electron Diffraction: Best Practices to Handle a New Tool. Symmetry. 2023;15(8):1555. doi: 10.3390/sym15081555. [DOI] [Google Scholar]
  44. Tang J., Li S., Chu Y., Xiao Y., Xu J., Deng F.. Solid-state NMR studies of the acidity of functionalized metal–organic framework UiO-66 materials. Magn. Reson. Chem. 2020;58(11):1091–1098. doi: 10.1002/mrc.4923. [DOI] [PubMed] [Google Scholar]
  45. Nguyen K. D., Kutzscher C., Ehrling S., Senkovska I., Bon V., de Oliveira M., Gutmann T., Buntkowsky G., Kaskel S.. Insights into the role of zirconium in proline functionalized metal-organic frameworks attaining high enantio- and diastereoselectivity. J. Catal. 2019;377:41–50. doi: 10.1016/j.jcat.2019.07.003. [DOI] [Google Scholar]
  46. Nandy A., Forse A. C., Witherspoon V. J., Reimer J. A.. NMR Spectroscopy Reveals Adsorbate Binding Sites in the Metal–Organic Framework UiO-66­(Zr) J. Phys. Chem. C. 2018;122(15):8295–8305. doi: 10.1021/acs.jpcc.7b12628. [DOI] [Google Scholar]
  47. Venel F., Volkringer C., Lafon O., Pourpoint F.. Probing adsorption of water and DMF in UiO-66­(Zr) using solid-state NMR. Solid State Nucl. Magn. Reson. 2022;120:101797. doi: 10.1016/j.ssnmr.2022.101797. [DOI] [PubMed] [Google Scholar]
  48. Kim M., Cahill J. F., Su Y., Prather K. A., Cohen S. M.. Postsynthetic ligand exchange as a route to functionalization of ‘inert’ metal–organic frameworks. Chem. Sci. 2012;3(1):126–130. doi: 10.1039/C1SC00394A. [DOI] [Google Scholar]
  49. Chen C.-X., Wei Z., Jiang J.-J., Fan Y.-Z., Zheng S.-P., Cao C.-C., Li Y.-H., Fenske D., Su C.-Y.. Precise Modulation of the Breathing Behavior and Pore Surface in Zr-MOFs by Reversible Post-Synthetic Variable-Spacer Installation to Fine-Tune the Expansion Magnitude and Sorption Properties. Angew. Chem., Int. Ed. 2016;55(34):9932–9936. doi: 10.1002/anie.201604023. [DOI] [PubMed] [Google Scholar]
  50. Furukawa H., Gándara F., Zhang Y.-B., Jiang J., Queen W. L., Hudson M. R., Yaghi O. M.. Water Adsorption in Porous Metal–Organic Frameworks and Related Materials. J. Am. Chem. Soc. 2014;136(11):4369–4381. doi: 10.1021/ja500330a. [DOI] [PubMed] [Google Scholar]
  51. Willems T. F., Rycroft C. H., Kazi M., Meza J. C., Haranczyk M.. Algorithms and tools for high-throughput geometry-based analysis of crystalline porous materials. Microporous Mesoporous Mater. 2012;149(1):134–141. doi: 10.1016/j.micromeso.2011.08.020. [DOI] [Google Scholar]
  52. Liu L., Chen Z., Wang J., Zhang D., Zhu Y., Ling S., Huang K.-W., Belmabkhout Y., Adil K., Zhang Y., Slater B., Eddaoudi M., Han Y.. Imaging defects and their evolution in a metal–organic framework at sub-unit-cell resolution. Nat. Chem. 2019;11(7):622–628. doi: 10.1038/s41557-019-0263-4. [DOI] [PubMed] [Google Scholar]
  53. Denayer J. F. M., Daems I., Baron G. V.. Adsorption and Reaction in Confined Spaces. Oil Gas Sci. Technol. - Rev. IFP. 2006;61(4):561–569. doi: 10.2516/ogst:2006025a. [DOI] [Google Scholar]
