This is the peer reviewed version of the following article: Reljic S., Broto-Ribas A., Cuadrado-Collados C., Jardim E.O., Maspoch D., Imaz I., Silvestre-Albero J.. Structural Deterioration of Well-Faceted MOFs upon H2S Exposure and Its Effect in the Adsorption Performance. Chemistry - A European Journal, (2020). . : - . 10.1002/chem.202002473, which has been published in final form at https://dx.doi.org/10.1002/chem.202002473. This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Use of Self-Archived Versions. Structural deterioration of well-faceted MOFs upon H2S exposure and its effect in the adsorption performance Snezana Reljic,[a] Anna Broto-Ribas,[b] Carlos Cuadrado-Collados,[a] Erika O. Jardim,[a] Daniel Maspoch,[b,c] Inhar Imaz,[b] and Joaquin Silvestre-Albero*[a] [a] S. Reljic, C. Cuadrado-Collados, Dr. E.O. Jardim, Prof. J. Silvestre-Albero Laboratorio de Materiales Avanzados, Departamento de Química Inorgánica-IUMA Universidad de Alicante Ctra. San Vicente-Alicante s/n, E-03690 San Vicente del Raspeig, Spain E-mail: joaquin.silvestre@ua.es [b] A. Broto-Ribas, Prof. D. Maspoch, Dr. I. Imaz Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC and Barcelona Institute of Science and Technology, Campus UAB, Bellaterra 08193, Barcelona, Spain [c] Prof. D. Maspoch ICREA Pg. Lluís Companys 23, 08010 Barcelona, Spain Supporting information for this article is given via a link at the end of the document. Abstract: The structural deterioration of archetypical, well-faceted materials can reach the actual DoE target (i.e., 263 v/v) at 10 metal organic frameworks (MOFs) has been evaluated upon MPa and 25ºC.[11] These carbon materials also possess a proper exposure to an acidic environment (H2S). Experimental results show performance for CO2 capture at 5.0 MPa with an adsorption that the structural damage highly depends on the nature of the capacity above 2065 mg/g at 25ºC.[12] Despite these excellent hybrid network (e.g., softness of the metal ions, hydrophilic values, the main drawback when working with activated carbon properties, among others) and the crystallographic orientation of the materials is the difficulty to tune a specific pore size and shape exposed facets. Microscopy images show that HKUST-1 with well- to achieve an optimum adsorption performance. Higher defined octahedral (111) facets is completely deteriorated, ZIF-8 with adsorption values have been described in the literature for preferentially exposed (110) facets exhibits a large external MOFs due to the presence of a regular, well-defined porous deterioration with the development of holes or cavities in the network. Pen et al. reported a storage capacity for methane as mesoporous range, whereas UiO-66-NH2 with (111) exposed facets, high as 230 v/v at 3.5 MPa and 270 v/v at 6.5 MPa for HKUST- and PCN-250 with (100) facets does not reflect any sign of surface 1.[4] Unfortunately, this excellent performance could not be damage. Despite the selectivity in the external deterioration, X-ray preserved after a conforming step due to the mechanical diffraction and gas adsorption measurements confirm that indeed all collapse of the internal porous structure. More recently, Tian et MOFs suffer an important internal deterioration, these effects being al. reported the successful development of HKUST-1 monoliths more severe for MOFs based on softer cations (e.g., Cu-based with a storage capacity as large as 259 v/v at 6.5 MPa, avoiding HKUST-1 and Fe-based PCN-250). These structural changes have the aforementioned limitations related with densification and inevitable important effects in the final adsorption performance for shaping.[13] A similar approach was applied by Connolly et al. to CO2 and CH4 at low and high pressures. grow UiO-66 monoliths with an exceptional performance for CH4, up to 211 v/v at 6.5 MPa, and for CO2 adsorption, up to 284 v/v at 4.0 MPa.[14] Despite these excellent numbers, preferentially when using Introduction MOFs, one has to be aware that these experiments are always performed using high-purity gases. However, real industrial The design of nanoporous materials able to store a large streams contain other gases that are able, not only to compete quantity of CH4 and CO2 at high pressure is one of the main for the active sites, but also to deteriorate the quality of the challenges nowadays in materials science due to the necessity porous network after use. For instance, this is the case of acidic to combine: i) a widely developed porous structure; ii) a proper molecules such as H2S. H2S is traditionally present in natural pore architecture (pore size and shape), able to promote a gas streams, in which, although methane is the main component, dense packing density of the adsorbed phase; iii) a large there is usually a H2S content of <1.0 % vol. H2S is also present framework density, to achieve an optimum adsorption capacity in flue gas streams (in addition to CO2, coal fired flue gas in the minimum volume (v/v); and iv) a proper structural stability contains around 100 ppm H2S) and in biogas (in addition to CH4 and cyclability. Previous studies described in the literature show and CO2, biogas contains <3% of H2S). that activated carbon materials and metal-organic frameworks To date, some previous studies have already shown that (MOFs) are currently the most promising candidates for CH4 and exposure of MOFs to acidic gases can give rise to important CO2 storage due to the possibility to tune their porous network to structural changes.