A Robust and Biocompatible Bismuth Ellagate MOF Synthesized Under Green Ambient Conditions.

The first bioinspired microporous metal-organic framework (MOF) synthesized using ellagic acid, a common natural antioxidant and polyphenol building unit, is presented. Bi2O(H2O)2(C14H2O8)·nH2O (SU-101) was inspired by bismuth phenolate metallodrugs, and could be synthesized entirely from nonhazardous or edible reagents under ambient aqueous conditions, enabling simple scale-up. Reagent-grade and affordable dietary supplement-grade ellagic acid was sourced from tree bark and pomegranate hulls, respectively. Biocompatibility and colloidal stability were confirmed by in vitro assays. The material exhibits remarkable chemical stability for a bioinspired MOF (pH = 2-14, hydrothermal conditions, heated organic solvents, biological media, SO2 and H2S), attributed to the strongly chelating phenolates. A total H2S uptake of 15.95 mmol g-1 was recorded, representing one of the highest H2S capacities for a MOF, where polysulfides are formed inside the pores of the material. Phenolic phytochemicals remain largely unexplored as linkers for MOF synthesis, opening new avenues to design stable, eco-friendly, scalable, and low-cost MOFs for diverse applications, including drug delivery.

Introduction

Tannins are natural polyphenols which can be found in a substantial part of the plant kingdom and are the second most abundant source of natural aromatic molecules after lignin.[1] They have traditionally been used for tanning leather, yet show great promise as a rich supply of precursors for green chemistry. Considering the general structure of tannin derivatives, having multiple aromatic rings with phenol functionalities, they stand out as a renewable source of rigid molecules.[1,2]

Rigid organic molecules capable of bridging metal cations are a principal component in the crystal chemistry of metal–organic frameworks (MOFs)—a class of porous hybrid materials which has shown an immense growth of compositions and topologies for over two decades, the main allure being the possibility of designing framework materials through a direct assembly of selected building units.[3,4] Considering the great variety of MOFs, the number of structures bridged only through phenolate ligands are few (representing <0.0008% of MOF reports), and existing examples involve synthetic linkers such as tetrahydroxybenzoate,[5] dihydroxybenzoquinone,[6−8] and hexahydroxytriphenylate ligands.[9−11] Numerous phenolate-based MOFs have been associated with higher chemical stability (especially in aqueous media) than carboxylate based MOFs—a consequence of the higher pKa of the phenolate groups, leading to strong chelation.[12,13]

Inspired by the potential stability of phenolate-based MOFs, together with the growing global demand to broaden the use of renewable resources, we have turned to tannins for the synthesis of green MOFs—specifically, gallotannins and ellagitannins, which contain gallic acid or ellagic acid subunits, respectively. While gallic acid, which consists of both carboxylic acid and phenol functional groups, has been utilized as a linker for the synthesis of a handful of MOFs,[14−19] surprisingly ellagic acid (Figure ) has seemingly never been used for MOF synthesis. Currently, the only entry in the Cambridge Structural Database (CSD) containing an ellagate anion together with a metal cation is a metal complex composed of two ruthenium cations linked by ellagate and capped by acetylacetonate.[20] Additionally, iron, manganese, and zinc ellagate materials have been developed for applications in phototherapy or as templates for porous carbons, yet no crystal structures nor porosities were reported.[21,22]

Figure 1

Ellagic acid—the phenolic building unit of ellagitannins.

