www.acsami.org Research Article Ammonia Capture in Rhodium(II)-Based Metal−Organic Polyhedra via Synergistic Coordinative and H‑Bonding Interactions Arnau Carné-Sánchez,* Jordi Martínez-Esaín, Tanner Rookard, Christopher J. Flood, Jordi Faraudo, Kyriakos C. Stylianou,* and Daniel Maspoch* Cite This: ACS Appl. Mater. Interfaces 2023, 15, 6747−6754 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. ABSTRACT: Ammonia (NH3) is among the world’s most widely produced bulk chemicals, given its extensive use in diverse sectors such as agriculture; however, it poses environmental and health risks at low concentrations. Therefore, there is a need for developing new technologies and materials to capture and store ammonia Downloaded via 79.152.129.231 on March 2, 2023 at 12:00:24 (UTC). safely. Herein, we report for the first time the use of metal−organic polyhedra (MOPs) as ammonia adsorbents. We evaluated three different rhodium-based MOPs: [Rh2(bdc)2]12 (where bdc is 1,3-benzene dicarboxylate); one function- alized with hydroxyl groups at its outer surface [Rh2(OH-bdc)2]12 (where OH- bdc is 5-hydroxy-1,3-benzene dicarboxylate); and one decorated with aliphatic alkoxide chains at its outer surface [Rh2(C12O-bdc)2]12 (where C12O-bdc is 5- dodecoxybenzene-1,3-benzene dicarboxylate). Ammonia-adsorption experiments revealed that all three Rh-MOPs strongly interact with ammonia, with uptake capacities exceeding 10 mmol/gMOP. Furthermore, computational and exper- imental data showed that the mechanism of the interaction between Rh-MOPs and ammonia proceeds through a first step of coordination of NH3 to the axial site of the Rh(II) paddlewheel cluster, which triggers the adsorption of additional NH3 molecules through H-bonding interaction. This unique mechanism creates H-bonded clusters of NH3 on each Rh(II) axial site, which accounts for the high NH3 uptake capacity of Rh-MOPs. Rh-MOPs can be regenerated through their immersion in acidic water, and upon activation, their ammonia uptake can be recovered for at least three cycles. Our findings demonstrate that MOPs can be used as porous hosts to capture corrosive molecules like ammonia, and that their surface functionalization can enhance the ammonia uptake performance. KEYWORDS: metal−organic polyhedra (MOPs), cages, ammonia capture, molecular dynamics, regeneration 1. INTRODUCTION researchers have introduced functional groups with high Ammonia (NH3) is industrially produced at an estimated rate affinity toward ammonia, including acidic and Lewis-acidic of ca. 175 million tons per year due to its vital role in the groups, into porous materials such as metal−organic frame- works (MOFs) and covalent-organic frameworks to yield development of fertilizers, which are crucial to securing the adsorbents with much higher ammonia storage capacity (up to food supply chain.1 Additionally, ammonia, which is gaseous at 20 mmol/g) than that of the previously mentioned room temperature, is increasingly being postulated as an materials.9−15 alternative combustion fuel, due to its high energy density Metal−organic polyhedra (MOPs) are a sub-class of (12.7 MJ/L), and as an alternative source for hydrogen, due to coordination cages16,17 that can store molecules within their its high gravimetric (17.7%) and volumetric (0.105 kg/L) H2 cavities, both in the solid state and in solution. Thus, MOPs energy density.2 That said, ammonia is also a pollutant that combine the merits of designed porosity exhibited by reticular threatens the environment and human health at a concen- materials with the solution processability of monomeric tration as low as 25 ppm.3 Therefore, the development of compounds. These features make MOPs an emerging class materials for the safe capture and storage of ammonia is highly of adsorbents18 that have already demonstrated promise desirable to enable the use of NH3 as an alternative fuel and to mitigate its harmful environmental and health effects. Development of stable adsorbents for ammonia has been Received: October 28, 2022 hindered by its corrosiveness. Initially proposed materials, Accepted: December 7, 2022 based on inorganic porous substances such as zeolites, Published: January 25, 2023 activated carbon, or porous silica, exhibit moderate adsorption capacity (7−9 mmol/g) and lack the structural and composi- tional versatility for further optimization.4−8 Recently, © 2023 The Authors. Published by American Chemical Society https://doi.org/10.1021/acsami.2c19206 6747 ACS Appl. Mater. Interfaces 2023, 15, 6747−6754 ACS Applied Materials & Interfaces www.acsami.org Research Article toward the storage and capture of gases such as methane,19 carbon dioxide,20 carbon monoxide,21 nitric oxide,21 and sulfur dioxide.22 However, given the inherent fragility of most MOPs, which are based on labile coordination bonds [i.e., Cu(II)- carboxylate and Pd(II)-pyridine], they have not been explored as potential porous hosts for ammonia.23 Herein, we introduce the use of MOPs24 as adsorbents for ammonia capture. To overcome the limitations related to MOP fragility, we have employed robust cuboctahedral Rh(II)-based MOPs of the general formula [Rh2(bdc)2]12 (where bdc is 1,3-benzene dicarboxylate) as porous media for the uptake of ammonia.21 The presence of 12 Rh(II) paddlewheel clusters in the structure of cuboctahedral Rh- MOPs endows them with high chemical and structural stability. Additionally, the desolvation of Rh-MOPs generates