[15,16] Interestingly, the activation energy achieve a proper adsorption performance.[1-10] In the specific associated with these gas-framework interactions, and the case of CH4, high-pressure adsorption measurements have associated side-effects, highly depends on the crystallographic shown that properly designed high-surface area carbon 1 orientation. For instance, Pang et al. demonstrated that the After exposure to H2S, the situation differed depending on the (110) facets of ZIF-8 crystals are more reactive than the (100) MOF evaluated. Thus, while UiO-66-NH2 and PCN-250 did not facets under humid SO2.[15] Furthermore, while this structural reflect external structural damages after H2S exposure at 75ºC damaged was mainly restricted to the external surface of ZIF-8 for 10 h, ZIF-8 started showing the formation of holes at the crystals, the bulk of these crystals remained mainly unchanged. boundaries between (110) facets even after 3 h at 75ºC (Figure Remarkably, Dutta et al. reported that the bulk properties of ZIF- 2a). Notably, a further treatment of ZIF-8 crystals with H2S for 10 8 were preserved after exposing it to H2S, showing similar h gave rise to more severe changes on the external surface of intraparticle diffusivities before and after acid treatment.[16] the (110) facets, with the creation of meso and macrocavities Anisotropic acid etching of MOFs using HCl has also been used (Figure 2b). The limiting case is HKUST-1. For this specific in the literature to obtain unprecedented morphologies due to sample, even 3h at 25ºC are sufficient to produce a drastic the different reactivity among crystallographic directions.[17] deterioration in the morphology of the crystals, with a complete These preliminary studies clearly anticipate that the composition destruction of the well-defined facets (Figure 2e). of the effluent gas and, more specifically, the presence of acidic contaminants can have a crucial role in the performance and (a) (b) lifetime of MOFs. However, the effect that exposure to these acidic gases could have in the subsequent adsorption performance for the main constituents of the industrial effluents containing H2S, i.e. CH4 and CO2, have been scarcely evaluated in the literature. Based on these premises, the main goal of this study is to evaluate the structural deterioration of four archetypical, well- (211) (111) (100) faceted MOFs (ZIF-8, UiO-66-NH2, PCN-250 and HKUST-1) upon H2S exposure under industrially relevant experimental conditions, and the subsequent evaluation of their adsorption performance for two relevant industrial/technological processes: CO2 capture and CH4 storage at high pressures. At this point it is (c) (d) important to highlight that, although some of them are not the best performing MOFs for this specific application, they constitute representative examples of the four main families of metallic nodes in MOFs (Zn-, Fe-, Zr- and Cu-based materials). In addition, they can be prepared with a high crystallographic quality and with well-defined facets, thus allowing a better (111) understanding of external and internal deterioration effects. (100) Results and Discussion Figure 1. Representative FE-SEM images of as-synthesized (a) ZIF-8, (b) External deterioration upon H2S exposure. UiO-66-NH2, (c) PCN-250 and (d) HKUST-1. One of the main characteristics of the four MOFs selected is the possibility to synthesize them in well-defined shapes with (a) (b) preferential orientation of specific crystallographic facets. This approach is of paramount importance to identify the selective attack of the acidic moieties to specific crystallographic domains. Figure 1 and 2 show FE-SEM images of the four MOFs before and after their exposure to H2S for 10 h at 75ºC (in the case of HKUST-1 exposure was limited to 25ºC for 3h because these conditions were sufficient to cause a severe structural deterioration). As shown in Figure 1, the four as-synthesized 500 nm 500 nm MOFs possess different crystal sizes, shapes and morphologies. ZIF-8 is constituted by  0.5 m in-size rhombic dodecahedral (c) (d) particles exposing twelve (100) facets.[17] UiO-66-NH2 exhibits uniform octahedral crystals with eight exposed (111) surfaces and an average crystal size of  0.6 m. PCN-250 is made of cubic crystals exposing six (100) facets with an average crystal size of  1 m. Although PCN-250 crystals are also quite uniform, smaller nanocrystals could be found attached to the main crystal units. Finally, HKUST-1 exhibits well-defined octahedral crystals with (111) exposed facets and an average 500 nm 500 nm crystal size of  15 m. 2 the XRPD peaks were clearly seen. Last but not least, XRPD analysis of HKUST-1 clearly anticipate that exposure to H2S for 3h at 25ºC are sufficient to completely deteriorate the quality of (e) the crystals, with a complete wash-out of the main XRPD contributions. Based on these observations, HKUST-1 exposure to H2S was not tested for higher temperatures and/or longer