Naturally high contents of ellagic acid are found in foods such as strawberries, raspberries, grapes, walnuts, pecans, and pomegranates, as well as in beverages such as wine, whiskey, and other spirits.[23−25] While the general benefit of consuming antioxidants, such as ellagic acid, remains up for debate due to challenges in maintaining oxidation homeostasis in healthy individuals,[26,27] their protective properties in oxidation-related chronic diseases, as well as their antimicrobial, antiviral, anti-inflammatory, and anticancer activities have been established.[28,29] In light of the scarcity with which phenolic bioderived molecules have been utilized for MOF synthesis, we find ellagic acid to be an appealing yet previously unexplored linker for the synthesis of bioinspired MOFs (BioMOFs) since (1) it is edible, biocompatible, bioactive, and has been consumed throughout human history (recommended intake ∼5 mg day–1 in adults,[30] reaching half-maximal inhibitory concentration (IC50) values of 23.5 μg mL–1 in diverse tumor cell lines or when administrating ∼60 mg kg–1 body weight for preclinical doses); (2) it is extracted from plant-based materials often considered as waste, such as tree bark and fruit peels; (3) can act as a rigid spacer between cations, contributing to the porosity of the resulting framework; (4) it has a centrosymmetric molecular structure, increasing the likelihood of it taking the role as a bridging linker; (5) it can chelate to metal cations, giving rise to strong five-membered rings; and (6) it has a low production cost.

Other polycomplexant biomolecules have previously been proposed as good candidates for the synthesis of BioMOFs,[31,32] and the resulting materials could potentially serve as delivery vehicles for active pharmaceutical ingredients (APIs) or as imaging agents.[33] However, many of the previously described BioMOFs are synthesized also using auxiliary nonbiological linker molecules and/or hazardous organic solvents. Despite the large appeal of using bioderived molecules for the synthesis of MOFs, examples are few and chemical stability is seldom thoroughly demonstrated, nor is biocompatibility.[34−38] To the best of our knowledge, only two microporous BioMOFs have been synthesized in water without auxiliary ligands: a magnesium gallate and an aluminum fumarate.[17,39] However, these MOFs demonstrate poor overall chemical stability, particularly in aqueous solutions.

In order to develop stable and biocompatible BioMOFs, the choice of metal cation is equally important. One nontoxic metal cation which has established biomedical applications is Bi3+.[35,40,41] As the second heaviest member of group 15 in the periodic table, bismuth is surrounded by notoriously toxic elements, yet bismuth itself has been found to have very low toxicity. With median lethal doses (LD50) of about 5 g kg–1 in rat models, the toxicity of bismuth and many of its compounds are comparable to or less than that of sodium chloride.[42] Furthermore, bismuth compounds have a two-century-long track record as APIs due to their inherent bioactivity and antimicrobial properties.[43] Bismuth subsalicylate (Pepto-Bismol) and bismuth subgallate are APIs made with the ligands salicylic acid and gallic acid, respectively.[44] Similar to ellagic acid, these are naturally occurring therapeutic phenolic molecules, making ellagic acid an appropriate choice of ligand to combine with Bi3+ for the development of a potentially biocompatible and bioactive MOF. So far, the use of bismuth in the synthesis of MOFs has been fairly limited, with only seven Bi(iii)-MOFs that have been reported to demonstrate porosity to nitrogen.[40,41,45−49] As all seven are exclusively based on synthetic carboxylate ligands, this begs the question, what new MOF structures can be acquired using Bi3+ and phenol-based linkers such as ellagic acid? The resulting material could potentially combine the biocompatible and bioactive features of both the metal cation and the organic linker, and would ideally be synthesized using green aqueous conditions.

In short, we report the synthesis, crystal structure, and physicochemical properties of a biocompatible and microporous bismuth ellagate framework (denoted SU-101)—the first MOF based on the bioactive phytochemical ellagic acid. For a bioinspired MOF, the material exhibits remarkable chemical stability and thermal stability under vacuum and in nitrogen, which led to investigations on its use as an adsorbent of the highly acidic gases SO2 and H2S, both of which are well-known air-pollutants associated with severe environmental and health issues.[50,51] Additionally, the environmentally friendly synthesis procedure, using a plant-derived linker molecule in water under ambient conditions, facilitates an easy scale-up process.