open-metal sites at the axial position of the Rh(II) paddlewheel cluster, which can serve as Lewis acid sites to bind NH3 molecules.25 Thus, Rh-MOPs boast several features that favor the design of adsorbents for ammonia, including micro- porosity, versatile surface functionalization, and a high density of open-metal sites. We assessed the impact of all these attributes by measuring the ammonia-uptake capacity of the archetypical [Rh2(bdc)2]12 MOP unit (hereafter named H- RhMOP) and two additional surface functionalized MOPs; one functionalized with hydroxyl groups [Rh2(OH-bdc)2]12 (hereafter named OH-RhMOP, where OH-bdc is 5-hydroxy- 1,3-benzene dicarboxylate)26 and another decorated with aliphatic alkoxide chains [Rh2(C12O-bdc)2]12 (hereafter named C12-RhMOP where C12O-bdc is 5-dodecoxybenzene- 1,3-benzene dicarboxylate) (Figures 1 and S1−S4). 27 Ammonia-adsorption experiments revealed that all three Rh- Figure 1. Structures of the three Rh-MOPs included in this study: H- MOPs strongly interact with ammonia, with uptake capacities RhMOP (a), OH-RhMOP (b), and C12-RhMOP (c). Color code: exceeding 10 mmol/g. Through computations and experi- carbon (gray); hydrogen (white); oxygen (red); and rhodium (blue). ments, we revealed that the interaction between Rh-MOPs and ammonia proceeds through a first step of coordination of NH3 to the axial site of the Rh(II) paddlewheel cluster, which triggers the adsorption of additional NH3 molecules through H-bonding interactions. This unique mechanism leads to the generation of H-bonded clusters of NH3 on each Rh(II) axial site, which accounts for the high NH3 uptake capacity of Rh- MOPs. 2. RESULTS AND DISCUSSION 2.1. Ammonia Adsorption in H-RhMOP: Uptake and Mechanism. We began the ammonia sorption experiments using H-RhMOP (Figures 2 and S5). The total NH3 uptake at 1 bar for this MOP was 12.9 mmol/g [81.6 NH3 molecules per Rh-MOP; 3.4 NH3 molecules per Rh(II) site], which is in the range of high-performing MOFs and other extended materials Figure 2. Ammonia-adsorption (solid dots) and desorption (outlined (Table S1).4,6,7,9,10,12−14,28−37 The ammonia-uptake profile dots) at 298 K of pristine activated H-RhMOP (black), H-RhMOP resembles a type-I isotherm and shows a steep increase in the after the first NH3-adsorption isotherm activated under vacuum uptake at low pressure. The desorption branch exhibited an (blue), and H-RhMOP after the first NH3-adsorption isotherm activated under vacuum and heat (130° C). open-loop hysteresis, in which only 26% of the adsorbed NH3 molecules had been desorbed at low pressure. Consequently, the performance of H-RhMOP on a second consecutive Based on the high initial uptake of ammonia by H-RhMOP, adsorption cycle dropped significantly, from the initial value of the presence of a large hysteresis loop in the desorption 12.9 mmol/g down to 3.6 mmol/g. Intriguingly, the uptake branch, and the difficulty to fully regenerate H-RhMOP after performance of H-RhMOP could only be partially recovered loading it with ammonia, we postulated that strong chemical after the NH3-loaded sample had been heated at 130 °C under interactions between the MOP and the ammonia have vacuum for 12 h: the resulting NH3 uptake at 1 bar was 6.3 occurred. We reasoned that the Lewis-acidic character of the mmol/g, which corresponds to 48% of the initial ammonia Rh(II) axial sites could play an essential role in such uptake (Figure 2). interactions, as previously observed for other gases with 6748 https://doi.org/10.1021/acsami.2c19206 ACS Appl. Mater. Interfaces 2023, 15, 6747−6754 ACS Applied Materials & Interfaces www.acsami.org Research Article Figure 3. Structures obtained from DFT geometry optimizations of Rh2(Ac)4 in the presence of 1 (a), 2 (b), 3 (c), and 4 (d) molecules of NH3 shown in ball and stick representation. Color code: carbon (black); hydrogen (white); nitrogen (blue); and oxygen (red). The dashed lines indicate distances involved in the adsorption interactions (Rh−N: ∼2.0 Å, O···H: ∼2.0−2.3 Å, and N···H: ∼1.9−2.0 Å). (e) Top view of the same structure shown in (d) in order to appreciate the structure of the H-bond network between the four adsorbed NH3 molecules (atoms shown as spheres with their van der Waals size). Figure 4. Results from MD simulation of a H-RhMOP in the presence of different amounts of NH3 at 298 K. Top row: screenshots of the instantaneous configurations obtained after equilibration in a simulation box of 11.4 (a), 28.4 (b), 45.7 (c), and 57.1 molecules/nm3 (d). For simplicity, only the NH3 molecules closer to the H-RhMOP are shown. The NH3 molecules and the Rh atoms are shown with its van der Waals size, and the rest of the H-RhMOP structure is shown as balls and sticks. Color code: coordinated NH3 molecules (pink); H-bonded NH3 molecules (cyan); and rhodium (green). Middle row: zoomed-in images of the corresponding environments of the Rh(II) paddlewheel cluster in each simulation experiment. Bottom row: detailed chemical environment of the Rh(II) paddlewheel cluster for each of the simulated systems. electron lone pairs, such as CO and NO.21,38 The first evidence site.40 Thus, the λmax of H-RhMOP shifted from 598 to 540 nm of the interaction of ammonia with the Rh(II) paddlewheel upon NH3 adsorption, indicating that the axial site of the cluster was provided by the direct observation of the sample Rh(II) paddlewheel clusters was indeed coordinated to NH3 before and after the NH3 sorption