times. ZIF-8 simulated ZIF-8 UiO-66-NH2 simulated UiO-66-NH2 ZIF-8 3 h-25ºC UiO-66-NH2 3 h-25ºC ZIF-8 3 h-75ºC UiO-66-NH2 3 h-75ºC Intensity (a.u.) ZIF-8 10 h-75ºC UiO-66-NH2 10 h-75ºC Intensity (a.u.) 500 nm Figure 2. Representative FE-SEM images of (a,b) ZIF-8, (c) UiO-66-NH2, (d) PCN-250 and (e) HKUST-1 after exposure to H2S at 75ºC for 10 h. Note that 6 8 10 12 14 16 18 20 image (a) was taken after a H2S exposure at 75ºC for 3 h and (e) was taken 2(º) 6 8 10 12 14 16 18 20 2(º) after a H2S exposure at 25ºC for 3h. HKUST-1 simulated PCN-250 simulated HKUST-1 HKUST-1 3h-25ºC PCN-250 PCN-250 3 h-25ºC These results clearly show that the external deterioration of PCN-250 3 h-75ºC Intensity (a. u.) Intensity (a.u.) PCN-250 10 h-75ºC MOF crystals upon exposure to an acidic media depends, not only on the crystallographic orientation, but also on the nature of the MOF. While UiO-66-NH2 and PCN-250 with exposed (111) and (100) facets, respectively, are highly resistant and does not experience visual damage, the (110) facets exposed in ZIF-8 are 6 8 10 12 14 16 18 20 6 8 10 12 14 16 18 20 more prone to suffer severe damage with the development of 2(º) 2 (º) highly extended nanometer size holes and cavities. This external deterioration of ZIF-8 crystals is in close agreement with recent Figure 3. XRPD patterns for ZIF-8, UiO-66-NH2, PCN-250 and HKUST-1 before and after exposure to H2S. results from Dutta et al.[16] In the case of HKUST-1, despite sharing a similar crystal morphology to UiO-66-NH2, with eight well-defined (111) facets, FE-SEM results clearly anticipate a It is important to highlight here that, contrariwise to the FE-SEM drastic deterioration of the crystal network even after exposure analysis described above, XRPD anticipates that the framework to H2S at 25ºC for 3h. This observation anticipates the crucial damage is more evident for UiO-66-NH2 and PCN-250 than for role of the MOF nature, and more specifically, of the metallic ZIF-8. Thus, both techniques are complementary but necessary nodes defining the extend of the structural deterioration in an to evaluate the internal and external deterioration of MOFs upon acidic environment. H2S exposure. They showed that the acid treatment clearly affects the external surface of ZIF-8 crystals but not their inner X-ray powder diffraction (XRPD) study of the internal core. However, for UiO-66-NH2 and PCN-250, while there is no deterioration. evidence of external deterioration upon H2S exposure, XRPD One of the main questions when MOF crystals are exposed to anticipates a certain degree of internal deterioration, which is acidic gases is the preservation of their quality after the different more pronounced for PCN-250. treatments. As abovementioned, FE-SEM characterization suggested the preservation of the crystal quality for both UiO-66- Textural characteristics before and after H2S exposure. NH2 and PCN-250, at least from a surface perspective. However, A very important aspect when a porous material is exposed to MOF crystals can also suffer internal deterioration, which acid gases is to ascertain its textural deterioration, i.e. the detection requires the use of bulk techniques. To start evaluating selective destruction of pores or cavities and/or the creation of the internal deterioration of our selected MOFs, we performed new pores. These details can provide further insight about the XRPD measurement before and after exposure of the four reaction mechanism with the exposed gases. One of the main MOFs to an H2S flow at different times (3 and 10 h) and techniques for the characterization of the porous structure in temperatures (25 and 75ºC) (except for HKUST-1). As it can be nanoporous solids is gas adsorption at cryogenic temperatures. appreciated in Figure 3, the patterns of the as-synthesized Among the different probe molecules available, N2 adsorption at samples perfectly fit with the simulated patterns, thus confirming -196ºC is the most widely applied to evaluate both micro- and their purity. In the case of ZIF-8, H2S exposure did not produce mesoporosity.[18] Figure 4 shows the N2 adsorption isotherms for important changes in the XRPD pattern, except a broadening of the four MOFs before and after exposure to H2S under different the main diffraction peaks and a slight shift to higher angles. The experimental conditions. Note that, for ZIF-8, the isotherm was internal deterioration via amorphization of the framework was plotted in logarithmic scale for a more careful evaluation of the more evident in UiO-66-NH2, mainly at 75ºC, at which a characteristic gate-opening effect taken place at mid-relative significant decrease of intensity and a shift of the main XRPD pressures.