Results and Discussion

Synthesis of SU-101

Bismuth ellagate (SU-101) was prepared from an aqueous suspension of bismuth acetate and ellagic acid stirred under ambient conditions (i.e., room temperature and atmospheric pressure in air) for 48 h. Phase purity and synthesis reproducibility were improved through the introduction of acetic acid as a modulator (6% by volume, from either glacial acetic acid or a distilled white vinegar stock solution, pH ≈ 2.3; see the Supporting Information, SI, for details). Such a concentration of acetic acid is well within the range of what is found in common household vinegar (5–12 vol % of acetic acid in distilled white vinegar).[52]

Two different sources of ellagic acid were utilized for the synthesis of SU-101: (1) 97% chemical reagent-grade ellagic acid isolated from chestnut tree bark and (2) inexpensive 90% dietary supplement-grade ellagic isolated from pomegranate hulls, marketed as an edible antioxidant. Both starting materials resulted in high quality and phase-pure SU-101 based on powder X-ray diffraction (PXRD) and sorption data (Figures , S11, and S12). In both cases, the mixtures remained as suspensions throughout the reaction but nonetheless resulted in phase-pure SU-101. High percent yields of SU-101 were obtained at a small 30 mL reaction scale (76% yield; 0.28 g). Through a linear scale-up, a 600 mL large-scale synthesis resulted in a similar percent yield (74%, 5.6 g) of material in a single batch. Although the estimated space-time-yield using these conditions (approximately 5 kg m–3 day–1) is lower than values obtained for highly scalable MOFs (>1000 kg m–3 day–1),[38,53] one should note the simplicity and cost-efficiency of the synthesis procedure, carried out at atmospheric pressure without heating or using any other forms of external energy input.

Figure 2

Powder X-ray diffraction patterns of SU-101 made from reagent-grade ellagic acid (97% ellagic acid, isolated from chestnut tree bark) and supplement-grade ellagic acid (90% ellagic acid, isolated from pomegranate hulls), as well as a simulated powder X-ray diffraction pattern of SU-101.

Structure of SU-101

Due to the small crystallite size, the structure of SU-101 could not be solved using single-crystal X-ray diffraction (SCXRD). Instead, the crystal structure was determined using 3D electron diffraction (3DED), a method previously used to solve the structure of the microcrystalline metallodrug bismuth subgallate, as well as several Bi(iii)-MOFs.[41,44,45] For structure determination, 3DED data were acquired from larger 3 μm long bladed crystals (Figure , Table S1) which were synthesized using an alternative synthesis method under hydrothermal conditions (see SI for details). The structure was solved from the collected 3DED data and further refined against high-resolution powder X-ray diffraction (HR-PXRD) data in the tetragonal space group P42/n (a = 18.62 Å, c = 5.55 Å).

Figure 3

Reciprocal space projection of 3DED data collected on SU-101, viewed along c*, and an image of the crystal studied (larger crystals were specifically prepared for structure determination using an alternative hydrothermal synthesis procedure).

The framework of SU-101 consists of rod-shaped inorganic building units (IBUs) which run parallel to the c-axis and are linked together by ellagate anions in the ab-plane (Figure a,c), forming one-dimensional pores roughly 6–7 Å in diameter, taking into account the van der Waals radii of the framework atoms. The asymmetric unit of SU-101 contains half of an ellagate anion and one Bi3+ cation, as well as three individual oxygen atoms interpreted as (1) a coordinated water molecule, (2) a guest water molecule within the one-dimensional channels, and last, (3) a single μ4-oxygen, resulting in an overall charge neutral framework with the formula Bi2O(H2O)2(C14H2O8)·nH2O (n = 2, Figure S5). The coordination number of Bi3+ is six, of which three bonds form with coordinating phenolates, two μ4-oxygens, and a terminally coordinated water molecule (Figure b). The phenolate groups chelate to Bi3+ cations, and all four phenolate groups of the ellagate are presumed to be deprotonated based on the short Bi–O bonds, between 2.1 and 2.3 Å. These short distances are indicative of strongly chelating phenolates as longer Bi–O distances are found in the previously published Bi(iii)-carboxylate MOFs (typically in the range of 2.4–2.8 Å). On each end of the ellagate linker, one of the phenolates coordinates to a single Bi3+, while the other bridges two Bi3+ along the rod-shaped IBU (Figure d). The shortest intermolecular carbon–carbon distance between neighboring ellagates is 3.51 Å indicating π–π stacking between the ellagate linkers down the c-axis, further stabilizing the structure. The non-coordinated water molecules occupying the one-dimensional channels form hydrogen bonds to neighboring water molecules as well as the carbonyl oxygens of the lactone rings pointing into the channels, with oxygen–oxygen distances in the range of 2.7–2.9 Å (Figure S6).