experiment. The sample (Figure S6). However, we also reasoned that the coordination became purple upon adsorption, which is indicative that the of a single NH3 molecule to each Rh(II) axial site alone could coordination environment of Rh(II) is different upon NH3 not explain the high amount of ammonia that was irreversibly loading, with the ammonia to be coordinated to the Rh(II) bound to H-RhMOP, which reached 3.4 molecules of NH3 per paddlewheel cluster through its N atom (Figure S6).39 We Rh(II) site at 1 bar (Figure S5). acquired further evidence of the Rh−N coordination through To elucidate the role of the Rh(II) paddlewheel cluster in diffusive reflectance UV−vis spectra by monitoring the the irreversible uptake of ammonia by H-RhMOP and the spectroscopic changes of the adsorption bands centered at interactions involved in this process at the atomistic level, we 500−600 nm (λmax), which is ascribed to the π* → σ* have performed density functional theory (DFT) geometry transitions of the Rh−Rh bond, which is highly sensitive to optimizations and molecular dynamics (MD) simulations. For alterations in the coordination environment of the Rh(II) axial our DFT calculations, we employed Gaussian 16 at the M06- 6749 https://doi.org/10.1021/acsami.2c19206 ACS Appl. Mater. Interfaces 2023, 15, 6747−6754 ACS Applied Materials & Interfaces www.acsami.org Research Article L/SDD level of theory, appropriate for the description of metal per Rh(II) site was reached (Figure 4b−d and Table S5). organic systems containing Rh (see the Supporting Informa- Interestingly, using Fourier transform infrared (FTIR) spec- tion for details).41,42 Instead of the full Rh(II)-based MOPs troscopy, we found experimental evidence that the carboxylic (which is too big for DFT calculations), we have considered group of the BDC ligand participates in the H-bonding Rhodium acetate [Rh2(Ac)4] as a surrogate. We have network of the NH3 molecules. The carboxylic region (C� calculated the optimized structures obtained after interaction Ostr) of the FTIR spectrum of the NH3-loaded H-RhMOP was of [Rh2(Ac)4] with increasing numbers of NH3 molecules (up shifted by ca. 12 cm−1 relative to the pristine sample, a value to four molecules per Rh(II) site; see Figure 3). Our results compatible with the O atoms of the carboxylic groups show that a single NH3 molecule coordinates to the axial site of contributing to formation of the H bonds (Figure S12), as the Rh(II) paddlewheel cluster with a N−Rh distance of 2.2 Å predicted by our simulations.12 and an estimated bonding energy of −31.8 kcal/mol (Figures The results of the DFT calculations and MD simulations 3a, S7 and Table S2). Subsequent addition of NH3 molecules indicate that adsorption of ammonia in H-RhMOP entails a to the complex is favored by approximately 10 kcal/mol, two-step mechanism that involves two different interactions. through H-bonding interactions that involve these additional The first step, occurring at low ammonia pressure, involves the NH3 and the oxygen atoms of the acetate ligand, with O···H coordination of one NH3 molecule to both axial sites of the distance of ∼2.2 Å (Figures 3 and S7). The adsorption of up to Rh(II) paddlewheel clusters of Rh-MOP. In the second step, a four NH3 molecules per Rh(II) site is possible due to the NH3 molecule coordinated with Rh(II) nucleates the formation of a coordination-assisted H-bonding network of adsorption of additional NH3 molecules, which are stabilized NH3 molecules: one NH3 molecule is coordinated to the axial through two types of H-bonds: N−H···H interactions between Rh(II) site, and it is surrounded by three H-bonded NH3 coordinated and non-coordinated NH3 and N−H···O�C molecules that are 120° apart from each other with N···H interactions between non-coordinated NH3 and BDC linkers. distance of ∼2.0 Å (Figures 3e and S8). It has to be noted that This second steps occurs at higher NH3 densities and the capacity of ammonia to form such H-bonded ammonia consequently at higher NH3 pressure. This mechanism would clusters in the liquid and solid state has been previously be responsible for the uptake of up to four molecules of NH3 described through computational and experimental meth- per Rh(II) site, as seen in our MD simulations, which ods.43,44 In our case, the initial Rh(II)-coordinated NH3 translates into a theoretical adsorption capacity of 15.0 mmol/ molecule is what serves as the seed to nucleate the formation g for H-RhMOP�a value that is close to the experimental of such clusters. Remarkably, DFT calculations showed that value of 12.9 mmol/g. The high stabilizing energies predicted this initial Rh(II)−NH3 coordination occurs even in the by our DFT calculations justify the irreversible adsorption of presence of water, highlighting the high affinity of Rh(II) NH3 in H-RhMOP experimentally found in the NH3- paddlewheel clusters toward ammonia (Figure S9). Overall, adsorption/desorption isotherms. DFT calculations did not reveal any destabilization, breakage, Based on the values above, we concluded that the uptake of or desymmetrization of Rh2(Ac)4 upon addition of ammonia, ammonia by H-RhMOP is dominated by the chemistry of the thus further corroborating the high chemical stability of this Rh(II) paddlewheel cluster and does not involve any metal−organic complex. significant contribution of physisorbed NH3 molecules within Next, we employed the results from DFT calculations to the