[19,20] peaks were observed. A similar amorphization was also found From these measurements, we found that the structural damage for PCN-250 upon H2S exposure, mainly at 75ºC. In this latter after exposure to H2S highly depends on the nature of the MOF case, a more pronounced shift and a decrease of the intensity of 3 material and the presence of functionalities in the organic linker. trends. However, at this point we cannot exclude an additional For ZIF-8, N2 adsorption data revealed a progressive contribution due to the hydrophilic character of the MOF, i.e., deterioration in the textural parameters (see Table S1 for further due to the presence of residual humidity in the material. details), mainly due to the degradation of the microporous Although Zn(II) is softer than Cu(II) and Fe(III), the hydrophobic network. The pore size distribution (Figure S1) confirm the character of ZIF-8 will prevent residual moisture to participate in degradation of the microporous network and the development of the degradation mechanism, whereas this is not the case for new pores around 1.8-2 nm. Furthermore, the apparent surface PCN-250 and, even less, for the hydrophilic HKUST-1. area decreased up to 16% after a H2S treatment at 75ºC for 10 h, Computational investigations by Pang et al. supported the from an initial value of 1860 m2/g down to a value of 1560 m2/g. reaction mechanism between the MOF framework and SO2, and A closer look to the N2 isotherms (see inset) showed that the the promoting effect by water through the formation of H2SO3 gate-opening behavior at p/p0 ~ 0.01 does not change or shifts species.[15] Similar experiment by Bhattacharyya et al. confirmed after the acid treatment, i.e., apparently the acid treatment the larger structural deterioration in ZIF-8 under humid SO2 produces an external deterioration of the crystal, without conditions, compared to the same system in dry conditions.[22] perturbation of the structural gate-opening phenomena. Similarly, UiO-66-NH2 suffered a progressive deterioration upon 500 350 ZIF-8 acid treatment; evidenced by the decrease of its apparent 400 ZIF-8 3 h-25°C ZIF-8 3 h-75°C 300 Volume adsorbed (cm STP/g) Volume adsorbed (cm STP/g) ZIF-8 10 h-75°C surface area from 1160 m2/g (as-synthesized sample) to 1022 250 3 3 m2/g after exposure to H2S (75ºC, 10 h). A closer look to the 300 200 500 textural values described in Table S2 showed that the 150 Volume adsorbed (cm STP/g) 200 UiO-66-NH2 400 3 UiO-66-NH2 3 h-25 °C deterioration affects preferentially to the microporous network, in O 100 UiO-66-NH2 3 h-75°C 100 300 UiO-66-NH2 10 h-75°C close agreement with the N2 isotherms. Interestingly, the 200 ZIF-8 ZIF-8 10 h-75°C 50 1E-3 0.01 0.1 structural deterioration after its exposure to H2S (75ºC, 10 h) 0 1E-5 1E-4 1E-3 0.01 p/p0 0.1 1 0 0.0 0.2 0.4 0.6 0.8 1.0 p/p0 p/p0 was the lowest among the evaluated MOFs (only 12%). This 500 400 observation is in close agreement with the improved stability Volume adsorbed (cm STP/g) described by Mounfield et al. for NH2-functionalized MIL-125 Volume adsorbed STP (cm3/g) 320 400 3 after exposure to humid SO2.[21] 240 300 HKUST-1 HKUST-1 3h-25°C N2 adsorption isotherms of PCN-250 after its exposure to H2S 160 200 showed an important deterioration of the framework with a PCN-250 PCN-250 3 h-25°C significant shrinkage of the microporous network, and the 80 PCN -250 3 h-75°C 100 PCN-250 10 h-75°C associated development of mesocavities. Indeed, the apparent 0 0 surface area decreased from 1300 m2/g (as-synthesized MOF) 0.0 0.2 0.4 p/po 0.6 0.8 1.0 0.0 0.2 0.4 p/p0 0.6 0.8 1.0 down to 830 m2/g upon the acid treatment for 10 h at 75ºC, i.e., Figure 4. N2 adsorption (full symbols)/desorption (empty symbols) isotherms a decrease of around 36%, much higher than that observed for for ZIF-8, UiO-66-NH2, PCN-250 and HKUST-1 before and after exposure to H2S. ZIF-8 and UiO-66-NH2. Furthermore, the mesopore volume scaled up to 0.14 cm3/g. The development of mesopores with a wide pore size distribution (from 3.0 nm to 6.0 nm) can be more clearly appreciated in Figure S1. X-ray photoelectron spectroscopy (XPS) analysis. Finally, when HKUST-1 was exposed to H2S, 3h at 25ºC were The chemical composition and environment of the four MOFs sufficient to produce a complete collapse of the 3D network. As before and after the acidic treatment with H 2S were evaluated it can be observed in Figure 4 and Table S4, the nitrogen using X-ray photoelectron spectroscopy (XPS). XPS is a surface adsorption capacity decreases more than 95%, the BET surface sensitive technique and consequently, it is suitable to evaluate area dropping from 1688 m2/g in the original MOF down to 74 the potential deterioration of the surface of the MOF crystals, m2/g in the treated sample. since this is the most susceptible region upon H2S exposure. In summary, the textural characterization of the four MOFs Table 1 reports the surface composition for ZIF-8, UiO-66-NH2, exposed to H2S showed that all of them suffer somehow an PCN-250 and HKUST-1 before and after the thermal treatment important structural deterioration. These effects are more drastic at 75ºC for 10 h under H2S flow conditions (in the case of for Cu(II)-based HKUST-1, followed by Fe(III)-based PCN-250, HKUST-1 the data correspond to 25ºC for 3h). In the case of Zn(II)-based ZIF-8, and finally the more robust Zr(IV)-based UiO- ZIF-8, the acid treatment gave rise to a decrease in the 66-NH2. Although previous studies by Mounfield et al. atomic % of carbon and nitrogen, whereas the oxygen content demonstrated that amine functionalization can improve the experienced a slight increase. These changes are clearly structural stability of MIL-125 towards acidic gases,[21] our results appreciated after normalization with the total amount of metal suggest that the structural stability is also defined by the (M). Although these results could suggest that imidazole ligands physicochemical properties of the MOF network. Our results are removed from the surface more efficiently than Zn(II) suggest that the stability of the MOFs is largely defined by the metallic nodes, in agreement with previous data reported in the acid-base interactions between the H2S molecule and the MOF literature, standard deviations are too large to provide a framework (HSAB concept), i.e., a soft base such as sulphur will conclusive answer.