  54. Santilli D. S., Harris T. V., Zones S. I.. Inverse shape selectivity in molecular sieves: Observations, modelling, and predictions. Microporous Mater. 1993;1(5):329–341. doi: 10.1016/0927-6513(93)80049-Z. [DOI] [Google Scholar]
  55. Denayer J. F. M., Ocakoglu A. R., Martens J. A., Baron G. V.. Investigation of inverse shape selectivity in alkane adsorption on SAPO-5 zeolite using the tracer chromatography technique. J. Catal. 2004;226(1):240–244. doi: 10.1016/j.jcat.2004.05.018. [DOI] [Google Scholar]
  56. Denayer J. F. M., Ocakoglu R. A., Arik I. C., Kirschhock C. E. A., Martens J. A., Baron G. V.. Rotational Entropy Driven Separation of Alkane/Isoalkane Mixtures in Zeolite Cages. Angew. Chem., Int. Ed. 2005;44(3):400–403. doi: 10.1002/anie.200454058. [DOI] [PubMed] [Google Scholar]
  57. Bárcia P. S., Guimarães D., Mendes P. A. P., Silva J. A. C., Guillerm V., Chevreau H., Serre C., Rodrigues A. E.. Reverse shape selectivity in the adsorption of hexane and xylene isomers in MOF UiO-66. Microporous Mesoporous Mater. 2011;139(1):67–73. doi: 10.1016/j.micromeso.2010.10.019. [DOI] [Google Scholar]
  58. Chang N., Yan X.-P.. Exploring reverse shape selectivity and molecular sieving effect of metal-organic framework UIO-66 coated capillary column for gas chromatographic separation. J. Chromatogr. A. 2012;1257:116–124. doi: 10.1016/j.chroma.2012.07.097. [DOI] [PubMed] [Google Scholar]
  59. Kolokolov D. I., Maryasov A. G., Ollivier J., Freude D., Haase J., Stepanov A. G., Jobic H.. Uncovering the Rotation and Translational Mobility of Benzene Confined in UiO-66 (Zr) Metal–Organic Framework by the 2H NMR–QENS Experimental Toolbox. J. Phys. Chem. C. 2017;121(5):2844–2857. doi: 10.1021/acs.jpcc.6b12001. [DOI] [Google Scholar]
  60. Bozbiyik B., Duerinck T., Lannoeye J., De Vos D. E., Baron G. V., Denayer J. F. M.. Adsorption and separation of n-hexane and cyclohexane on the UiO-66 metal–organic framework. Microporous Mesoporous Mater. 2014;183:143–149. doi: 10.1016/j.micromeso.2013.07.035. [DOI] [Google Scholar]
  61. Duerinck T., Bueno-Perez R., Vermoortele F., De Vos D. E., Calero S., Baron G. V., Denayer J. F. M.. Understanding Hydrocarbon Adsorption in the UiO-66 Metal–Organic Framework: Separation of (Un)­saturated Linear, Branched, Cyclic Adsorbates, Including Stereoisomers. J. Phys. Chem. C. 2013;117(24):12567–12578. doi: 10.1021/jp402294h. [DOI] [Google Scholar]
  62. Sławek A., Jajko G., Ogorzały K., Dubbeldam D., Vlugt T. J. H., Makowski W.. The Influence of UiO-66 Metal–Organic Framework Structural Defects on Adsorption and Separation of Hexane Isomers. Chem. - Eur. J. 2022;28(29):e202200030. doi: 10.1002/chem.202200030. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ja5c05854_si_001.pdf (8.9MB, pdf)

Data Availability Statement

The data underlying this study is available in the University of Liverpool Data Repository at 10.17638/datacat.liverpool.ac.uk/2986.

Articles from Journal of the American Chemical Society are provided here courtesy of American Chemical Society

RESOURCES