cavities of H-RhMOP. We pondered whether the high parametrize appropriate force fields (see the Supporting ammonia uptake in H-RhMOP could be achieved without Information) for performing MD simulations between the structuring Rh(II) paddlewheel clusters in a porous cage. To entire H-RhMOP and NH3 molecules. The simulations were test this hypothesis, we checked the ammonia-adsorption designed, performed, and analyzed using NAMD and VMD uptake capacity of Rh2(Ac)4. Interestingly, we found that software (see full details of the simulations in the Supporting Rh2(Ac)4 could adsorb 1.5 mmol/g of NH3 [0.3 NH3 Information).45,46 Briefly, the simulated systems consisted of a molecules per Rh(II) site] at 1 bar, which we attributed to H-RhMOP placed in the middle of a simulation box containing the lack of diffusion pathways throughout the sample, as this a fixed amount of NH3 molecules in the gas phase, would disfavor interactions between NH3 molecules and the thermostated at 298 K. We considered simulations with the potentially available Rh(II) sites (Figure S13). Furthermore, same amount of molecules but different volumes, correspond- we reasoned that close packing between adjacent Rh2(Ac)4 in ing to different gas densities and consequently mimicking the solid state would not leave any space to establish the H- different points of the experimental adsorption isotherms. We bonding network of the NH3 molecules organized around the observed that even for those simulations with the largest coordinated NH3. Therefore, we concluded that cavities and volume (lowest gas density), each Rh(II) axial site was diffusion pathways are essential to generate the coordination- occupied with one coordinated NH3 molecule (Figures 4a and induced H-bonding network of NH3 molecules at the Rh(II) S10). Additionally, 60% of the Rh(II) paddlewheel clusters axial sites. contained one additional NH3 molecule, which was stabilized 2.2. Regeneration of H-RhMOP. We targeted the through H-bonding interactions (Table S5). Thus, at low regeneration of the Rh(II) axial sites of H-RhMOP through ammonia pressure, the average number of NH3 per Rh(II) site a chemical method based on a ligand-exchange reaction with was 1.6. This result agrees with the experimental value of 1.5 water, which benefits from the high hydrolytic stability of Rh- NH3 per Rh(II) site that we had previously found at low MOPs, even at low pH.47 Thus, NH3-loaded H-RhMOP was ammonia pressure (19.7 mmHg, 0.03 bar) (Figure S5). immersed in acidic water (pH = 2) for 5 min, after which the Further shrinking of the simulation box induced adsorption sample changed from purple to green, indicating that the of more NH3 molecules to the Rh(II) sites, via H-bonding ligand at the axial site of the Rh(II) paddlewheel cluster had interactions with the already-coordinated NH3 molecule and exchanged from an N-donor (i.e., NH3) to an O-donor (i.e., with the O atom of the carboxylic group of the BDC ligand, H2O) (Figure S14). Low pH accelerates this process as it until an equilibrium configuration of ca. four NH3 molecules promotes the formation of non-coordinating NH4+ cations. 6750 https://doi.org/10.1021/acsami.2c19206 ACS Appl. Mater. Interfaces 2023, 15, 6747−6754 ACS Applied Materials & Interfaces www.acsami.org Research Article Evidence of ligand exchange was acquired from diffusive reflectance UV−vis spectra. The spectrum obtained after incubating the NH3-loaded H-RhMOP in acidic water revealed a λmax of 593 nm, which confirmed the release of the NH3 molecule from the axial site of Rh(II) (Figure S6). Once the axial site of the H-RhMOP was regenerated, we investigated the efficiency of the regeneration process, by measuring the second cycle of ammonia adsorption. To this end, the activated H-RhMOP was heated at 130 °C under vacuum for 12 h, and the NH3-adsorption isotherm was collected. As depicted in Figure 5, the ammonia uptake of the Figure 5. Ammonia-adsorption (solid dots) and desorption (outlined dots) at 298 K of pristine activated H-RhMOP (black) and regenerated H-RhMOP (red). Inset: ammonia uptake of H-RhMOP at 1 bar and 298 K for each of the five successive ammonia- adsorption/aqueous-regeneration cycles. regenerated sample was 12.1 mmol/g, which corresponds to an overall regeneration efficiency of 94%. This chemical and thermal regeneration method was repeated for five consecutive cycles, and no significant loss in the ammonia-uptake capacity of H-RhMOP was observed (Figure 5, inset). 2.3. Ammonia Adsorption in Surface-Functionalized Figure 6. (a) Ammonia-adsorption (solid dots) and desorption Rh-MOPs. Based on our proposed mechanism by which the (outlined dots) at 298 K of pristine activated OH-RhMOP. Inset: basic cuboctahedral H-RhMOP unit interacts with NH3 ammonia uptake of OH-RhMOP at 1 bar and 298 K for the two molecules, we then evaluated the impact of surface successive NH3-adsorption/aqueous−regeneration cycles. (b) Am- functionality on ammonia uptake by measuring the NH3 monia-adsorption (solid dots) and desorption (outlined dots) at 298 sorption of two differently surface-functionalized Rh-MOPs: K of pristine activated C12-RhMOP. Inset: ammonia uptake of C12- OH-RhMOP and C12-RhMOP. As depicted in Figure 6, OH- RhMOP at 1 bar and 298 K for the two successive NH3-adsorption/ aqueous−regeneration cycles. RhMOP and C12-RhMOP each exhibited a type-I adsorption isotherm, with maximum values for ammonia uptake (at 1 bar) of 14.7 mmol/g [100.8 NH3 molecules per Rh-MOP/4.2 NH3 confirmation of the integrity of these Rh-MOPs after two molecules per Rh(II) site] and 10.7 mmol/g [115.2 NH3 consecutive NH3-adsorption/aqueous-regeneration cycles molecules per Rh-MOP/4.8 NH3 molecules per Rh(II) site], (Figure S17). Interestingly, the lower BET surface areas respectively. Moreover, both functionalized Rh-MOPs pre- (SBET) of OH-RhMOP (SBET = 480 m2/g) and of C12-RhMOP sented a large, open-loop hysteresis in the corresponding (non-porous to N2) relative to H-RhMOP (SBET = 940 m2/g) desorption branch. As with H-RhMOP, the performance could do not correlate to the ammonia-uptake capacity of these only be recovered after the water and thermal treatments materials. Conversely, the ammonia uptake for both OH- (Figures S15 and S16). The overall regeneration efficiencies RhMOP (4.2 NH3 molecules per Rh(II) site) and C12- were 83% for OH-RhMOP and 96% for C12-RhMOP (Figure RhMOP (4.8 NH3 molecules per Rh(II) site) exceeded that 6). We ascribed the lower recovery efficiency of OH-RhMOP for H-RhMOP (3.4 NH3 molecules per Rh(II) site. We to the presence of acidic hydroxyl protons at the surface, which attributed this difference to the presence of available H-donors can protonate ammonia to generate ammonium phenolate, (hydroxyl groups, in OH-RhMOP) or H-acceptors (ether which could be more challenging to regenerate compared to groups, in C12-RhMOP) at the periphery of the functionalized H-bonded and Rh(II)-coordinated NH3 molecules.32,48,49 Rh-MOPs. Thus, we concluded that the uptake of function- Matrix-assisted laser desorption ionization time-of-flight alized Rh-MOPs could be increased via peripheral H-bonding measurements of both functionalized Rh-MOPs enabled interactions between the surface functional groups and NH3 6751 https://doi.org/10.1021/acsami.2c19206 ACS Appl. Mater. Interfaces 2023, 15, 6747−6754 ACS Applied Materials & Interfaces www.acsami.org Research Article molecules. This interpretation was supported by additional MD simulations performed for functionalized Rh-MOPs ■ AUTHOR INFORMATION Corresponding Authors (Figures S18 and S19). We found that the activities of the Arnau Carné-Sánchez − Catalan Institute of Nanoscience and Rh(II) sites in OH-RhMOP and in C12-RhMOP are equivalent Nanotechnology (ICN2), CSIC, and Barcelona Institute of to that which we found for H-RhMOP. Thus, each Rh(II) site Science and Technology, 08193 Bellaterra, Barcelona, Spain; can take up to four molecules of NH3, through the Departament de Química, Facultat de Ciències, Universitat coordination-induced H-bonding mechanism described Autònoma de Barcelona, 08193 Bellaterra, Spain; above. Furthermore, functionalized Rh-MOPs showed addi- orcid.org/0000-0002-8569-6208; Email: arnau.carne@ tional H-bonding interactions between their external surface icn2.cat groups and their adsorbed NH3 molecules, which would Kyriakos C. Stylianou − Materials Discovery Laboratory account for their superior ammonia-uptake relative to H- (MaD Lab), Department of Chemistry, Oregon State RhMOP (Figures S20 and S21). Overall, these results coincide University, Corvallis, Oregon 97331-4003, United States; with the observed trend in MOFs containing open-metal sites, orcid.org/0000-0003-1670-0020; in which the interaction strength of binding sites is a better Email: kyriakos.stylianou@oregonstate.edu predictor of ammonia uptake than is the surface area.10,13 Daniel Maspoch − Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC, and Barcelona Institute of 3. CONCLUSIONS Science and Technology, 08193 Bellaterra, Barcelona, Spain; In summary, we have reported the first-ever study on MOPs Departament de Química, Facultat de Ciències, Universitat for the uptake of gaseous ammonia. We investigated three Autònoma de Barcelona, 08193 Bellaterra, Spain; ICREA, isostructural cuboctahedral Rh-MOPs with different surface 08010 Barcelona, Spain; orcid.org/0000-0003-1325- functionalities, all interacting strongly with NH3 molecules 9161; Email: daniel.maspoch@icn2.cat through their Rh(II) axial sites. We unveiled the mechanism Authors that underpins the interaction between Rh-MOPs and NH3 Jordi Martínez-Esaín − Catalan Institute of Nanoscience and through experiments and simulations, finding an NH 3 Nanotechnology (ICN2), CSIC, and Barcelona Institute of molecule that initially interacts with the Rh(II) paddlewheel Science and Technology, 08193 Bellaterra, Barcelona, Spain; cluster by coordinating with their axial sites. The coordinated orcid.org/0000-0002-8420-8559 NH3 molecule then templates the subsequent addition of NH3 Tanner Rookard − Materials Discovery Laboratory (MaD molecules, which are stabilized through H-bonding inter- Lab), Department of Chemistry, Oregon State University, actions. This coordination-induced H-bonding network of Corvallis, Oregon 97331-4003, United States NH3 on top of each Rh(II) axial site increases the number of Christopher J. Flood − Materials Discovery Laboratory NH3 molecules that can be taken up per Rh(II) paddlewheel (MaD Lab), Department of Chemistry, Oregon State cluster, which, at the saturation point, reaches eight NH3 University, Corvallis, Oregon 97331-4003, United States molecules per Rh(II) paddlewheel cluster. The coordination- Jordi Faraudo − Institut de Ciència de Materials de Barcelona induced NH3-cluster formation that we have described here is (ICMAB-CSIC), 08193 Bellaterra, Spain; orcid.org/ consistent with