www.acsami.org Research Article Influence of the Surface Chemistry of Metal−Organic Polyhedra in Their Assembly into Ultrathin Films for Gas Separation Inés Tejedor, Miguel A. Andrés, Arnau Carné-Sánchez, Mónica Arjona, Marta Pérez-Miana, Javier Sánchez-Laínez, Joaquín Coronas, Philippe Fontaine, Michel Goldmann, Olivier Roubeau, Daniel Maspoch, and Ignacio Gascón* Cite This: ACS Appl. Mater. Interfaces 2022, 14, 27495−27506 Read Online ACCESS * See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. Metrics & More Article Recommendations sı Supporting Information ABSTRACT: The formation of ultrathin films of Rh-based porous metal− organic polyhedra (Rh-MOPs) by the Langmuir−Blodgett method has been explored. Homogeneous and dense monolayer films were formed at the air− Downloaded via 158.109.48.66 on May 10, 2023 at 10:56:03 (UTC). water interface either using two different coordinatively alkyl-functionalized Rh-MOPs (HRhMOP(diz)12 and HRhMOP(oiz)12) or by in situ incorpo- ration of aliphatic chains to the axial sites of dirhodium paddlewheels of another Rh-MOP (OHRhMOP) at the air−liquid interface. All these Rh- MOP monolayers were successively deposited onto different substrates in order to obtain multilayer films with controllable thicknesses. Aliphatic chains were partially removed from HRhMOP(diz)12 films post-synthetically by a simple acid treatment, resulting in a relevant modification of the film hydrophobicity. Moreover, the CO2/N2 separation performance of Rh-MOP- supported membranes was also evaluated, proving that they can be used as selective layers for efficient CO2 separation. KEYWORDS: metal−organic polyhedra (MOP), ultrathin films, surface functionalization, CO2 separation, MOP assembly 1. INTRODUCTION enables using them under homogeneous conditions or shaping Intrinsically porous metal−organic polyhedra1 (MOPs) are a them into functional macroscopic objects. In such a context, class of metal−organic cages2 obtained from the self-assembly we15 and Yin and co-workers16 recently demonstrated that of metal ions and organic linkers. Compared to metal−organic alkyl-functionalized MOPs can form stable monolayers at the frameworks,3 widely used for different applications that make air−liquid interface. The self-assembly of these MOP films can use of their permanent porosity (separations,4 sensing,5 and be studied in situ during film formation at different surface drug delivery,6 among others), MOPs present some relevant densities (obtained from surface pressure−area isotherms) advantages due to their molecular nature, such as tailorable using specific characterization techniques such as Brewster solubility and easy processability.7 These properties make angle microscopy (BAM) or neutron reflectivity (NR), which MOPs excellent candidates for molecular separation processes provide relevant information about the organization and in liquid phases, including cargo transfer between immiscible structure of the films. Moreover, these films can be sequentially solvents,8 the formation of ion channels,9 and the development transferred onto appropriate solid substrates by horizontal of membranes for gas separation.10 (Langmuir−Schaefer, LS) or vertical (Langmuir−Blodgett, Previous studies have shown that cuboctahedral rhodium- LB) depositions to obtain multilayer films with the required based MOPs (Rh-MOPs), which are assembled from 1,3- thickness. benzenedicarboxylate and dirhodium paddlewheels,11 offer Dense and ultrathin MOP films would be the basis for the very interesting possibilities for separation processes. The development of efficient MOP-based membranes for separa- cavity of Rh-MOPs can be used for selective entrapment of tion technologies. For instance, LS multilayer films (ca. 60 nm) gases/molecules,12 whereas their reactive surface can be used to coordinatively attach selected molecules.13 The chemical robustness of Rh-MOPs enables them to sustain their Received: April 7, 2022 performance even when exposed to harsh conditions, such as Accepted: May 24, 2022 exposure to water, acids, or competing coordinating ligands.14 Published: June 3, 2022 In addition, the surface of Rh-MOPs can be functionalized to modulate their solubility from water to polar/non-polar organic solvents, which increases their processability and © 2022 The Authors. Published by American Chemical Society https://doi.org/10.1021/acsami.2c06123 27495 ACS Appl. Mater. Interfaces 2022, 14, 27495−27506 ACS Applied Materials & Interfaces www.acsami.org Research Article fabricated with the cuboctahedral Rh-MOP named reproducibility of the results before further experiments (BAM image C12RhMOP, with the formula [Rh2(C12-bdc)2]12 (where C12- acquisition, film transfer onto different substrates, etc.) were carried bdc is 5-dodecyloxybenzene-1,3-dicarboxylate), showed re- out. BAM images were acquired using a KSV NIMA Micro BAM markable CO2 permeance and CO2/N2 selectivity values in equipped with a red laser (659 nm, 50 mW) as the light source. The incidence angle was fixed at 53.1°, and a black quartz plate was placed post-combustion.15 inside the trough as a light trap. The optics of the system provided a In this study, we expand the fabrication of Rh-MOP films to spatial resolution of 6 μm per pixel in the water surface plane and a different compounds with aliphatic chains coordinated to Rh- field of view of 3600 × 4000 μm2. MOP axial sites. These chains can be incorporated before LS films were fabricated using a KSV-NIMA model KN 2003 (580 spreading Rh-MOP onto the liquid surface or in situ at the air− × 145 mm), also arranged in a symmetrical double-barrier liquid interface. Moreover, an acid treatment after film configuration. formation allows a partial ligand cleavage, which modifies the Both troughs were placed inside closed cabinets in a dedicated hydrophobicity of the film surface, opening the possibility to facility to limit the presence of dust and at a constant temperature (20 post-synthetically modify the properties of ultrathin MOP- ± 1 °C). Ultra-pure Milli-Q water (resistivity 18.2 MΩ·cm) was used in all the experiments as the subphase. The films were compressed in based films. all cases at a constant speed of 8 cm2/min, and surface pressure was continuously monitored during the experiments by means of a 2. MATERIALS AND METHODS Wilhelmy balance using a filter paper plate. 2.1. Synthesis and Post-Synthetic Modification of Rh-MOPs. LS films were transferred after film compression and stabilization of Both ligands 1-dodecyl-1H-imidazole (diz) and 1-octyl-1H-imidazole the desired surface pressure, keeping the substrates parallel to the (oiz) were synthesized according to a reported procedure.17 water surface using a vacuum pump-based horizontal dipping clamp OHRhMOP (formula: [Rh2(OH-BDC)2]12, where OH-BDC is 5- (KSV KN-006). The substrate was approached to the water surface at hydroxy-1,3-benzenedicarboxylate) was synthesized according to the a vertical speed of 1 mm/min. Once the substrate touched the film, it reported procedure.14 HRhMOP(diz)12 (formula: was withdrawn at a vertical speed of 10 mm/min. After each transfer, [Rh2(BDC)2(diz)1]12, where BDC is 1,3-benzenedicarboxylate) was films were dried with N2 at ambient temperature, and the synthesized according to the reported procedure.7b HRhMOP(oiz)12 transferences were repeated, when required, as many times as (formula: [Rh2(BDC)2(oiz)1]12) was synthesized following a necessary to obtain films with the desired number of Rh-MOP layers. procedure similar to HRhMOP(diz)12. To this end, 100 mg of The UV−vis spectra of the Rh-MOP solutions and films deposited HRhMOP (formula: [Rh2(BDC)2]12) was dispersed in 20 mL of onto the quartz substrates were recorded on a