[15] Compared to the literature, the increase in have a higher affinity and will form stronger bonds with softer the amount of oxygen in our study was rather limited due to the metal ions (transition metals with a high occupation of d orbitals). dry nature of the H2S stream used. The detailed spectra for the Based on these premises, sulphur should have higher affinity for C 1s, N 1s, O 1s, S 2p and Zn 2p can be found in the Supporting softer cations (e.g., Cu(II), Zn(II) and Fe(III)) versus harder Information (Figure S2-S5). These spectra confirmed the above cations (e.g., Zr(IV)), thus explaining the observed stability described data and the presence of a significant amount of sulphur anchored to the ZIF-8 surface, with a peak centered at a 4 binding energy of  162-163 eV (atomic % 0.7); very close to the Finally, HKUST-1 exhibits a significant content of carbon and binding energy of ZnS.[23] Exposure of ZIF-8 to H2S also shifted oxygen in the as-prepared sample, whereas nitrogen content is the Zn 2p3/2 signal to higher binding energies (a shoulder negligible. However, upon H2S exposure the chemical analysis appeared at  1023.1 eV). This shift is most probably associated shows a significant decrease in the amount of carbon and a with Zn-S species. However, care must be taken in the large increase in the oxygen content, most probably associated interpretation of this data due to difficulty to differentiate the with the formation of -OH surface species promoted by moisture. chemical state of Zn through the evaluation of the Zn 2p3/2 signal. As expected, the amount of sulphur incorporated in HKUST-1 is Concerning UiO-66-NH2, the XPS analysis showed that the the highest among of samples evaluated (up to 3.2 wt.%), in surface of this material contains preferentially carbon and close agreement with the softer character of the Cu metallic oxygen, with a small contribution of nitrogen (Figure S3 shows a nodes. detailed evaluation of these components). Upon H2S exposure, the carbon and oxygen ratio (normalized to the amount of Zr) High-pressure CO2 adsorption isotherms. decreased, while the nitrogen ratio remains stable. These results One of the open questions at this point is how these structural support the hypothesis that the acid treatment promotes the changes produced by an acidic environment can alter the deterioration of the ligand (maybe through acetate defects). adsorption performance of these MOFs for two industrially Interestingly, the amount of sulphur retained by UiO-66-NH2 relevant gases, i.e. CH4 and CO2. Previous studies described in after the acid treatment was negligible (atomic % = 0), thus the literature using carbide-derived carbons (TiC-CDC) have confirming the weak affinity of sulphur for a hard metallic cation anticipated that CO2 adsorption at atmospheric pressure (0.1 such as Zr. MPa) requires the presence of pores around 0.8 nm in size; i.e., angstrom-size pores allow an optimum packing density under these experimental conditions.[26] However, larger pores (around Table 1. Chemical composition (carbon, oxygen, metal, nitrogen and sulphur 2.0-3.0 nm) govern the adsorption performance for high- content) of ZIF-8, UiO-66-NH2,PCN-250 and HKUST-1 samples (atomic %) pressure applications.[27] Based on these premises, although and atomic ratio, deduced from XPS analysis before and after their exposure to H2S at 75ºC for 10 h (25ºC for 3h in the case of HKUST-1). deterioration of the microporous structure in MOFs must be a priori detrimental for the adsorption of CO2 at atmospheric Sample C O M[a] N S C/M N/M O/M pressure, the potential development of mesopores, through the (%) (%) (%) (%) (%) destruction of micropores, can be useful to promote the adsorption at high pressures. ZIF-8 Before 75.6 5.1 4.2 15.1 0 18.1 3.6 1.2 ±1.2 ±0.7 ±1.2 ±3.7 ±0.0 Figure 5 shows the CO2 excess adsorption isotherms at 25ºC and up to 4.0 MPa for the four MOFs before and after exposure After 72.4 7.4 4.8 14.6 0.7 14.9 3.0 1.5 to an acidic environment. As it can be appreciated, the as- ±1.0 ±0.5 ±0.5 ±1.0 ±0.1 synthesized MOFs exhibit a type I isotherm with a large adsorption capacity at low pressures due to the presence of UiO-66- Before 73.4 20.7 3.4 2.5 0 21.8 0.74 6.2 NH2 ±0.4 ±0.0 ±0.1 ±0.3 ±0.0 microporosity and a certain increased capacity above 3.0 MPa; most probably due to the presence of mesocavities in the After 67.9 24.3 4.5 3.1 0.2 14.9 0.70 5.3 interparticle space. The total adsorption capacity for CO2 at 4.0 ±3.9 ±2.4 ±1.1 ±0.6 ±0.1 MPa is 435 mg/g for ZIF-8, 505 mg/g for UiO-66-NH2, 635 mg/g PCN- Before 72.3 22.5 1.7 3.0 0 42.5 1.8 13.2 for PCN-250 and 700 mg/g for HKUST-1. These values are in ±3.4 ±2.6 ±0.6 ±0.1 ±0 250 the same range observed for carbon materials exhibiting 74.0 20.5 2.0 2.5 0.9 37.0 1.2 10.2 After exclusively micropores,[27] and much lower than those reported ±4.5 ±4.1 ±0.2 ±1.1 ±0.3 65.0 27.8 6.4 0.45 0 10.1 0.07 4.3 for high-surface area carbon materials and exceptional MOFs HKUST- Before 1 ±0.1 ±0.7 ±0.5 ±0.1 ±0.0 (values larger than 1200 mg/g have been reported for samples like carbon VR-5, MOF-177, and MOF-200, at 4.0 MPa).