previously postulated ammonia-adsorption 0000-0002-6315-4993 mechanisms for some of the best-performing MOFs such as M2Cl2BBTA [where M is Co(II), Ni(II), and Cu(II) and BTA Complete contact information is available at: https://pubs.acs.org/10.1021/acsami.2c19206 is 1H,5H-benzo(1,2-d),(4,5-d′)bistriazole]50 and MFM-300 MOF.30 However, those MOFs relay on high-nuclearity Notes clusters to trigger the formation of NH3 clusters, Rh-MOPs instead depend on low-nuclearity Rh(II) paddlewheel clusters The authors declare no competing financial interest. as NH3-binding sites, thus highlighting the uniqueness of this class of MOPs. Furthermore, we discovered that the overall ammonia-uptake efficiency of Rh-MOPs could be further ■ ACKNOWLEDGMENTS This work was supported by the Spanish MINECO (project enhanced by the judicious use of surface functional groups that PID2021-124297NB-C33); the Catalan AGAUR (project can establish H-bonding interactions, such as hydroxyls (as H- 2017 SGR 238); the CERCA Program/Generalitat de donors) or ethers (as H-acceptors). Overall, our results Catalunya; the MCIN/AEI/10.13039/501100011033; and by demonstrate the viability of using discrete molecular cages as the European Union “NextGenerationEU”/PRTR (EUR2020- porous hosts for capture of ammonia. We believe that the 112294). ICN2 is supported by the Severo Ochoa program capacity of MOPs to be processed into materials such as from the Spanish MINECO (grant SEV-2017-0706). We also films,51 gels, and monoliths,52 or even porous solutions53 thank the “Severo Ochoa” Program for Centers of Excellence should contribute to developing a new class of shaped in R&D Grant CEX2019-000917-S awarded to ICMAB. The adsorbents for ammonia. project that generated these results received support from the “la Caixa” Foundation (ID 100010434), through a fellowship ■ * ASSOCIATED CONTENT sı Supporting Information (LCF/BQ/PR20/11770011). We thank CESGA Supercom- puting center for technical support and computer time at the supercomputer Finisterrae III. K.C.S. thanks the Department The Supporting Information is available free of charge at of Chemistry at Oregon State University (OSU) and the https://pubs.acs.org/doi/10.1021/acsami.2c19206. College of Science at OSU for support through start-up funding and SciRis-ii, respectively. C.J.F. acknowledges support Detailed experimental procedures, material character- from OSU’s URSA Engage and Summer Undergraduate ization, methods, and ammonia isotherms (PDF) Research Experience (SURE) Science Program. 6752 https://doi.org/10.1021/acsami.2c19206 ACS Appl. Mater. Interfaces 2023, 15, 6747−6754 ACS Applied Materials & Interfaces www.acsami.org Research Article ■ REFERENCES (1) Erisman, J. W.; Sutton, M. A.; Galloway, J.; Klimont, Z.; Polyhedra Assemblies through Solution Processing. Chem. Sci. 2018, 9, 6463−6469. (21) Furukawa, S.; Horike, N.; Kondo, M.; Hijikata, Y.; Carné- Winiwarter, W. How a Century of Ammonia Synthesis Changed the Sánchez, A.; Larpent, P.; Louvain, N.; Diring, S.; Sato, H.; Matsuda, World. Nat. Geosci. 2008, 1, 636−639. R.; Kawano, R.; Kitagawa, S. Rhodium−Organic Cuboctahedra as (2) Valera-Medina, A.; Xiao, H.; Owen-Jones, M.; David, W. I. F.; Porous Solids with Strong Binding Sites. Inorg. Chem. 2016, 55, Bowen, P. J. Ammonia for Power. Prog. Energy Combust. Sci. 2018, 69, 10843−10846. 63−102. (22) Valencia-Loza, S. de J.; López-Olvera, A.; Martínez-Ahumada, (3) National Research Council (US) Committee on Acute Exposure E.; Martínez-Otero, D.; Ibarra, I. A.; Jancik, V.; Percástegui, E. G. SO2 Guideline Levels. Acute Exposure Guideline Levels for Selected Capture and Oxidation in a Pd6L8 Metal−Organic Cage. ACS Appl. Airborne Chemicals. Ammonia Acute Exposure Guideline Levels; Mater. Interfaces 2021, 13, 18658−18665. National Academies Press (US): Washington (DC), 2008; Vol. 6, p 2. (23) Tateishi, T.; Yoshimura, M.; Tokuda, S.; Matsuda, F.; Fujita, (4) Lucero, J. M.; Crawford, J. M.; Wolden, C. A.; Carreon, M. A. D.; Furukawa, S. Coordination/Metal−Organic Cages inside Out. Tunability of Ammonia Adsorption over NaP Zeolite. Microporous Coord. Chem. Rev. 2022, 467, 214612. Mesoporous Mater. 2021, 324, 111288. (24) Tranchemontagne, D. J.; Ni, Z.; O’Keeffe, M.; Yaghi, O. M. (5) Jaramillo, E.; Chandross, M. Adsorption of Small Molecules in Reticular Chemistry of Metal−Organic Polyhedra. Angew. Chem., Int. LTA Zeolites. 1. NH3, CO2, and H2O in Zeolite 4A. J. Phys. Chem. B Ed. 2008, 47, 5136−5147. 2004, 108, 20155−20159. (25) Lo Schiavo, S.; Cardiano, P.; Donato, N.; Latino, M.; Neri, G. A (6) Qajar, A.; Peer, M.; Andalibi, M. R.; Rajagopalan, R.; Foley, H. Dirhodium(I,I) Complex as a Highly Selective Molecular Material for C. Enhanced Ammonia Adsorption on Functionalized Nanoporous Ammonia Detection: QCM Studies. J. Mater. Chem. 2011, 21, Carbons. Microporous Mesoporous Mater. 2015, 218, 15−23. 