Varian Cary 50 Bio dichloromethane (DCM). Then, 42.05 mg of oiz (15 equiv) was spectrophotometer. added. The solution was sonicated for 5 min and filtered. The filtrate LS films were characterized through atomic force microscopy was evaporated under vacuum. The solid residue was washed twice (AFM) under ambient conditions using either a NTEGRA Aura with ether to remove unreacted oiz. Finally, the obtained purple solid microscope from NT-MDT (semicontact mode employing a was dried under vacuum. SF005&AU006NTF head) or a Bruker Multimode 8 microscope, OHRhMOP was alkyl-functionalized in solution via coordination which is equipped with a NanoScope V control unit (tapping mode). chemistry with an excess (1:25) of the N-donor ligand (diz) to obtain AFM data were collected using silicon tips with typical spring OHRhMOP(diz)12 (formula: [Rh2(OH-BDC)2(diz)1]12). The same constants and resonant frequencies of 3.5−5 N·m−1 and 90−210 kHz, ligand was also used to study the in situ functionalization of respectively. OHRhMOP at the air−liquid interface. The bulk synthesis was The Rh-MOP mass deposited onto QCM substrates (5 MHz AT- performed by adding 0.9 μL of diz (ca. 3.8 × 10−3 mmol) to a cut QCM crystals from Stanford Research Systems) was determined dispersion of 1 mg of OHRhMOP (ca. 1.5 × 10−4 mmol) in 2 mL of using a Stanford Research Systems (SRS) QCM200 system equipped THF. The insoluble green powder of OHRhMOP was immediately with a QCM25 crystal oscillator working at 5 MHz. QCM crystals dissolved after the addition of diz. The resultant purple solution were placed in a O100RH Kynar crystal holder before and after film presented its maximum absorption in the visible region at ca. 552 nm transfer, and the mass increases were calculated from frequency (Figure S1), which indicates that all the dirhodium paddlewheels in changes using the Sauerbrey equation, Δf = −CfΔm, where Δf is the the Rh-MOP structure are coordinated to one diz, as it has been observed frequency change in Hz, Cf is the sensitivity factor of the previously reported.14 OHRhMOP dissolved in a 1:5 methanol/ QCM crystal provided by the manufacturer (0.0566 Hz·cm2/ng), and chloroform mixture (the same mixture used for the preparation of Δm is the change in mass per unit area. Langmuir films) showed its maximum absorbance at 588 nm (also X-ray photoelectron spectroscopy (XPS) spectra were acquired on plotted in Figure S1 for comparison). a Kratos AXIS ultra DLD spectrometer with a monochromatic Al Kα 2.2. Langmuir and LS Film Fabrication and Character- X-ray source (1486.6 eV) using a pass energy of 20 eV. The ization. Diluted solutions (ca. 3.0 × 10−6 M) of C12RhMOP, photoelectron take-off angle was 90° with respect to the sample plane. HRhMOP(diz)12, and HRhMOP(oiz)12 were prepared in DCM, The XPS binding energies reported in this work were referenced to while OHRhMOP was dissolved in a mixture of 1:5 methanol/ the C 1s peak at 284.6 eV. The data treatment was done using chloroform (ca. 6.0 × 10−5 M). Appropriate volumes of these CasaXPS software. The Rh 3d high-resolution spectra were simulated solutions (ca. 1000−1200 μL for C12RhMOP, HRhMOP(diz)12, and using Shirley-type background and the components detailed in Table HRhMOP(oiz)12 and 5000 μL for OHRhMOP) were spread on the S1. water subphase for the formation of Rh-MOP films. 2.3. Synchrotron Characterization of Langmuir Films. In situ For the interfacial formation of alkyl-functionalized Rh-MOPs at grazing-incidence X-ray diffraction (GIXD) and grazing-incidence the air−water interface, 3150 μL of OHRhMOP solution and an small-angle X-ray scattering (GISAXS) measurements at the air− appropriate volume of diz solution (concentration of 2.25 × 10−4 M water interface were performed in situ at SIRIUS Beamline of in DCM), in order to achieve the desired molar proportion Rh-MOP/ Synchrotron SOLEIL (Saint Aubin, France).18 The measurements ligand, were spread successively onto the air/water interface. were performed using an incident X-ray beam of 8 keV (λ = 0.155 In all experiments, after solvent evaporation (ca. 15 min), the nm), a beam size of 0.1 × 0.5 mm2 (V × H) for GISAXS and 0.1 × 2.0 floating molecules were compressed to induce their assembly into an mm2 (V × H) for GIXD, and an incidence angle of 2.0 mrad with the ordered film. water surface below the total external reflection critical angle value of A NIMA 702BAM Langmuir trough (720 × 100 mm) equipped the air−water interface (2.7 mrad at 8 keV). GISAXS measurements with two mobile symmetrical barriers was used to obtain surface were done using a 2D PILATUS3 1M (Dectris, Switzerland) detector pressure versus area (π−A) isotherms and BAM images. Each π−A located at 4.5 m from the sample. A vertical tungsten rod was used as isotherm was registered at least three times in order to check the a beam-stop for the direct and reflected beams. A flight path tube 27496 https://doi.org/10.1021/acsami.2c06123 ACS Appl. Mater. Interfaces 2022, 14, 27495−27506 ACS Applied Materials & Interfaces www.acsami.org Research Article Figure 1. Schematic representation of the three types of Rh-MOPs included in this study: without alkyl substituents (left, OHRhMOP), with 12 alkyl chains added through coordination of an imidazole ligand at the outer axial position of each Rh pair (middle, HRhMOP(diz)12), and with 24 alkyl chains covalently anchored on the organic ligand (right, C12RhMOP). Representations are, respectively, based on the reported X-ray structures of OHRhMOP and HRhMOP(tertPy)14 and on the Cu analogue to C12RhMOP.20 The structure of the other Rh-MOP studied (HRhMOP(oiz)12) is similar to that of HRhMOP(diz)12, although the alkyl chains are shorter (octyl instead of dodecyl). The maximum diameter of the alkyl-functionalized MOPs was calculated considering all the alkyl chains in an extended configuration. flushed with helium gas was located between the sample and the 3. RESULTS AND DISCUSSION detector in order to reduce absorption and scattering by air. The GIXD setup used a 2D PILATUS2 (Dectris, Switzerland) detector 3.1. Assembly of Soluble Rh-MOPs into Thin Films at combined with a Soller collimator of 0.05° resolution. This detection the Air−Liquid Interface. We initially studied the monolayer setup was continuously scanned over the in-plane 2θ angle in order to film formation of our soluble Rh-MOPs (see Figure 1) to record the horizontal and vertical distribution intensity. Peak unveil the impact of surface functionalization on their self- adjustment was performed with the Qz-integrated intensity. assembly behavior at the air−liquid interface. Monolayers were prepared in a dedicated Langmuir trough18 To this end, the appropriate volume for each compound was enclosed in a gastight chamber flushed by water-saturated helium gas spread on the water subphase. The floating films obtained after flow to reduce gas scattering and to avoid the damage of the solvent evaporation were compressed at a constant speed in monolayer by the beam. The temperature was kept constant thanks to order to obtain the characteristic surface pressure−area a water circulating bath at 20 ± 1 °C. isotherms (π−A) for each compound (Figure 2). 