Figure 4

(a) The structure of SU-101 as viewed down the c-axis. Hydrogen atoms and water molecules in the pores are omitted for clarity. (b) The coordination environment around Bi3+. The bond to a coordinated water molecule is represented as a dashed line. (c, d) Chelation of ellagate toward the bismuth oxo rods. (e) The tiling of the svd net. (f) The underlying svd net. (g, h) Choice of nodes for the deconstruction of the infinite IBU.

To simplify the description of SU-101, and to classify the structure, a topological analysis was performed. The choice of nodes when deconstructing the rod-shaped inorganic building unit was carried out as described by O’Keeffe and Yaghi,[54] where the point of extension of the linker, in this case, the midpoint between each pair of phenolates in the ellagate anions, ortho to one another, was selected as a node (Figure g,h). These nodes were then linked to neighboring vertices outlining the rod, resulting in a 6-c node, which, upon considering the linkage of rods by ellagate anions, gives the uninodal 7-coordinated net svd (Figure e,f).

Physicochemical Characterization

Thermogravimetric analysis (TGA) of SU-101 is in agreement with the proposed sum formula of Bi2O(H2O)2(C14H2O8)·nH2O (n = 1, Figure S7). The lower water content compared to that determined from HR-PXRD (n = 2) in a sealed capillary, and the immediate loss of water observed in the TGA data indicates that some of the water molecules in the pores are readily lost from the material at conditions close to room temperature.

The thermal stability of SU-101 was assessed by variable temperature powder X-ray diffraction (VT-PXRD), revealing that the framework remains crystalline under vacuum or in a nitrogen atmosphere at temperatures of up to 500 °C (Figures S8 and S9). While VT-PXRD tests are not always employed to study the thermal stability of MOFs, hindering an overall comparison, an intact and open framework at 500 °C (independent of atmosphere) has only been shown for a small handful of MOFs, such as CFA-2 and ZIF-8.[55] The framework remains intact up to 250 °C in air, after which the ellagate anion likely oxidizes, leading to a loss of crystallinity (Figure S10). This has also been observed with other MOFs composed of phenolate linkers and has been attributed to the relative ease with which phenolates are oxidized.[12]

The permanent microporosity of SU-101 was confirmed by N2 sorption experiments at 77 K after activating the material at 150 °C under vacuum for 10 h. Type I N2 isotherms, typical for microporous materials, were observed (Figures S11 and S12), with a Brunauer–Emmett–Teller surface area (SBET) of 412 m2g–1 when using reagent-grade ellagic acid for synthesis and 350 m2 g–1 for a washed large-scale batch produced from supplement-grade ellagic acid. These values are within the range of surface areas previously reported for Bi(iii)-MOFs (70–1100 m2 g–1).[40,41,45−49]

The average pore size of SU-101 was estimated to be approximately 6.8 Å (Figure S13), in good agreement with the crystallographic model (6–7 Å). The pore size distribution was calculated using the density functional theory (DFT) method with a N2 slit pore model. The trend toward lower surface areas obtained by Bi(iii)-MOFs in units of m2 g–1 is in part due to the heavy nature of Bi3+ compared to the significantly lighter metal cations found in the large majority of other MOFs. The adsorption of other environmentally and industrially relevant gases, CO2 and CH4, was also investigated (Figure S14), where SU-101 displays a higher affinity toward CO2 over N2 and CH4, due to the lower saturation pressure of CO2. The CO2/N2 selectivity for a hypothetical flue gas mixture, containing 15 kPa CO2 and 85 kPa N2 was estimated using the equation s = (q1/q2)/(p1/p2), giving a CO2/N2 selectivity of ∼35, yet other MOFs and porous materials have shown higher selectivity and CO2 uptakes under similar conditions.[56] Even so, bismuth-based MOFs are of interest as precursors to electrocatalysts for CO2 reduction, as demonstrated using CAU-7.[57]