18034−18041. (7) Helminen, J.; Helenius, J.; Paatero, E.; Turunen, I. Adsorption (26) Carné-Sánchez, A.; Albalad, J.; Grancha, T.; Imaz, I.; Juanhuix, Equilibria of Ammonia Gas on Inorganic and Organic Sorbents at J.; Larpent, P.; Furukawa, S.; Maspoch, D. Postsynthetic Covalent and 298.15 K. J. Chem. Eng. Data 2001, 46, 391−399. Coordination Functionalization of Rhodium(II)-Based Metal− (8) Helminen, J.; Helenius, J.; Paatero, E.; Turunen, I. Comparison Organic Polyhedra. J. Am. Chem. Soc. 2019, 141, 4094−4102. of Sorbents and Isotherm Models for NH3-Gas Separation by (27) Kawano, R.; Horike, N.; Hijikata, Y.; Kondo, M.; Carné- Adsorption. AIChE J. 2000, 46, 1541−1555. Sánchez, A.; Larpent, P.; Ikemura, S.; Osaki, T.; Kamiya, K.; Kitagawa, (9) Liu, J.; Lu, Z.; Chen, Z.; Rimoldi, M.; Howarth, A. J.; Chen, H.; S.; Takeuchi, S.; Furukawa, S. Metal-Organic Cuboctahedra for Alayoglu, S.; Snurr, R. Q.; Farha, O. K.; Hupp, J. T. Ammonia Synthetic Ion Channels with Multiple Conductance States. Chem Capture within Zirconium Metal−Organic Frameworks: Reversible 2017, 2, 393−403. and Irreversible Uptake. ACS Appl. Mater. Interfaces 2021, 13, 20081− (28) Cao, R.; Chen, Z.; Chen, Y.; Idrees, K. B.; Hanna, S. L.; Wang, 20093. X.; Goetjen, T. A.; Sun, Q.; Islamoglu, T.; Farha, O. K. Benign (10) Van Humbeck, J. F.; McDonald, T. M.; Jing, X.; Wiers, B. M.; Integration of a Zn-Azolate Metal−Organic Framework onto Textile Zhu, G.; Long, J. R. Ammonia Capture in Porous Organic Polymers Fiber for Ammonia Capture. ACS Appl. Mater. Interfaces 2020, 12, Densely Functionalized with Brønsted Acid Groups. J. Am. Chem. Soc. 47747−47753. 2014, 136, 2432−2440. (29) Chen, Y.; Du, Y.; Liu, P.; Yang, J.; Li, L.; Li, J. Removal of (11) Doonan, C. J.; Tranchemontagne, D. J.; Glover, T. G.; Hunt, J. Ammonia Emissions via Reversible Structural Transformation in R.; Yaghi, O. M. Exceptional Ammonia Uptake by a Covalent Organic M(BDC) (M = Cu, Zn, Cd) Metal−Organic Frameworks. Environ. Framework. Nat. Chem. 2010, 2, 235−238. Sci. Technol. 2020, 54, 3636−3642. (12) Yang, Y.; Faheem, M.; Wang, L.; Meng, Q.; Sha, H.; Yang, N.; (30) Han, X.; Lu, W.; Chen, Y.; da Silva, I.; Li, J.; Lin, L.; Li, W.; Yuan, Y.; Zhu, G. Surface Pore Engineering of Covalent Organic Sheveleva, A. M.; Godfrey, H. G. W.; Lu, Z.; Tuna, F.; McInnes, E. J. Frameworks for Ammonia Capture through Synergistic Multivariate L.; Cheng, Y.; Daemen, L. L.; MPherson, L. J. M.; Teat, S. J.; Frogley, and Open Metal Site Approaches. ACS Cent. Sci. 2018, 4, 748−754. M. D.; Rudić, S.; Manuel, P.; Ramirez-Cuesta, A. J.; Yang, S.; (13) Rieth, A. J.; Dincă, M. Controlled Gas Uptake in Metal− Schröder, M. High Ammonia Adsorption in MFM-300 Materials: Organic Frameworks with Record Ammonia Sorption. J. Am. Chem. Dynamics and Charge Transfer in Host−Guest Binding. J. Am. Chem. Soc. 2018, 140, 3461−3466. Soc. 2021, 143, 3153−3161. (14) Kim, D. W.; Kang, D. W.; Kang, M.; Choi, D. S.; Yun, H.; Kim, (31) Lyu, P.; Wright, A. M.; López-Olvera, A.; Mileo, P. G. M.; S. Y.; Lee, S. M.; Lee, J.-H.; Hong, C. S. High Gravimetric and Zárate, J. A.; Martínez-Ahumada, E.; Martis, V.; Williams, D. R.; Volumetric Ammonia Capacities in Robust Metal−Organic Frame- Dincă, M.; Ibarra, I. A.; Maurin, G. Ammonia Capture via an works Prepared via Double Postsynthetic Modification. J. Am. Chem. Unconventional Reversible Guest-Induced Metal-Linker Bond Dy- Soc. 2022, 144, 9672−9683. namics in a Highly Stable Metal−Organic Framework. Chem. Mater. (15) Nguyen, T. N.; Harreschou, I. M.; Lee, J.-H.; Stylianou, K. C.; 2021, 33, 6186−6192. Stephan, D. W. A Recyclable Metal−Organic Framework for (32) Moribe, S.; Chen, Z.; Alayoglu, S.; Syed, Z. H.; Islamoglu, T.; Ammonia Vapour Adsorption. Chem. Commun. 2020, 56, 9600−9603. Farha, O. K. Ammonia Capture within Isoreticular Metal−Organic (16) Yoshizawa, M.; Klosterman, J. K.; Fujita, M. Functional Frameworks with Rod Secondary Building Units. ACS Mater. Lett. Molecular Flasks: New Properties and Reactions within Discrete, Self- 2019, 1, 476−480. Assembled Hosts. Angew. Chem., Int. Ed. 2009, 48, 3418−3438. (33) Barin, G.; Peterson, G. W.; Crocellà, V.; Xu, J.; Colwell, K. A.; (17) Zarra, S.; Wood, D. M.; Roberts, D. A.; Nitschke, J. R. Nandy, A.; Reimer, J. A.; Bordiga, S.; Long, J. R. Highly Effective Molecular Containers in Complex Chemical Systems. Chem. Soc. Rev. Ammonia Removal in a Series of Brønsted Acidic Porous Polymers: 2015, 44, 419−432. Investigation of Chemical and Structural Variations. Chem. Sci. 2017, (18) Gosselin, A. J.; Rowland, C. A.; Bloch, E. D. Permanently 8, 4399−4409. Microporous Metal−Organic Polyhedra. Chem. Rev. 2020, 120, (34) Li, J.; Xiao, Y.; Shui, F.; Yi, M.; Zhang, Z.; Liu, X.; Zhang, L.; 8987−9014. You, Z.; Yang, R.; Yang, S.; Li, B.; Bu, X.-H. Extremely Stable Sulfuric (19) Rowland, C. A.; Lorzing, G. R.; Gosselin, A. J.; Trump, B. A.; Acid Covalent Organic Framework for Highly Effective Ammonia Yap, G. P. A.; Brown, C. M.; Bloch, E. D. Methane Storage in Capture. Chin. J. Chem. 2022, 40, 2445−2450. Paddlewheel-Based Porous Coordination Cages. J. Am. Chem. Soc. (35) Han, Y.-S.; An, S.; Dai, J.; Hu, J.; Xu, Q.; Song, F.; Li, M.; Peng, 2018, 140, 11153−11157. C.; Liu, H. Defect-Engineering of Anionic Porous Aromatic (20) Craig, G. A.; Larpent, P.; Kusaka, S.; Matsuda, R.; Kitagawa, S.; Frameworks for Ammonia Capture. ACS Appl. Polym. Mater. 2021, Furukawa, S. Switchable Gate-Opening Effect in Metal−Organic 3, 4534−4542. 6753 https://doi.org/10.1021/acsami.2c19206 ACS Appl. Mater. Interfaces 2023, 15, 6747−6754 ACS Applied Materials & Interfaces www.acsami.org Research Article (36) Kang, D. W.; Kang, M.; Moon, M.; Kim, H.; Eom, S.; Choe, J. works with Open Mn, Co, and Ni Sites. J. Am. Chem. Soc. 2016, 138, H.; Lee, W. R.; Hong, C. S. PDMS-Coated Hypercrosslinked Porous 9401−9404. Organic Polymers Modified via Double Postsynthetic Acidifications (51) Tejedor, I.; Andrés, M. A.; Carné-Sánchez, A.; Arjona, M.; for Ammonia Capture. Chem. Sci. 2018, 9, 6871−6877. Pérez-Miana, M.; Sánchez-Laínez, J.; Coronas, J.; Fontaine, P.; (37) Hu, T.