2.4. Rh-MOP Films for Gas Separation Studies. For gas separation studies, Rh-MOP LS films were deposited onto permeable poly[1-(trimethylsilyl)-1-propyne] (PTMSP) supports of a thickness of ca. 80 μm. Scheme S1 illustrates the LS sequential deposition of MOP films onto the PTMSP membrane in order to achieve the desired number of MOP monolayers in the selective layer. PTMSP supports were prepared following a procedure previously described.19 PTMSP (Cymit Quimica, >95%) was first dissolved in toluene (analytical reagent purchased from VWR Chemicals, >99.5%) at room temperature (solution concentration 1.9 wt %). Then, the solution was poured on a glass Petri dish and allowed to dry for 72 h at room temperature. The obtained films were immersed in methanol (Sigma- Aldrich, 99.8%) for 24 h to remove the traces of toluene. Before using them, the PTMSP supports were gently dried with paper sheets. Rh-MOP/PTMSP membranes were cut in circular areas of 2.12 cm2 for conducting gas separation studies at 35 °C and two different Figure 2. Representative surface pressure−area (π−A) isotherms feed pressures (1 and 3 bar). The membranes were assembled into a obtained for the four Rh-MOPs studied. The C12RhMOP isotherm15 module consisting of two stainless steel pieces and a 316LSS is plotted for comparison purposes (dashed line). macroporous disk support (from Mott Co.) with a 20 μm nominal pore size, gripped inside with Viton O-rings. The permeation module was placed in a UNE 200 Memmert oven to control the temperature π−A isotherms show a similar behavior for all Rh-MOPs of the module. Gas separation measurements were carried out by functionalized with alkyl chains, including the compound feeding a 10/90 in volume CO2/N2 mixture (100 cm3(STP)/min) to C12RhMOP previously reported by us15 (plotted as the dashed the feed side by means of two mass flow controllers (Alicat Scientific, line in Figure 2 for comparison). For these compounds, the MC-100CCM-D), while the permeate side of the membrane was surface pressure starts to rise at areas per molecule close to 9 swept with a 4.5 cm3(STP)/min mass flow-controlled stream of He at nm2/molecule, and further compression leads to a marked 1 bar (Alicat Scientific, MC-5CCM-D). The permeate concentrations increase of the surface pressure, revealing the formation of of CO2 and N2 were analyzed online by an Agilent 3000 A micro-gas condensed films. However, the surface pressure reaches higher chromatograph equipped with a thermal conductivity detector . values for the Rh-MOP functionalized with 24 covalently Permeance was calculated in GPU [10−6 cm3(STP)/(cm2 s cmHg)] bounded alkyl chains (C12RhMOP) than for the two Rh- once the steady state was reached (after about 3 h). The separation MOPs with 12 coordinatively attached alkyl chains selectivity was obtained as the ratio of CO2 and N2 permeances. At (HRhMOP(diz)12 and HRhMOP(oiz)12). The areas per least two membrane samples of each type were fabricated and molecule are also higher for C12RhMOP than for HRhMOP- measured to provide the corresponding error estimations. (diz)12 and HRhMOP(oiz)12. These results indicate that Rh- 27497 https://doi.org/10.1021/acsami.2c06123 ACS Appl. Mater. Interfaces 2022, 14, 27495−27506 ACS Applied Materials & Interfaces www.acsami.org Research Article Figure 3. BAM images obtained during Rh-MOP film compression at indicated surface pressures and the corresponding areas per molecule. MOPs functionalized with 12 alkyl chains assemble into denser BAM images obtained at this stage reveal the formation of films than those functionalized with 24 alkyl chains due to the thick films as evidenced by the brightness of the images. reduced steric hindrance of the former type of Rh-MOPs. On In order to further investigate the behavior of Rh-MOPs at the other side, the surface pressure for OHRhMOP starts to the air−water interface and determine if the assembled increase at an area per molecule close to 1.6 nm2/molecule and continuous films show some regular crystalline arrangement, reaches the maximum slope of the π−A curve between 1.2 and GIXD and GISAXS studies were performed. The two Rh- 1.0 nm2/molecule. Close to 20 mN/m, the slope of the MOPs with dodecyl chains, C12RhMOP and HRhMOP(diz)12, isotherm decreases and a pseudo-plateau can be observed. and that with 24 hydroxo groups, OHRhMOP, were first Finally, at areas below 0.45 nm2/molecule, the surface pressure studied in GIXD configuration (q range 0.55−1.8 Å−1). Very increases again markedly. The areas obtained for this Rh-MOP, similar results are obtained for the two Rh-MOPs with dodecyl even at low surface pressures, are lower than the expected size chains, at both tested pressures (5 and 10 mN/m) and after of a single molecule, revealing that, possibly, the formed film is collapse (Figures 4 and S2). For C12RhMOP and HRhMOP- not a homogeneous monolayer. (diz)12, two very broad humps are observed at ca. 0.76/0.71 and 1.29/1.21 Å−1, respectively. These are likely resulting from The BAM images obtained simultaneously to π−A the diffusion form factors of the Rh-MOPs, which in first isotherms (Figure 3) show thin film domains and uncovered approximation can be considered as core−shell spheres. water regions at low surface pressures for all tested Rh-MOPs. Indeed, a reasonably good simulation can be obtained with These domains merge and form homogeneous monolayers that either a core−shell or a core−double shell model and the completely cover the water surface at surface pressures above characteristics of the Rh-MOPs. In addition, a sharp peak is 10 mN/m for HRhMOP(diz)12 and 15 mN/m for HRhMOP- observed systematically for both C12RhMOP and HRhMOP- (oiz)12, while for C12RhMOP, it has been reported that it is (diz)12 at q = 1.51 Å−1, characteristic of ordered alkyl chains.21 necessary to increase the surface pressure up to 25 mN/m to Interestingly, the peak arises solely from in-plane scattering fully cover the water surface.15 However, OHRhMOP film (Figure S3), thus indicating that the chain organization a priori domains are brighter, which indicates that these domains are takes place parallel to the subphase. Indeed, the Bragg rod has thicker than those observed for the Rh-MOPs functionalized only a very limited qz extension (Figure S3), in disagreement with aliphatic chains, even at a very low surface pressure. This with the form factor of dodecyl chains21 or other classical indicates that the floating films obtained at low pressures are amphipathic molecules that extend perpendicular to the water not true monolayers but rather contain multilayer domains. At subphase. This in turn suggests that Rh-MOP molecules are a low surface pressure of ca. 2 mN/m, these surface aggregates forming a film in which they are softly connected through assemble into a film that covers the entire water phase. The interdigitation of some of their dodecyl chains. An additional 27498 https://doi.org/10.1021/acsami.2c06123 ACS Appl. Mater. Interfaces 2022, 14, 27495−27506 ACS Applied Materials & Interfaces www.acsami.org Research Article range of 0.27−0.42 Å−1 (see Figure 4). These peaks are reasonably reproduced, together with those at higher q, considering the diffusion form factor of core−double-shell spheres with an empty core (calculation made using Sas- View22); that is, with zero scattering length density (SLD), of 5 Å radius, a first dense shell of 10.5 Å, thickness corresponding to the Rh-ligand structure (SLD = 2 × 10−6 Å−2), and an outer less dense shell of 18 Å thickness corresponding to the alkyl arms (SLD = 0.6 × 10−6 Å−2). This is in line with the characteristic sizes of both Rh-MOPs, which have a core of ca. 28 Å diameter and a total size of up to 52 Å diameter considering extended dodecyl chains (and thus, a shell thickness of up to 24 Å). It is worth noting that the broad humps at ca. 0.4 and 0.7 Å−1 are in good agreement with SAXS data obtained on Cu(II) analogues of C12RhMOP with varying Figure 4. In situ GIXD (q > 0.55 Å−1) and GISAXS (q < 0.45 Å−1) alkyl chain lengths,16 thus supporting our model. On the data for the studied Rh-MOPs at the gas−water interface at 5 mN/m. contrary, our experimental observation of in-plane ordering of Code: C12RhMOP (■), OHRhMOP (○), HRhMOP(diz) 12 (+), and some of