The chemical stability of SU-101 was investigated by mimicking different industrially relevant media (e.g., for catalysis). Hence, as to cover a large range of chemical properties, the selected solvents were acetone, acetonitrile, chlorobenzene, cyclohexane, dichloromethane, 1,4-dioxane, dimethylformamide, dimethyl sulfoxide, ethanol, ethyl acetate, methanol, n-pentane, pyrrolidine, tetrahydrofuran, and toluene. PXRD results indicated that the structure remains intact in all of these solvents at room temperature (Figure S15), as well as when heated to 80 °C for 24 h in these solvents, with the exception of toluene where additional peaks, consistent with partial degradation, were observed (Figure S16). The high chemical stability of SU-101 arises from the presence of the two deprotonated phenolate oxygen atoms, chelating strongly to the Bi3+. As described for MIL-163,[12] the strong metal–linker interaction leads to a material that is resistant to harsh chemical conditions.

An ongoing challenge for MOFs is their stability in various aqueous conditions. As such, the pH-dependent stability of SU-101 was evaluated by exposing the material to aqueous solutions of HCl or NaOH, showing that the framework remains intact over a large pH-range, 2 < pH ≤ 14 (Figure S18). The structure of SU-101 also remained unchanged under hydrothermal conditions (180 °C for 24 h; Figure S16). However, at highly acidic conditions (pH ≤ 2), SU-101 is converted to BiOCl (Figure S18).

Immersion of SU-101 in potentially coordinative carboxylic acids (propionic acid, acetic acid and oxalic acid) was also investigated. The material remains intact at room temperature but decomposes at 80 °C in propionic acid. In acetic acid SU-101 undergoes partial transformation at room temperature and full degradation at 80 °C. In a concentrated oxalic acid solution decomposition occurs at room temperature (Figures S15 and S16).

As the ability to withstand the presence of coordinative species is of great importance for both biological as well as industrial applications, SU-101 was exposed to simulated physiological conditions such as phosphate-buffered saline (PBS) solution and cell culture medium (RPMI) at 37 °C. As seen in Figure , the material remains intact despite the presence of glucose, salts, vitamins, antibiotics, and amino acids, even for longer periods of time (24 h). Specifically, the stability of SU-101 was also evaluated in the presence of l-cysteine and l-cystine, showing that the material remains intact in solutions with these sulfur-containing biomolecules (Figure S17). Considering these results, the overall chemical stability of SU-101 appears to be better or similar to that of hallmark MOFs such as UiO-66 and ZIF-8, with the added value of using nontoxic reagents in a water-based synthesis.[58]

Figure 5

Powder X-ray diffraction patterns of SU-101 before and after exposure to a variety of aqueous conditions including a wide pH-range, hydrothermal conditions and simulated biological media. The limit of stability in acidic solutions can be seen at pH = 2, where a small peak appears at 2θ = 12°, attributed to the formation of BiOCl.