[12,28,29] After 53.5 36.3 6.5 0.36 3.2 8.3 0.06 5.6 500 ±2.7 ±2.1 ±0.8 ±0.1 ±0.1 400 400 Amount adsorbed (mg/g) Amount adsorbed (mg/g) 300 [a] M corresponds to the different metals in the MOFs, i.e. Zn for ZIF-8, Zr for 300 UiO-66-NH2, Fe for PCN-250 and Cu for HKUST-1. 200 200 UiO-66-NH2 ZIF-8 UiO-66-NH2 3 h-25°C 100 ZIF-8 3 h-25°C ZIF-8 3 h-75°C 100 UiO-66-NH2 3 h-75°C PCN-250 also contains a significant amount of carbon, oxygen ZIF-8 10 h-75°C UiO-66-NH2 10 h-75°C and nitrogen. However, the acid treatment significantly 0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 00.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Pressure (MPa) Pressure (MPa) decreased the amount of these three components referred to Fe, 700 as shown in Table 1. The evaluation of the Fe 2p signal in the 600 600 as-synthesized PCN-250 showed a broad band centered at Amount adsorbed (mg/g) 500 Amount adsorbed (mg/g) 500 711.7 eV (Figure S4), which was attributed to Fe(III). After 400 400 exposure to H2S, the main contribution developed a shoulder at 300 HKUST-1 300 HKUST-1 3h-25°C lower binding energies (ca. 709.7 eV), due to the development 200 PCN-250 200 of Fe-S species.[24,25] The retention of sulphur in the MOF 100 PCN-250 3 h-25°C PCN-250 3 h-75°C PCN-250 10 h-75°C 100 network was clearly appreciated in the S 2p3 region with a 0 0 significant contribution at 164 eV (0.9 at.%), attributed to the 0.0 0.5 1.0 1.5 2.0 2.5 Pressure (MPa) 3.0 3.5 4.0 0.0 0.5 1.0 1.5 2.0 2.5 Pressure (MPa) 3.0 3.5 4.0 formation of Fe-based thiol (-SH) or disulfide (S2-) species. 5 Figure 5. High-pressure CO2 adsorption isotherms (mg/g) at 25ºC for ZIF-8, UiO-66-NH2, PCN-250 and HKUST-1 evaluated before and after exposure to H2S. Figure 6. High-pressure CH4 adsorption isotherms (wt.%) at 25ºC for ZIF-8, UiO-66-NH2, PCN-250 and HKUST-1 before and after exposure to H2S. After exposure to an acidic environment, the four MOFs exhibited a market decrease in the amount adsorbed, preferentially at pressure below 2.0 MPa, due to the Figure 6 shows the high-pressure CH4 isotherms for the four deterioration of the inner microporous network. This reduction MOFs before and after the acid treatment. As expected, the was significant for ZIF-8 and PCN-250, with a decrease in the largest adsorption capacity corresponded to the as-synthesized amount adsorbed at 2.0 MPa of around 20% for the sample samples with a total capacity at 4.0 MPa around 8.2 wt.% for exposed at 75ºC for 10 h (see Tables S5-S8). In these two ZIF-8, 9.8 wt.% for UiO-66-NH2, 12.1 wt.% for PCN-250 and MOFs, however, the decrease in the adsorption capacity upon 14.4 wt.% for HKUST-1 (see Tables S5-S8 for further details). H2S exposure was less significant at 4.0 MPa; this value being These values are slightly lower than those described in the around 5-6 % compared to the un-treated ZIF-8 and PCN-250. literature for the best carbon materials and MOFs (e.g., HKUST- In the case of UiO-66-NH2, the decrease in the amount of CO2 1 with a capacity of 17-20 wt.% at 4.0 MPa).[11] adsorbed was larger at higher pressures, whereas at 2.0 MPa, In all cases, exposure to H2S gave rise to a decrease in the the effect is rather limited. This observation reflects the lower amount of CH4 stored over the whole pressure range evaluated, deterioration of the microporous structure in UiO-66-NH2, in this effect being rather similar at 2.0 MPa and 4.0 MPa. This close agreement with the N2 adsorption data described above. observation reflects that CH4 storage under these experimental On the contrary, the development of mesocavities or holes in conditions is affected by the amount and extend of micropores ZIF-8 and PCN-250 upon deterioration explains the additional and not by the presence of wider pores. The decrease in the increase in the amount of CO2 adsorbed above 3.0 MPa. adsorption capacity was rather similar for ZIF-8 and UiO-66-NH2 HKUST-1 constitutes an exception to the observed behavior. In (around 15 % after the acid treatment at 75ºC for 10 h), whereas this sample, after H2S exposure, a drastic decrease in the the effect for PCN-250 was much larger (around 25 %). In the amount of CO2 adsorbed can be appreciated over the whole specific case of HKUST-1, the decrease was close to 87%, with pressure range evaluated (at 4.0 MPa CO2 uptake does not a final uptake close to 1.8 wt.