the alkyl chains in films of C12RhMOP and HRhMOP(oiz) 12 (×). The data have been shifted vertically for HRhMOP(diz)12 a priori contradicts the alkyl chain/Rh- clarity. The red line is the diffusion form factor of core−double shell spheres with an empty (SLD = 0) core of 5 Å radius, a first dense shell MOP cores/alkyl chains three-layer model used to simulate the of 10.5 Å thickness (SLD = 2 × 10−6 Å−2), and an outer less dense NR data, albeit limited to q < 0.1 Å−1, of the LB film of C18- shell of 18 Å thickness, considering a pinhole instrumental smearing δ CuMOP.16 Q/Q of 10%. This model was calculated using SasView.22 Note that it Altogether, the GIXD/GISAXS data supports the fact that has been scaled differently for the two data ranges. Vertical arrows Rh-MOP molecules with alkyl chains form a sort of glassy film highlight the peak at 1.51 Å−1 corresponding to alkyl chain ordering. at the air−water interface, with the Rh-MOPs maintaining motional degrees of liberty, thus having variable orientations weaker peak at ca. 1.68 Å−1 is also detected in the collapsed and inter-MOP separations. The film is formed at rather low film for C12RhMOP, which could indicate a herringbone pressure, probably as soon as inter-MOP interactions are arrangement of the dodecyl chains. The fact that this peak is favored, through interdigitation of the alkyl chains. Although not observed in the case of HRhMOP(diz)12 may be ascribed the chains show some in-plane order, this does not result in a to its lower density of dodecyl chains (12 vs 24). regular organization of the Rh-MOPs in the film. In the case of The GIXD data obtained for OHRhMOP are similar for the OHRhMOP, the experimental scattering data is likely the sum three pressures studied (1, 5, and 10 mN/m) but differ of diffraction arising from a crystalline arrangement in multi- significantly from those of the Rh-MOPs with dodecyl chains layer Rh-MOP aggregates, as pointed out by BAM (Figures 4 and S2). Four broad peaks are clearly distinguished observations and diffusion from Rh-MOPs in other, at ca. 0.57, 0.63, 0.77, and 0.87 Å−1, in addition to several unorganized, areas of the film. weaker and poorly resolved features in the 1.1−1.5 Å−1 range. 3.3. Rh-MOP Film Deposited onto Solid Substrates. While the diffusion form factor for a core−shell sphere of the Having characterized the Rh-MOP behavior at the air−liquid characteristics of OHRhMOP does coincide with the two interface, Rh-MOP films have been transferred onto solid stronger peaks at 0.63 and 0.87 Å−1 (Figure S4), the rest of substrates. The first step was the optimization of the features, and in particular the peaks at 0.57 and 0.77 Å−1, experimental conditions for the sequential deposition of cannot be reasonably predicted by this model. This could homogeneous monolayers in order to build Rh-MOP multi- mean that, in the case of OHRhMOP, the observed features layers with the desired thickness onto different substrates. are also in part resulting from a regular organization of the Rh- Previous studies carried out with C12RhMOP showed that MOPs. The corresponding higher order peaks are unfortu- LS Rh-MOP films can be used, for instance, as selective nately not scanned by GIXD due to the high background at materials in composite membranes for CO2 separation.15 LS low q. Nevertheless, GISAXS data for OHRhMOP do present deposition is especially suitable for this application since it is a relatively sharp peak at 0.28 Å−1, which is in reasonably good faster than LB deposition (which is highly time consuming for agreement with the expected Bragg peak arising from a close- the deposition of a large number of layers), and the selective packed arrangement of OHRhMOP, which has a size of ca. 28 film is deposited only on one side of the porous support, which Å. The peak at 0.57 Å−1 in GIXD can thus be ascribed to the is required to obtain asymmetric composite membranes. second order, altogether supporting the presence of domains of In this work, the LS films of HRhMOP(diz)12 and regularly organized OHRhMOP. HRhMOP(oiz)12 were first transferred onto quartz substrates HRhMOP(diz)12 and the Rh-MOP with shorter octyl chains at 20 mN/m, whereas OHRhMOP LS films were deposited at HRhMOP(oiz)12 were also studied in GISAXS configuration. 2 mN/m. All the films were characterized by AFM and UV−vis For both Rh-MOPs with alkyl chains, no low q peak is detected spectroscopy. that could correspond to the inter-MOP separation D in a The comparison of UV−vis spectra for LS monolayer films crystalline arrangement, typically expected in the range 0.14− of HRhMOP(diz)12 and HRhMOP(oiz)12 with those of 0.18 Å−1 considering Rh-MOP sizes in the range 40−52 Å in a dichloromethane solutions (Figure S5) shows a very good close to hexagonal packing and thus D = 2π/qxy(√3). This agreement between both types of films and solution spectra at supports the assumption that homogeneous films formed by wavelengths above the solvent cut-off, which confirms that film Rh-MOPs with alkyl substituents do not present crystalline formation does not affect the integrity of the materials. organization. Furthermore, two broad peaks are detected in the Additionally, AFM characterization of LS monolayers of 27499 https://doi.org/10.1021/acsami.2c06123 ACS Appl. Mater. Interfaces 2022, 14, 27495−27506 ACS Applied Materials & Interfaces www.acsami.org Research Article Figure 5. Representative AFM topography (top) and phase images (bottom) from HRhMOP(diz)12 and HRhMOP(oiz)12 LS films transferred onto quartz substrates at 20 mN/m. The observed scratches correspond to the defects on the surface of the quartz plates used (see Figure S7) since phase images confirm the homogeneity of the samples. Figure 6. UV−vis spectra from HRhMOP(diz)12 and HRhMOP(oiz)12 sequential deposition of LS films transferred onto quartz at 20 mN/m. HRhMOP(diz)12 and HRhMOP(oiz)12 (Figure 5) shows flat (diz)12 and HRhMOP(oiz)12. Moreover, AFM characterization and homogeneous films that completely cover quartz (Figure S8) shows films with an average thickness of ca. 3 nm substrates at 20 mN/m. The average thickness of these films and an rms roughness of ca. 2.3 nm due to the presence of was estimated by measuring different film defects and borders several defects of lateral size between 250 and 400 nm and with the AFM tip (Figure S6). HRhMOP(diz)12 monolayers heights between 30 and 50 nm, which confirms that this Rh- are ca. 2.5 nm thick, similarly to C12RhMOP films, whereas MOP without alkyl chains does not form homogeneous HRhMOP(oiz)12 monolayers are slightly thinner (ca. 2.3 nm), monolayers at the air−water interface. which is related to the shorter alkyl chain length for this The sequential deposition of multilayer films of HRhMOP- compound. The root-mean-square (rms) roughness of MOP (diz)12 and HRhMOP(oiz)12 at 20 mN/m onto quartz and monolayers was also evaluated using topography images. quartz crystal microbalance disks was then investigated. Figure HRhMOP(diz)12 and HRhMOP(oiz)12 films present a similar 6 shows the continuous increment of the film absorbance with rms roughness data (0.95 and 0.86 nm, respectively). the number of Rh-MOP monolayers deposited, without Remarkably, these values are even lower than the rms altering the position of the absorption bands. Moreover, the roughness of the uncovered quartz substrates used for film absorbance at the wavelength of maximum absorption (ca. 215 deposition (Figure S7), which is close to 1.3 nm, due to the nm) was plotted against the number of LS monolayers presence of defects on the surface of the quartz plates. deposited, showing a linear increase of the absorbance with the The OHRhMOP LS films transferred onto quartz substrates number