Colloidal Stability and Biocompatibility

In addition to the structural and chemical stability of a material in biorelevant conditions, insight into the physicochemical properties such as charge, surface chemistry, and colloidal stability will affect the affinity to different biological structures, which in turn governs the efficacy and biodistribution of the material.[59−61] Aqueous suspensions of SU-101 exhibit a small and monodisperse particle size of 129 ± 2 nm (PdI ∼ 0.3), a size which is a priori compatible with intravenous drug administration.[62] This hydrodynamic diameter, obtained by dynamic light scattering (DLS) measurements, assumes a spherical particle shape in solution, yet is in fair agreement with the microscopic observations acquired by TEM (Figures S1 and S2), showing polydispersed elongated crystals at its dry state (176 ± 88 nm length and 51 ± 18 nm wide; measured from 200 particles). This slight particle size difference between methods has been previously observed.[62] In addition, previous intravenous toxicological studies using elongated MOF nanoparticles did not show any signs of in vivo toxicity associated with the MOF morphology.[62,63] As such, a biocompatible character of SU-101 is expected (see below). When monitoring the particle size over time (24 h), SU-101 exhibited suitable colloidal stability in water (average size ∼137 ± 4 nm; Figure S19) with a strongly negative ζ-potential (−35 mV), possibly due to the presence of partially coordinated ellagate anions or hydroxyl groups completing the coordination sphere of superficial Bi3+. In turn, the resulting electrostatic repulsions might be responsible for the observed colloidal stability. In RPMI, the observed particle size is slightly larger than that in water (180 ± 26 nm), while also showing a less negative ζ-potential (−10 mV). This is expected as proteins in the RPMI media (supplemented with FBS) can adhere to the particle surfaces, hampering direct interparticle contact through the formation of a protein corona, leading to lower surface charge values (from −35 ± 1 to −10 ± 0 mV), in agreement with previous colloidal MOF behaviors.[62,64] Colloidal stability is maintained over time, with a particle size of 200 ± 11 nm over 24 h, further indicating properties suitable for biological applications.

To explore the intrinsic qualities and potential applications of this green material, in vitro cytotoxicity tests were performed to assess the biocompatibility of SU-101. For these purposes, a promyelocytic cell line (HL-60) was selected as a model for differentiated phagocytic cells and granulocytes involved in the body’s defense. Remarkably, after 24 h of cell contact, a very low cytotoxicity profile was observed (80% maximum inhibitory concentration, IC80 = 1000 μgmL–1), even when exposed to very high concentrations of SU-101 (up to 1200 μgmL–1; Figure S20). This is in agreement with the absence of severe toxicity observed for the individual reagents—ellagic acid and bismuth acetate. Therefore, SU-101 seems to be a promising candidate for environmental and biological applications, such as depollution and drug delivery, both being chemically stable and biocompatible.

SO2 and H2S Capture in SU-101

Due to the remarkable chemical stability demonstrated by SU-101, adsorption of the acidic gases SO2 and H2S was investigated. Sulfur dioxide (SO2) is a toxic gas which was classified by the World Health Organization as one of the most hazardous air pollutants, associated with increased respiratory problems[50,65] and mortality.[66] Similarly, hydrogen sulfide (H2S) is a highly toxic chemical and exposure to concentrations of over 100 ppm can be fatal as it is rapidly absorbed into the bloodstream, limiting O2 uptake at a cellular level.[51,67,68] However, few MOFs have demonstrated stability upon exposure to either SO2 or H2S, principally due to the formation of strong and irreversible metal–sulfur bonds.[69−76]

To investigate the adsorptive properties of SU-101, SO2 sorption isotherms were acquired at pressures between 0 to 1 bar at 298 K on an activated sample using a Dynamic Gravimetric Gas/Vapor Sorption Analyzer, DVS vacuum (see SI). The resulting SO2 isotherm shows a total uptake of 2.2 mmol g–1 (Figure S21). Although this value is considerably lower than the uptake of other MOFs,[70,72,77,78] the stability of SU-101 upon adsorption–desorption cycling was investigated and structural integrity after both dry and humid SO2 sorption was confirmed by PXRD (Figures S25 and S27). Furthermore, the host–guest interaction between SU-101 and SO2 was quantified, evaluating the isosteric heat of adsorption (ΔH) for SO2 at low coverage for SU-101 (estimated by fitting two adsorption isotherms at 303 and 308 K to a Clausius–Clapeyron equation, Figures S22 and S23). The low isosteric heat of adsorption for SO2 (ΔH = −29.6 kJ mol–1) indicates that no M–SO2 interactions occur (>90 kJ mol–1).[79] Since the heat of adsorption was low, cycling experiments were carried out, where activation of SU-101 between cycles was achieved by applying vacuum (1.7 × 10–6 Torr) for only 15 min at 298 K. Using this procedure, it was demonstrated that the SO2 capture capacity (1 bar, 298 K) remains constant during 50 adsorption–desorption cycles (2.200 ± 0.004 mmol g–1, Figure S24) with the structure remaining intact (Figure S25).