%. These results are in agreement reach 150 mg/g). with the decrease in the micropore volume obtained from the N2 isotherms after the H2S treatment (Tables S1-S4). This decrease High-pressure CH4 adsorption isotherms. was around 11 % for ZIF-8 and UiO-66-NH2, around 39 % for Another important adsorption process susceptible of acid PCN-250 and 96% for HKUST-1. In any case, these results contamination is CH4 storage, either from natural gas or biogas. confirm that the effect of H2S exposure in the CH4 storage Although the amount of H2S in these streams can be reduced to capacity is rather similar to the effect in CO2 adsorption, at least ppm levels using a proper guard bed, the presence of traces of at intermediate pressures. At high pressures, CO2 can H2S or any other acid molecule can have a tremendous impact compensate and mitigate these detrimental effects due to the in the adsorption performance of the porous material after presence of mesoporosity available for CO2 condensation at several cycles. To this end, excess methane adsorption capacity high pressures, whereas CH4 does not. was also evaluated for the four MOFs at 25ºC and up to 4.0 MPa. Under these experimental conditions and contrary to CO2, only micropores must be susceptible to participate in the adsorption Conclusion process. This assumption is based on the large difference in the critical temperature between the two probe molecules: CO2 The structural stability of four archetypical, well-faceted MOFs (31.13ºC) and CH4 (-82.45ºC). At 25ºC, CO2 adsorption takes was evaluated upon exposure to an acidic environment. place in subcritical conditions, i.e., it is susceptible to experience Experimental results showed that the structural deterioration condensation processes in wide pores, whereas CH4 adsorption depends on the nature of the MOF network (e.g., nature of the takes place in supercritical conditions, i.e., it is non-condensable. metallic nodes, hydrophilic properties, among others), and the 9 10 crystallographic orientation of the exposed facets. While ZIF-8 8 8 exhibits a significant external deterioration on the crystal surface 7 with the development of holes or cavities, FE-SEM images of Amount adsorbed (wt.%) Amount adsorbed (wt.%) 6 5 6 UiO-66-NH2 and PCN-250 does not allow to ascertain any on- 4 4 surface structural damage. On the contrary, HKUST-1 exhibits a 3 UiO-66-NH2 UiO-66-NH2 3 h-25°C rather complete structural deterioration even after a treatment at 2 ZIF-8 ZIF-8 3 h-25°C ZIF-8 3 h-75°C 2 UiO-66-NH2 3 h-75°C UiO-66-NH2 10 h-75°C 25ºC for 3h. XRPD and N2 adsorption measurements confirmed 1 0 ZIF-8 10 h-75°C 0 the presence of a significant amorphization of the inner structure, 0.0 0.5 1.0 1.5 2.0 Pressure (MPa) 2.5 3.0 3.5 4.0 0.0 0.5 1.0 1.5 2.0 Pressure (MPa) 2.5 3.0 3.5 4.0 mainly for ZIF-8 and PCN-250, associated with the preferential interaction of a soft base as sulphur with soft acid cations such 12 14 as Fe(III) and Zn(II). 10 12 Notably, high-pressure adsorption measurements confirm that Amount adsorbed (wt.%) Amount adsorbed (wt.%) 10 8 these structural changes have a dramatic effect in the 8 6 adsorption performance of the four MOFs. Here, high-pressure 6 HKUST-1 4 PCN-250 HKUST-1 3h-25°C CO2 adsorption show that the structural deterioration and the PCN-250 3 h-25°C 4 PCN-250 3 h-75°C associated decrease in the storage capacity is much larger for 2 PCN-250 10 h-75°C 2 0 0 6 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 Pressure (MPa) Pressure (MPa) HKUST-1, followed by ZIF-8 and PCN-250. The presence of - propan-2-ol. Both solutions were mixed at room temperature for 5 min. NH2 functionalities and the lower affinity of Zr(IV) for sulphur Meanwhile, 3.1 g of FeCl3·6H2O were dissolved at room temperature in 9 limits the deterioration of the inner crystal structure of UiO-66- mL of propan-2-ol. Finally, the metal solution was added to the ligand solution, and the resulting slurry was stirred at 85 °C for 24 h. A NH2. However, the associated development of mesopores or crystalline powder was recovered by centrifugation, washed with distilled holes in ZIF-8 and PCN-250 is able to compensate, at least water and ethanol and dried under vacuum. partially, the detrimental effect of the acid treatment in the inner micropores. These observations suggest that, contrary to HKUST-1 literature, the structural deterioration in ZIF-8 is not exclusively a surface phenomenon and some internal amorphization must For the synthesis of HKUST-1, a solution A was prepared by dissolving take place, in close agreement with our XRPD measurements. 0.420 g of benzene-1,3,5-tricarboxylic acid in 12 mL ethanol and a Contrariwise, high-pressure CH4 adsorption confirm that the solution B was prepared by dissolving 0.8742 g of Cu(NO3)2·5H2O in 12 presence of acidic moieties in industrial streams will have a mL of ultrapure water. Once dissolved, both solution were mixed under significant effect in the CH4 storage capacity, mainly due to the stirring conditions for 30 min, and later on transferred into a 50 mL teflon- deterioration of the inner microporous network. Although these lined stainless steel autoclave. The autoclave containing the mixture was placed in the oven at 110ºC for 18h. Finally, the obtained crystals were drawbacks will be more drastic for MOFs with a large affinity washed with ethanol and dried at 80ºC overnight. between the metallic nodes and sulphur (such as HKUST-1 and PCN-250), we hypothesized that, in general, many MOFs will H2S exposure experiments experience limitations to outperform under real industrial streams, unless guard beds are appropriately considered. The synthesized MOFs were exposed to a flow of 1000 ppm H2S in N2 (30 ml/min) using an U-shaped quartz reactor. The acidic treatment was performed at two industrially relevant temperature conditions (25 ºC and Experimental Section 75 ºC) and for a total duration of 3 h. Note that, only at 75 ºC, the acidic treatment was extended to 10 h. The main purpose of this step was to Synthesis of MOFs expose MOF crystals to H2S in the temperature range relevant to industrial biogas and flue gas streams, i.e. 25 ºC-75 ºC. ZIF-8 Physico-chemical characterization of the MOFs ZIF-8 was synthesized according to a previously reported synthetic prodecure.