of deposited layers, which demonstrates an almost show similar spectra to those in solution (Figure S5). constant Rh-MOP deposition in each transfer (Figure S9). However, the absorbance of the film is significantly higher Rh-MOP films deposited onto QCM disks were used to than the absorbance of the monolayer films of HRhMOP- determine the correlation between the deposited mass and the 27500 https://doi.org/10.1021/acsami.2c06123 ACS Appl. Mater. Interfaces 2022, 14, 27495−27506 ACS Applied Materials & Interfaces www.acsami.org Research Article number of monolayers transferred (see Figure S10). From the molecule, compared to pure OHRhMOP, is obtained when 15 slopes, almost constant mass depositions were obtained for diz ligands per OHRhMOP are spread on the water surface. each Rh-MOP monolayer transferred of HRhMOP(oiz)12 However, the areas per Rh-MOP are still far from the values (0.32 μg/cm2) and HRhMOP(diz)12 (0.35 μg/cm2). These obtained for alkyl-functionalized Rh-MOPs. When the amount values give a similar mean molecular area for both compounds of diz is increased up to 25 or 50 diz ligands per OHRhMOP, in LS films (ca. 4.4 nm2/molecule), which is smaller than the the area per Rh-MOP is also increased almost proportionally to areas obtained for the same Rh-MOPs at the air−water the increment of diz amount. Moreover, the shape of these interface (around 5.4 nm2/molecule). The area obtained from isotherms is quite similar to the isotherm of HRhMOP(diz)12, C12RhMOP multilayer films was significantly higher (5.8 nm2/ which seems to indicate that the chemical reaction at the air− molecule). Moreover, for this compound, the area was very water interface leads to the formation of an alkyl-functionalized close to the value at the air−water interface. This points to a Rh-MOP in a similar manner as observed in solution. BAM densification of multilayer films deposited onto solid substrates images were also acquired during film compression (Figure for the Rh-MOPs with 12 alkyl chains, whereas the films of the S11), revealing the formation of dense films at surface Rh-MOP with 24 alkyl chains retain almost exactly the same pressures above 15 mN/m, although more defects were superficial density obtained at the air−liquid interface. These observed when 50 diz ligands per OHRhMOP are spread. results, which are in line with the observation derived from the Consequently, the proportion 1:25 was chosen for further π−A isotherms, further support the relationship between film studies. density and the extent of aliphatic functionalization on the The films obtained after spreading 25 diz ligands per surface of the Rh-MOP. OHRhMOP molecule were transferred by LS deposition onto 3.4. Air−Water Interfacial Formation of Alkyl-Func- different substrates at 20 mN/m. Figure 8 shows a tionalized Rh-MOP Films. Preceding sections have shown representative AFM image of one LS film transferred onto a that ex situ aliphatic chain functionalization on the surface of quartz substrate. A continuous film with a thickness of ca. 2.8 Rh-MOPs enables their processing into ultrathin films through nm is obtained, although some aggregates of height up to 200 the LB technique. Such aliphatic chains can be incorporated on nm and lateral size between 400 and 600 nm can also be the surface of the Rh-MOP before the formation of the film observed. A plausible explanation is that these aggregates could through ligand design (C12RhMOP) or through coordinative be formed by unreacted OHRhMOP since this Rh-MOP has a post-synthetic modifications (HRhMOP(oiz) 1 2 and tendency to self-aggregate at the air−water interface as shown HRhMOP(diz)12). in the previous characterization of OHRhMOP films. On the other hand, OHRhMOP showed to aggregate upon Additional characterization of LS films transferred onto film formation. We envisaged that the degree of aggregation of Si(100) substrates was made by XPS. Figure 9 presents the OHRhMOP could be controlled by functionalizing OHRh- comparison of the XPS spectra of solid OHRhMOP and alkyl- MOP films in situ during film formation. With this aim, functionalized Rh-MOP films obtained by the OHRhMOP different amounts of separate solutions of the diz ligand and reaction with excess of diz (1:25 proportion) in THF or at the OHRhMOP were spread successively onto the air−water air−water interface. Previous literature studies23 indicate that interface. Figure 7 shows some representative π−A isotherms the Rh 3d high-resolution spectrum is composed of a doublet peak (3d5/2 and 3d3/2) with a spin−orbital coupling energy difference of 4.74 eV. The OHRhMOP spectrum shows these peaks at ca. 309.0 and 313.8 eV, in good agreement with the literature. However, the samples obtained after MOP reaction with excess diz present broader peaks, and the deconvolution of these peaks allows obtaining some interesting conclusions. The 3d5/2 peak deconvolution shows two contributions at ca. 308.7 and 310.1 eV, while the deconvolution of the 3d3/2 peak results in two bands at ca. 313.5 and 315.0 eV. The bands at 308.7 and 313.5 eV are reasonably close to the peaks obtained for OHRhMOP and can be ascribed to unmodified Rh paddlewheels, while the new peaks at ca. 310.1 and 315 eV present a significant shift of ca. 1.1 eV to a higher binding energy, which is likely a consequence of diz ligand Figure 7. Surface pressure−area (π−A) isotherms obtained for the coordination to Rh. Indeed, diz is a stronger donor than the imidazole ligand (diz) and different OHRhMOP/diz proportions (for O-donors, either H2O or MeOH, likely occupying the axial these systems, the area per molecule is calculated considering only the position of Rh paddlewheels in OHRhMOP. amount of Rh-MOP molecules). A HRhMOP(diz)12 isotherm has also been plotted for comparison (dashed line). Interestingly, the relative amount of Rh atoms with modified 3d bands is higher in the film obtained at the air−liquid interface, although the solution spectra obtained in THF obtained in this study. The diz ligand almost does not show confirm that all the Rh paddlewheels are coordinated to diz in any activity at the air−water interface, and the surface pressure solution. A plausible explanation to this fact can be the only rises slightly at very small surface areas, revealing that diz evaporation of THF after drop-cast deposition, which results in alone cannot form stable Langmuir films. a partial uncoordination of diz ligands (or exchange with water Given that the stoichiometric relationship OHRhMOP/diz molecules from the air) in the unstructured film, while LS films is 1:12 if all the axial sites are occupied by imidazole ligands, are densely arranged, preventing diz uncoordination. Overall, different OHRhMOP/diz proportions were tested. As shown XPS measurements confirm the in situ coordination of diz to in Figure 7, a significant increase of the area per Rh-MOP OHRhMOP at the water−air interface. 