H2S capture by SU-101 was investigated with a series of breakthrough experiments (4.3 vol % H2S with 95.7 vol % of N2). Samples were first activated at 120 °C for 6 h under a flow of dry N2 gas, after which the material was cooled down to 25 °C. The resultant breakthrough H2S experiment led to a gas uptake of 15.95 mmol g–1, i.e., 542 mg g–1, (Figures and S28) equivalent to 515 cm3g–1. Surprisingly, considering that the SBET of SU-101 is 412 m2g–1, this H2S uptake is one of the highest reported, and is comparable to the top-performing MOFs, such as Ni-CPO-27 (12.0 mmol g–1, SBET = 1545 m2 g–1),[80] MFM-300(Sc) (16.5 mmolg–1, SBET = 1360 m2 g–1)[81] or MIL-53(Al)-TDC (18.1 mmol g–1, SBET = 1150 m2g–1).[82] After the initial exposure to H2S, the sample was reactivated and used for a second H2S adsorption cycle. Interestingly, the capacity initially observed was lost (0.2 mmol g–1, Figure ) despite an apparent preservation of the framework, as observed by PXRD (Figure S29). To further investigate the seemingly lost porosity, a N2 sorption experiment at 77 K was performed, showing a specific surface area of 15 m2 g–1. As the framework remains intact, we hypothesize that H2S is captured by SU-101 via a chemisorption process, leading to the formation of polysulfides. To test this hypothesis, Raman spectra of SU-101 were acquired before and after H2S adsorption (Figure S30). After exposure to H2S two additional peaks appear, one at 243 cm–1 corresponding to different S2– species[83,84] and another at 445 cm–1 corresponding to S42–.[83,85,86] These characteristic peaks support the hypothesized chemisorption of H2S within SU-101, resulting in the formation of polysulfides. We have recently published a similar phenomenon for MFM-300(Sc),[81] where the MOF can oxidize H2S, forming low order polysulfides (n = 2) species. Presumably, the electrochemical potential of SU-101 should be similar to the one for MFM-300(Sc) (2.29 V),[81] considering that S42– species were also found for MFM-300(Sc). The mechanism of polysulfide formation from H2S includes the adsorption of H2S, dissociation, oxidation, and recombination of sulfur species.[81,87] Diffuse-reflectance Fourier-transform infrared spectroscopy experiments (Figure S31) were conducted on SU-101 before and after H2S adsorption to explore the possible perturbation of the band characteristic of the phenolate stretching vibration in the ellagate anion (3408 cm–1). This band was shifted to a lower wavenumber (3100 cm–1) after H2S sorption, suggesting a relatively strong interaction between the phenolate group of SU-101 and the polysulfides, as previously reported for other MOFs.[81,82] This polysulfide formation, paired with the high stability of the material, could pave the way toward applications such as the construction of a MOF-based lithium/sulfur battery.

Figure 6

Breakthrough curve of H2S adsorption by SU-101 at 25 °C and 1 bar. The inset shows the comparative H2S adsorption capacities for each cycle.

Conclusions

As the first MOF synthesized from ellagic acid, SU-101 exemplifies the use of a renewable phenol-functionalized and plant-based linker to construct a highly chemically stable and biocompatible MOF under scalable, green, and ambient synthesis conditions. Bi3+ was the first cation we combined with ellagic acid in an attempt to develop a new MOF, resulting in SU-101. We therefore expect that a plethora of new crystal structures, including MOFs and coordination polymers, can be synthesized from ellagic acid as well as other previously unexplored phenolic phytochemicals. As the synthesis of SU-101 was carried out at room temperature and atmospheric pressure, further scale-up is expected to be feasible. Combining sustainable green solvents, such as water, with renewable and inexpensive linkers as well as biocompatible inorganic species in the design and synthesis of new framework materials paves the way for further development of MOFs as green materials for environmental and health applications, including, but certainly not limited to, gas capture and drug delivery.