-T.; Liu, F.; Dou, S.; Zhong, L.-B.; Cheng, X.; Shao, Z.- Goldmann, M.; Roubeau, O.; Maspoch, D.; Gascón, I. Influence of D.; Zheng, Y.-M. Selective Adsorption of Trace Gaseous Ammonia the Surface Chemistry of Metal−Organic Polyhedra in Their from Air by a Sulfonic Acid-Modified Silica Xerogel: Preparation, Assembly into Ultrathin Films for Gas Separation. ACS Appl. Mater. Characterization and Performance. Chem. Eng. J. 2022, 443, 136357. Interfaces 2022, 14, 27495−27506. (38) Pirillo, J.; Hijikata, Y. Trans Influence across a Metal−Metal (52) Carné-Sánchez, A.; Craig, G. A.; Larpent, P.; Guillerm, V.; Bond of a Paddle-Wheel Unit on Interaction with Gases in a Metal− Urayama, K.; Maspoch, D.; Furukawa, S. A Coordinative Solubilizer Organic Framework. Inorg. Chem. 2020, 59, 1193−1203. Method to Fabricate Soft Porous Materials from Insoluble Metal− (39) Carné-Sánchez, A.; Craig, G. A.; Larpent, P.; Hirose, T.; Organic Polyhedra. Angew. Chem., Int. Ed. 2019, 58, 6347−6350. Higuchi, M.; Kitagawa, S.; Matsuda, K.; Urayama, K.; Furukawa, S. (53) Ma, L.; Haynes, C. J. E.; Grommet, A. B.; Walczak, A.; Parkins, Self-Assembly of Metal−Organic Polyhedra into Supramolecular C. C.; Doherty, C. M.; Longley, L.; Tron, A.; Stefankiewicz, A. R.; Polymers with Intrinsic Microporosity. Nat. Commun. 2018, 9, 2506. Bennett, T. D.; Nitschke, J. R. Coordination Cages as Permanently (40) Boyar, E. B.; Robinson, S. D. Rhodium(II) Carboxylates. Coord. Porous Ionic Liquids. Nat. Chem. 2020, 12, 270−275. Chem. Rev. 1983, 50, 109−208. (41) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Petersson, G. A.; Nakatsuji, H.; Li, X.; Caricato, M.; Marenich, A. V.; Bloino, J.; Janesko, B. G.; Gomperts, R.; Mennucci, B.; Hratchian, H. P.; Ortiz, J. V.; Izmaylov, A. F.; Sonnenberg, J. L.; Williams-Young, D.; Ding, F.; Lipparini, F.; Egidi, F.; Goings, J.; Peng, B.; Petrone, A.; Henderson, T.; Ranasinghe, D.; Zakrzewski, V. G.; Gao, J.; Rega, N.; Zheng, G.; Liang, W.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Throssell, K.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M. J.; Heyd, J. J.; Br, Kudin, K. N.; Staroverov, V. N.; Keith, T. A.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A. P.; Burant, J. C.; Ivengar, S. S.; Tomasi, J.; Cossi, M.; Millam, J. M.; Klene, M.; Adamo, C.; Cammi, R.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Farkas, O.; Foresman, J. B.; Fox, D. J., et al. Gaussian 16, Revision B.01; Gaussian, Inc.: Wallingford, CT, 2016. (42) Zhao, Y.; Truhlar, D. G. The M06 Suite of Density Functionals for Main Group Thermochemistry, Thermochemical Kinetics, Noncovalent Interactions, Excited States, and Transition Elements: Recommended by ACS Two New Functionals and Systematic Testing of Four M06-Class Functionals and 12 Other Functionals. Theor. Chem. Acc. 2008, 120, 215−241. Institution of Metal–Organic Frameworks as a Highly (43) Malloum, A.; Fifen, J. J.; Dhaouadi, Z.; Engo, S. G. N.; Jaidane, Sensitive and Selective Layer In-Field Integrated Soil- N.-E. Structures and Relative Stabilities of Ammonia Clusters at Moisture Capacitive Sensor Different Temperatures: DFT vs. Ab Initio. Phys. Chem. Chem. Phys. Norah Alsadun, Mohamed Eddaoudi, et al. 2015, 17, 29226−29242. JANUARY 20, 2023 (44) Boese, R.; Niederprüm, N.; Bläser, D.; Maulitz, A.; Antipin, M. ACS APPLIED MATERIALS & INTERFACES READ Yu.; Mallinson, P. R. Single-Crystal Structure and Electron Density Distribution of Ammonia at 160 K on the Basis of X-Ray Diffraction Functionality-Induced Locking of Zeolitic Imidazolate Data. J. Phys. Chem. B 1997, 101, 5794−5799. Frameworks (45) Phillips, J. C.; Braun, R.; Wang, W.; Gumbart, J.; Tajkhorshid, Tongtong Xu, Yue-Biao Zhang, et al. E.; Villa, E.; Chipot, C.; Skeel, R. D.; Kalé, L.; Schulten, K. Scalable JANUARY 01, 2023 Molecular Dynamics with NAMD. J. Comput. Chem. 2005, 26, 1781− CHEMISTRY OF MATERIALS READ 1802. (46) Humphrey, W.; Dalke, A.; Schulten, K. VMD: Visual Molecular Flexing of a Metal–Organic Framework upon Hydrocarbon Dynamics. J. Mol. Graphics 1996, 14, 33−38. Adsorption: Atomic Level Insights from Neutron Scattering (47) Hernández-López, L.; Cortés-Martínez, A.; Parella, T.; Carné- Sánchez, A.; Maspoch, D. PH-Triggered Removal of Nitrogenous Benjamin A. Trump, Shane G. Telfer, et al. Organic Micropollutants from Water by Using Metal-Organic JANUARY 31, 2023 CHEMISTRY OF MATERIALS READ Polyhedra. Chem.�Eur. J. 2022, 28, No. e202200357. (48) Zhong, F.-Y.; Peng, H.-L.; Tao, D.-J.; Wu, P.-K.; Fan, J.-P.; Huang, K. Phenol-Based Ternary Deep Eutectic Solvents for Highly Computational Design of Metal–Organic Frameworks with Efficient and Reversible Absorption of NH3. ACS Sustainable Chem. Unprecedented High Hydrogen Working Capacity and High Eng. 2019, 7, 3258−3266. Synthesizability (49) Liu, J.; Chen, Z.; Wang, R.; Alayoglu, S.; Islamoglu, T.; Lee, S.- Junkil Park, Jihan Kim, et al. J.; Sheridan, T. R.; Chen, H.; Snurr, R. Q.; Farha, O. K.; Hupp, J. T. DECEMBER 23, 2022 Zirconium Metal−Organic Frameworks Integrating Chloride Ions for CHEMISTRY OF MATERIALS READ Ammonia Capture and/or Chemical Separation. ACS Appl. Mater. Interfaces 2021, 13, 22485−22494. Get More Suggestions > (50) Rieth, A. J.; Tulchinsky, Y.; Dincă, M. High and Reversible Ammonia Uptake in Mesoporous Azolate Metal−Organic Frame- 6754 https://doi.org/10.1021/acsami.2c19206 ACS Appl. Mater. Interfaces 2023, 15, 6747−6754