27501 https://doi.org/10.1021/acsami.2c06123 ACS Appl. Mater. Interfaces 2022, 14, 27495−27506 ACS Applied Materials & Interfaces www.acsami.org Research Article Figure 8. Representative AFM topography and phase images from an OHRhMOP + diz LS film transferred at 20 mN/m onto a quartz substrate. The film was prepared using a 1:25 OHRhMOP/diz proportion at the air−liquid interface. Then, multilayer LS films of the interfacially functionalized hydrophobicity, which can be consistent with a partial removal Rh-MOP were fabricated by sequential deposition of Langmuir of diz ligands from the film. films onto quartz and QCM substrates. After each deposition, The comparison of UV spectra of the samples with one the samples were washed with hexane with the aim of monolayer deposited onto a quartz substrate before and after eliminating, at least partially, the observed aggregates without HCl treatment shows that the band of maximum absorption at damaging the Rh-MOP film. Figure S12 shows the continuous ca. 215 nm is not affected by the acid treatment, which seems increment of the film absorbance and the surface mass density to indicate that the MOP polyhedral core surface density is with the number of layers deposited, respectively, onto quartz similar after the acid treatment (Figure S13). Note that the and QCM substrates. These results are very similar to the detachment of diz from the Rh-MOP films could not be behavior observed for alkyl-functionalized Rh-MOPs. In monitored by UV−vis. The π* → σ* transition of the Rh(II) particular, the increment of the absorbance per deposited paddlewheel (band I), which is sensitive to the coordination layer is almost the same than for HRhMOP(oiz)12, although environment of the Rh(II) axial site, is a forbidden transition the wavelength of the maximum absorbance is slightly shifted with low molar attenuation coefficient.25 The minute amount to 221 nm. Meanwhile, the deposited mass increases up to 0.44 of deposited Rh-MOP on the substrate makes not possible to μg/cm2 probably due to the unreacted OHRhMOP aggregates analyze the changes of band I on these samples. and diz chains that are present on the films and the denser Finally, the AFM images of a monolayer film deposited onto structure of the films derived from the π−A isotherm. Si(100) before and after acid treatment for 30 min were also 3.5. Post-Synthetic Modification of Alkyl-Function- obtained (Figure S14). Phase images show that the surface alized Rh-MOP Films. We have shown that it is possible to coverage is similar after acid treatment, although more functionalize Rh-MOP films at the air−water interface pinholes are present in the film, as can be observed in the coordinating a N-donor ligand (diz) to Rh dimers’ axial sites topography images. However, these defects scarcely modify the through an interfacial reaction. However, the reversible nature rms roughness of the samples, which increases from 0.29 to of the interaction between diz and the Rh-MOP suggests that it 0.41 nm (uncovered Si(100) substrates present a rms should also be possible to post-synthetically detach diz ligands roughness of 0.26 nm, see Figure S7). from previously assembled HRhMOP(diz)12 films. With this Altogether, these results point to a cleavage of the diz ligands aim, LS films of different thicknesses (one and three layers) of exposed to the air by acid treatment, while the interdigitated HRhMOP(diz)12, deposited onto quartz and Si(100) sub- alkyl chains are not affected, thus keeping a similar film strates, were exposed to HCl vapors at room temperature and coverage after the acid treatment. Therefore, we have shown subsequently washed with n-hexane. The acid treatment was that thin films made by coordinatively functionalized Rh- selected because it has been previously employed with success MOPs can be post-synthetically modified to tune their to detach N-donor ligands from Rh-MOPs in solution.24 The hydrophobicity, which in turn might impact on the film acid-treated films were characterized through UV adsorption applications. spectra, water contact angle (WCA), and AFM to analyze the 3.6. Rh-MOP Films in CO2/N2 Separation. HRhMOP- chemical and morphological changes induced by acid treat- (diz)12 LS multilayer films were deposited onto the permeable ment. polymer PTMSP in order to fabricate selective Rh-MOP layers Table 1 presents the WCA values of LS films deposited onto with the desired film thickness. Here, multilayer films formed Si(100) and quartz substrates before and after the treatment by 10−30 Rh-MOP monolayers were tested for CO2/N2 with HCl vapor. The deposition of HRhMOP(diz)12 ultrathin separation in post-combustion conditions (temperature 35 LS films significantly increases the hydrophobicity of the °C, feed pressures 1 and 3 bar, and CO2/N2 mixture substrates, reaching values around 91° on Si and 88° on quartz composition 10/90 in volume). CO2 permeance and CO2/ substrates, which suggest an efficient packing of the alkyl N2 selectivity average values are shown in Figure 10. The CO2/ chains from diz ligands on the upper side of Rh-MOP films. N2 selectivity of the PTMSP membrane at 1 bar is 4.1 and Only 2 min after being exposed to HCl vapors, the WCA of constantly increases with the deposition of HRhMOP(diz)12 these samples diminish between 10 and 18°. Longer treatment films up to values close to 13 when 30 Rh-MOP monolayers (30 min) does not change significantly the WCA values, except are deposited, while the CO2 permeance diminishes from 415 in the case of the monolayer film deposited on quartz that to ca. 120 GPU (at 1 bar). Comparing the performance of reduces to ca. 57°. These results reveal a decrease of film HRhMOP(diz)12 membranes to the values obtained for 27502 https://doi.org/10.1021/acsami.2c06123 ACS Appl. Mater. Interfaces 2022, 14, 27495−27506 ACS Applied Materials & Interfaces www.acsami.org Research Article Table 1. WCA Values (Average ± Standard Deviation) for Ultrathin Films of HRhMOP(diz)12 Transferred onto Si(100) or Quartz Substratesa WCA film after HCl treatment uncovered pristine film on sample substrate the substrate t = 2 min t = 30 min 1 layer 44.9 ± 0.4 91.6 ± 0.2 77.8 ± 0.6 76.0 ± 0.2 Si(100) 3 layers 91.2 ± 0.2 73.4 ± 0.3 73.1 ± 0.4 Si(100) 1 layer 15.4 ± 0.6 80.7 ± 0.2 72.9 ± 4.3 56.8 ± 0.5 quartz 3 layers 88.1 ± 0.3 78.1 ± 0.6 74.4 ± 1.4 quartz a For comparison purposes, uncoated substrates were also analyzed. separation. Figure 10 shows the CO2 permeance and CO2/N2 selectivity average values. The CO2/N2 selectivity increases with the number of layers deposited at the same time that the CO2 permeance diminishes, similarly to the results obtained for alkyl-functionalized MOPs. However, in this system, the changes between the samples containing 25 or 30 layers are more limited, and the deviations between different samples incorporating the same number of layers are smaller with 25 than with 30 layers. Interestingly, the CO2 permeance values at 1 and 3 bar are almost identical for the membranes obtained with the interfacially functionalized Rh-MOP and bigger than the permeances obtained for alkyl-functionalized Rh-MOPs, maintaining CO 2 /N 2 selectivities similar to that of C12RhMOP. This confirms that the interfacial reaction between diz ligands and MOPs that do not possess alkyl chains is an efficient method to obtain ultrathin MOP films that can be incorporated onto different substrates and opens an interesting methodology for the development of MOP-based films with different structures and functionalities. Although the performances of the MOP films fabricated herein are below the commercial target for post-combustion CO2 capture,26 it should be highlighted that they have been evaluated for the separation of CO2 from N 2 under experimental conditions relevant to post-combustion CO2 capture (10% CO2 and 90% N2 and two different feed Figure 9. XPS spectra of (a) OHRhMOP (powder), (b) 1 LS film pressures, 1 and 3 bar). Moreover, the performances of MOP deposited at 20 mN/m after OHRhMOP + diz (1:25) reaction at the films are comparable to those previously reported for LS films air−liquid interface, and (c) drop-cast film obtained after OHRhMOP fabricated with different polymers of intrinsic microporosity + diz (1:25) reaction in THF. The black lines represent the (PIMs),19 as shown in Table S2. In particular, the performance simulation of these spectra using CasaXPS and a Shirley-type of HRhMOP(diz)12 films is very similar to that of PIM films, background. The components depicted as colored lines are detailed both in terms of CO2 permeance and CO2/N2 selectivity. Also, in Table S1. C12RhMOP and interfacially functionalized MOP films present larger CO2 permeances than PIM films (up to 1.8 times C12RhMOP in the same conditions, the higher CO2/N2 larger), which leads to lower CO2/N2 selectivity values. selectivity values obtained for HRhMOP(diz)12 samples can Consequently, we have shown that the approach introduced be highlighted (maximum values for 30 monolayer films of in this study allows fabricating dense MOP films with tailorable C12RhMOP are close to 10), which logically leads to lower performance in CO2/N2 separation. Moreover, the LB CO2 permeance values (ca. 195 GPU for C12RhMOP). These technique is the only method reported up to the date that results are in good agreement with the denser packing of the allows obtaining pure MOP selective layers that can be used HRhMOP(diz)12 films discussed in previous sections and pave for gas separation. the way for the modulation of selectivity and permeance of MOP films in gas separations by controlling the number of 4. CONCLUSIONS alkyl chains attached to a given MOP polyhedral core. The formation and deposition of ultrathin films of Rh-based Additionally, multilayer films of the interfacially modified MOPs through the LB technique have been successfully OHRhMOP with diz ligands were also deposited onto PTMSP expanded to Rh-MOPs bearing alkyl substituents introduced substrates in order to investigate its performance in CO2/N2 through coordination at the Rh paddlewheel external sites, 27503 https://doi.org/10.1021/acsami.2c06123 ACS Appl. Mater. Interfaces 2022, 14, 27495−27506 ACS Applied Materials & Interfaces www.acsami.org Research Article Figure 10. CO2 permeance and CO2/N2 selectivity measurements of HRhMOP(diz)12 and OHRhMOP + diz samples transferred onto PTMSP. Measurements were made at 35 °C and at two different pressures (1 and 3 bar). Error bars were obtained from measurements on at least two different samples. C12RhMOP data previously reported by us15 have also been included for comparison. instead of those with substituents on the organic ligand of the MOP LS film deposition; UV−vis and AFM character- Rh-MOP previously reported. The obtained films are dense ization of MOP films deposited onto quartz substrates; and homogeneous monolayers, resulting from partial inter- Rh-MOP mass deposited onto QCM disks; BAM images digitation of alkyl chains. In situ modification of Rh-MOP films obtained during OHRhMOP + diz film formation; UV− can be done by either partially grafting alkyl substituents vis and AFM characterization of the post-synthetic through coordination at the air−water interface or by partially modification of MOP films; parameters used to simulate removing these through exposure of pre-formed films to acid the Rh 3d high-resolution XPS spectra of MOP films; vapors. Multilayered films of Rh-MOPs with alkyl substituents and comparison of the performance of MOP and PIM all act as selective layers for efficient CO2 separation, including ultrathin films in CO2/N2 separation (PDF) ■ those obtained by interfacial modification, thus opening a useful path to variations of MOP film structure and functionality. Overall, the results presented here enable to AUTHOR INFORMATION expand the composition of MOP-based films by incorporating Corresponding Author previously inaccessible MOPs (i.e., insoluble MOPs) and by Ignacio Gascón − Instituto de Nanociencia y Materiales de post-synthetic modification of the MOP-based films. Aragón (INMA), CSIC and Universidad de Zaragoza, ■ * ASSOCIATED CONTENT sı Supporting Information Zaragoza 50009, Spain; Departamento de Química Física, Universidad de Zaragoza, Zaragoza 50009, Spain; orcid.org/0000-0002-3492-6456; Email: igascon@ The Supporting Information is available free of charge at unizar.es https://pubs.acs.org/doi/10.1021/acsami.2c06123. Authors UV−vis characterization of OHRhMOP alkyl function- Inés Tejedor − Instituto de Nanociencia y Materiales de alization with diz in solution; GIXD characterization Aragón (INMA), CSIC and Universidad de Zaragoza, during MOP film formation; schematic representation of Zaragoza 50009, Spain; Departamento de Química Física, 27504 https://doi.org/10.1021/acsami.2c06123 ACS Appl. Mater. Interfaces 2022, 14, 27495−27506 ACS Applied Materials & Interfaces www.acsami.org Research Article Universidad de Zaragoza, Zaragoza 50009, Spain; 2017-0706). I.T. and M.P.-M. gratefully acknowledge their orcid.org/0000-0002-8267-9306 DGA PhD fellowship from Government of Aragon. The Miguel A. Andrés − Instituto de Nanociencia y Materiales de microscopy work was carried out in the Laboratorio de Aragón (INMA), CSIC and Universidad de Zaragoza, Microscopias Avanzadas at the Instituto de Nanociencia y Zaragoza 50009, Spain; Departamento de Química Física, Materiales de Aragon (LMA-INMA). This work benefited Universidad de Zaragoza, Zaragoza 50009, Spain; from the use of the SasView application, originally developed orcid.org/0000-0003-3691-3437 under NSF award DMR-0520547. SasView contains code Arnau Carné-Sánchez − Catalan Institute of Nanoscience and developed with funding from the European Union’s Horizon Nanotechnology (ICN2), CSIC and The Barcelona Institute 2020 research and innovation programme under the SINE2020 of Science and Technology, Barcelona 08193, Spain; project, grant agreement no. 654000. The authors thank the orcid.org/0000-0002-8569-6208 synchrotron SOLEIL for beamtime provision under projects Mónica Arjona − Departamento de Química Física, 20190435 and 20191874. Universidad de Zaragoza, Zaragoza 50009, Spain Marta Pérez-Miana − Instituto de Nanociencia y Materiales de Aragón (INMA), CSIC and Universidad de Zaragoza, ■ REFERENCES (1) Eddaoudi, M.; Kim, J.; Wachter, J. B.; Chae, H. K.; O’Keeffe, M.; Zaragoza 50009, Spain; Chemical and Environmental Yaghi, O. M. Porous Metal-Organic Polyhedra: 25 Å Cuboctahedron Engineering Department, Universidad de Zaragoza, Constructed from 12 Cu2(CO2)4 Paddle-Wheel Building Blocks. J. Zaragoza 50018, Spain Am. Chem. Soc. 2001, 123, 4368−4369. Javier Sánchez-Laínez − Instituto de Nanociencia y Materiales (2) Zhang, D.; Ronson, T. K.; Zou, Y.-Q.; Nitschke, J. R. Metal- de Aragón (INMA), CSIC and Universidad de Zaragoza, Organic Cages for Molecular Separations. Nat. Rev. Chem. 2021, 5, 168−182. Zaragoza 50009, Spain; Chemical and Environmental (3) Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M. The Engineering Department, Universidad de Zaragoza, Chemistry and Applications of Metal-Organic Frameworks. Science Zaragoza 50018, Spain; orcid.org/0000-0001-6627- 2013, 341, 1230444. 0079 (4) (a) Adil, K.; Belmabkhout, Y.; Pillai, R. S.; Cadiau, A.; Bhatt, P. Joaquín Coronas − Instituto de Nanociencia y Materiales de M.; Assen, A. H.; Maurin, G.; Eddaoudi, M. Gas/Vapour Separation Aragón (INMA), CSIC and Universidad de Zaragoza, Using Ultra-Microporous Metal-Organic Frameworks: Insights into Zaragoza 50009, Spain; Chemical and Environmental the Structure/Separation Relationship. Chem. Soc. Rev. 2017, 46, Engineering Department, Universidad de Zaragoza, 3402−3430. (b) Rojas, S.; Horcajada, P. Metal-Organic Frameworks Zaragoza 50018, Spain; orcid.org/0000-0003-1512- for the Removal of Emerging Organic Contaminants in Water. Chem. 4500 Rev. 2020, 120, 8378−8415. (5) Fang, X.; Zong, B. Y.; Mao, S. Metal-Organic Framework-Based Philippe Fontaine − Synchrotron SOLEIL, L’Orme des Sensors for Environmental Contaminant Sensing. Nano-Micro Lett. Merisiers, Gif-sur-Yvette 91192, France; orcid.org/0000- 2018, 10, 64. 0003-3394-6508 (6) Horcajada, P.; Chalati, T.; Serre, C.; Gillet, B.; Sebrie, C.; Baati, Michel Goldmann − Synchrotron SOLEIL, L’Orme des T.; Eubank, J. F.; Heurtaux, D.; Clayette, P.; Kreuz, C.; Chang, J.-S.; Merisiers, Gif-sur-Yvette 91192, France; Institut des Hwang, Y. K.; Marsaud, V.; Bories, P.-N.; Cynober, L.; Gil, S.; Férey, NanoSciences de Paris, UMR 7588 CNRS, Sorbonne G.; Couvreur, P.; Gref, R. Porous Metal-Organic Framework Université, Paris Cedex 05 75252, France Nanoscale Carriers as a Potential Platform for Drug Delivery and Olivier Roubeau − Instituto de Nanociencia y Materiales de Imaging. Nat. Mater. 2010, 9, 172−178. Aragón (INMA), CSIC and Universidad de Zaragoza, (7) (a) Gosselin, A. J.; Rowland, C. A.; Bloch, E. D. Permanently Zaragoza 50009, Spain; orcid.org/0000-0003-2095- Microporous Metal-Organic Polyhedra. Chem. Rev. 2020, 120, 8987− 9014. (b) Carné-Sánchez, A.; Craig, G. A.; Larpent, P.; Guillerm, V.; 5843 Urayama, K.; Maspoch, D.; Furukawa, S. A Coordinative Solubilizer Daniel Maspoch − Catalan Institute of Nanoscience and Method to Fabricate Soft Porous Materials from Insoluble Metal- Nanotechnology (ICN2), CSIC and The Barcelona Institute Organic Polyhedra. Angew. Chem. Int. Ed. 2019, 58, 6347−6350. of Science and Technology, Barcelona 08193, Spain; (c) Sánchez-González, E.; Tsang, M. Y.; Troyano, J.; Craig, G. A.; orcid.org/0000-0003-1325-9161 Furukawa, S. Assembling Metal-Organic Cages as Porous Materials. Complete contact information is available at: Chem. Soc. Rev. 2022, DOI: 10.1039/d1cs00759a. (8) Grancha, T.; Carné-Sánchez, A.; Hernández-López, L.; Albalad, https://pubs.acs.org/10.1021/acsami.2c06123 J.; Imaz, I.; Juanhuix, J.; Maspoch, D. Phase Transfer of Rhodium(II)- Based Metal-Organic Polyhedra Bearing Coordinatively Bound Cargo Notes Enables Molecular Separation. J. Am. Chem. Soc. 2019, 141, 18349− The authors declare no competing financial interest. 18355. ■ (9) Kawano, R.; Horike, N.; Hijikata, Y.; Kondo, M.; Carné-Sánchez, A.; Larpent, P.; Ikemura, S.; Osaki, T.; Kamiya, K.; Kitagawa, S.; ACKNOWLEDGMENTS Takeuchi, S.; Furukawa, S. Metal-Organic Cuboctahedra for Synthetic This work was funded by MCIN/AEI/10.13039/ Ion Channels with Multiple Conductance States. Chem 2017, 2, 393− 501100011033 and ERDF “A way of making Europe” (grant 403. PID2019-105881RB-I00). The authors also acknowledge the (10) Liu, X.; Wang, X.; Bavykina, A. V.; Chu, L.; Shan, M.; support from the Spanish MINECO (project RTI2018- Sabetghadam, A.; Miro, H.; Kapteijn, F.; Gascon, J. Molecular-Scale Hybrid Membranes Derived from Metal-Organic Polyhedra for Gas 095622-B-I00) and the Catalan AGAUR (project 2017 SGR Separation. ACS Appl. Mater. Interfaces 2018, 10, 21381−21389. 238). It was also funded by the CERCA Programme/ (11) Furukawa, S.; Horike, N.; Kondo, M.; Hijikata, Y.; Carné- Generalitat de Catalunya and through a fellowship (LCF/ Sánchez, A.; Larpent, P.; Louvain, N.; Diring, S.; Sato, H.; Matsuda, BQ/PR20/11770011) from “la Caixa” Foundation (ID R.; Kawano, R.; Kitagawa, S. Rhodium-Organic Cuboctahedra as 100010434). ICN2 is supported by the Severo Ochoa Porous Solids with Strong Binding Sites. Inorg. Chem. 2016, 55, programme from the Spanish MINECO (grant no. SEV- 10843−10846. 27505 https://doi.org/10.1021/acsami.2c06123 ACS Appl. Mater. Interfaces 2022, 14, 27495−27506 ACS Applied Materials & Interfaces www.acsami.org Research Article (12) Craig, G. A.; Larpent, P.; Kusaka, S.; Matsuda, R.; Kitagawa, S.; Furukawa, S. Switchable Gate-Opening Effect in Metal-Organic Polyhedra Assemblies through Solution Processing. Chem. Sci. 2018, 9, 6463−6469. (13) Hernández-López, L.; Martínez-Esaín, J.; Carné-Sánchez, A.; Grancha, T.; Faraudo, J.; Maspoch, D. Steric Hindrance in Metal Coordination Drives the Separation of Pyridine Regioisomers Using Rhodium(II)-Based Metal-Organic Polyhedra. Angew. Chem. Int. Ed. 2021, 133, 11507−11514. (14) Carné-Sánchez, A.; Albalad, J.; Grancha, T.; Imaz, I.; Juanhuix, J.; Larpent, P.; Furukawa, S.; Maspoch, D. Postsynthetic Covalent and Coordination Functionalization of Rhodium(II)-Based Metal-Organic Polyhedra. J. Am. Chem. Soc. 2019, 141, 4094−4102. (15) Andrés, M. A.; Carné-Sánchez, A.; Sánchez-Laínez, J.; Roubeau, O.; Coronas, J.; Maspoch, D.; Gascón, I. Ultrathin Films of Porous Metal-Organic Polyhedra for Gas Separation. Chem. Eur J. 2020, 26, 143−147. (16) Li, M.; Zhang, M.; Lai, Y.; Liu, Y.; Halbert, C.; Browning, J. F.; Liu, D.; Yin, P. Solvated and Deformed Hairy Metal-Organic Polyhedron. J. Phys. Chem. C 2020, 124, 15656−15662. (17) Al-Mohammed, N. N.; Duali Hussen, R. S.; Alias, Y.; Abdullah, Z. Tris-Imidazolium and Benzimidazolium Ionic Liquids: a New Class of Biodegradable Surfactants. RSC Adv. 2015, 5, 2869−2881. (18) Fontaine, P.; Ciatto, G.; Aubert, N.; Goldmann, M. Soft Interfaces and Resonant Investigation on Undulator Source: A Surface X-ray Scattering Beamline to Study Organic Molecular Films at the SOLEIL Synchrotron. Sci. Adv. Mater. 2014, 6, 2312−2316. (19) (a) Benito, J.; Sánchez-Laínez, J.; Zornoza, B.; Martín, S.; Carta, M.; Malpass-Evans, R.; Téllez, C.; McKeown, N. B.; Coronas, J.; Gascón, I. Ultrathin Composite Polymeric Membranes for CO2/N2 Separation with Minimum Thickness and High CO2 Permeance. Chemsuschem 2017, 10, 4014−4017. (b) Benito, J.; Vidal, J.; Sánchez- Laínez, J.; Zornoza, B.; Téllez, C.; Martín, S.; Msayib, K. J.; Comesaña-Gándara, B.; McKeown, N. B.; Coronas, J.; Gascón, I. The Fabrication of Ultrathin Films and their Gas Separation Performance from Polymers of Intrinsic Microporosity with Two-Dimensional (2D) and Three-Dimensional (3D) Chain Conformations. J. Colloid Recommended by ACS Interface Sci. 2019, 536, 474−482. (20) Furukawa, H.; Kim, J.; Plass, K. E.; Yaghi, O. M. Crystal Structure, Dissolution, and Deposition of a 5 nm Functionalized Water-Stable etb-MOFs for Methane and Carbon Dioxide Metal-Organic Great Rhombicuboctahedron. J. Am. Chem. Soc. 2006, Storage 128, 8398−8399. Constantinos Tsangarakis, Pantelis N. Trikalitis, et al. (21) Vollhardt, D.; Brezesinski, G.; Siegel, S.; Emrich, G. Phase MARCH 28, 2023 Transition in Adsorbed Monolayers of Sodium Dodecyl Sulfate/ INORGANIC CHEMISTRY READ Dodecanol Mixtures. J. Phys. Chem. B 2001, 105, 12061−12067. (22) SasView website. https://www.sasview.org/.(accessed on Feb Enhanced Thermoelectric Properties of a Semiconducting 23, 2022). Two-Dimensional Metal–Organic Framework via Iodine (23) Men, S.; Lovelock, K. R. J.; Licence, P. X-Ray Photoelectron Loading Spectroscopy as a Probe of Rhodium-Ligand Interaction in Ionic Liquids. Chem. Phys. Lett. 2016, 645, 53−58. Maria de Lourdes Gonzalez-Juarez, Iris Nandhakumar, et al. JANUARY 23, 2023 (24) Legrand, A.; Liu, L.-H.; Royla, P.; Aoyama, T.; Craig, G. A.; ACS APPLIED MATERIALS & INTERFACES READ Carné-Sánchez, A.; Urayama, K.; Weigand, J. J.; Lin, C.-H.; Furukawa, S. Spatiotemporal Control of Supramolecular Polymerization and Gelation of Metal-Organic Polyhedra. J. Am. Chem. Soc. 2021, 143, Multicomponent, Functionalized HKUST-1 Analogues 3562−3570. Assembled via Reticulation of Prefabricated Metal–Organic (25) Boyar, E. B.; Robinson, S. D. Rhodium(II) Carboxylates. Coord. Polyhedral Cavities Chem. Rev. 1983, 50, 109−208. Akim Khobotov-Bakishev, Daniel Maspoch, et al. (26) Merkel, T. C.; Lin, H. Q.; Wei, X. T.; Baker, R. Power Plant AUGUST 16, 2022 Post-Combustion Carbon Dioxide Capture: an Opportunity for JOURNAL OF THE AMERICAN CHEMICAL SOCIETY READ Membranes. J. Membr. Sci. 2010, 359, 126−139. Computational Exploration of a Metal(II) Catecholate- Functionalized UiO-66 Nanoporous Metal–Organic Framework for Effective NOx Capture Karuppasamy Gopalsamy, Guillaume Maurin, et al. SEPTEMBER 19, 2022 ACS APPLIED NANO MATERIALS READ Get More Suggestions > 27506 https://doi.org/10.1021/acsami.2c06123 ACS Appl. Mater. Interfaces 2022, 14, 27495−27506