[17] A solution of 0.3 g of Zn(OAc)2·2H2O in 5 mL of deionized Textural properties of the evaluated MOFs were determined using N 2 water was added into a solution of 1.12 g of 2-methylimidazolate (2-MIM) adsorption at cryogenic temperatures. N2 data were obtained in a home- in 5 mL of water, and the resulting mixture was homogenized by stirring it made manometric equipment designed and constructed by the LMA for a few seconds. Then, the mixture was let at room temperature for 24 group and now commercialized by G2Mtech (www.g2mtech.com). Before h. White crystals were collected and washed several times with methanol the adsorption measurement, samples were outgassed under ultra-high and finally, dried under vacuum. vacuum conditions (UHV) for 4 h at 200 ºC. N2 adsorption data were used to estimate the apparent surface area (SBET), micropore volume (Vo), mesopore volume (Vmeso) and total pore volume (VT). UiO-66-NH2 Powder X-Ray Diffraction (PXRD) patterns were recorded on an X'Pert A solution of 0.36 g of ZrCl4 in 50 mL of DMF was added to a solution of PRO MPD analytical diffractometer (Panalytical) at 45 kV, 40 mA using 0.26 g of 2-aminoterephthalic acid and 35 mL of acid acetic glacial in 110 CuKα radiation (λ = 1.5419 Å). Field-Emission Scanning Electron mL DMF. The resulting mixture was then heated at 120 ºC in an oven for Microscopy (FE-SEM) images were collected on a FEI Magellan 400L 24 h. After this period, a crystalline powder was recovered by scanning electron microscope at an acceleration voltage of 1.0 kV, using centrifugation, and washed twice with DMF and twice with MeOH. aluminum as support. Synthesis of 3,30,5,50-azobenzene tetracarboxylic acid (H4-Tazb) X-ray photoelectron spectroscopy (XPS) spectra were recorded in a XPS K-ALPHA Thermo Scientific. All spectra were collected using an Al-K TazbH4 was prepared following a previous procedure.[30] 9.5 g of 5- radiation (1486.6 eV), monochromatized by a twin crystal nitroisophthalic acid, 25 g of NaOH and 125 mL of distilled water were monochromator, yielding focused X-ray spot elliptical shaped with a first placed into a 1 L 3-neck round bottom flask and stirred at 60 ºC for major axis length of 400 µm at 3 mA x 12 kV. The alpha hemispherical 1h. This mixture formed a pink slurry that turned yellow after 1 h. analyser was operated at the constant energy mode with survey scan Meanwhile, 50 g of glucose were dissolved into 75 mL of distilled water pass energies of 200 eV to measure the whole energy band and 50 eV in at 60 ºC. The glucose solution was added slowly to the slurry, turning a narrow scan to selectively measure the desired elements. Charge dark brown. The final mixture was left to cool down for 10 min, and air compensation was achieved with the system flood gun that provides low was bubbled into it for 16 h under stirring at room temperature. After this energy electrons and low energy argon ions from a single source. The period, the mixture was cooled in an ice-bath and the sodium salt of Tazb CHx in carbon 1s score level was used as reference binding energy recovered by filtration. The solid was then dissolved in 200 mL of distilled (284.6 eV). The powder samples were pressed and mounted on the water and then, this solution was acidified to pH = 1, using 37% HCl, sample holder and placed in the vacuum chamber. Before the spectrum yielding a bright orange precipitate. The product was recovered by recording, the samples were maintained in the analysis chamber until a filtration, washed with distilled water and dried at 120 ºC. residual pressure of ca. 5 × 10−7 Nm-2. The peaks deconvolution was performed by a quantitative analysis calculating the integral of each peak, PCN-250 after subtracting the S-shaped background, and by fitting the experimental curve to a combination of Lorentzian (30%) and Gaussian PCN-250 was synthetized by a slight modification of the synthesis (70%) model. Due to the chemical heterogeneity of the MOF crystals process previously described.[31] 0.56 g of NaOH was dissolved in 1 mL after the H2S treatment, XPS analysis have been performed in different of distilled water. 2.0 g of H4-TazBz were suspended in 11.12 mL of 7 positions, the chemical composition reported in Table 1 being an average [17] C. Avci, J. Ariñez-Soriano, A. Carné-Sánchez, V. Guillerm, C. Carbonell, of these analyses. I. Imaz, D. Maspoch, Angew. Chem. Int. Ed. 2015, 54, 14417-14421. [18] F. Rouquerol, J. Rouquerol, K.S.W. Sing in Adsorption by Powders and High-pressure CO2 and CH4 measurements Porous Solids, Academic Press, 1999. [19] D. Fairen-Jimenez, S.A. Moggach, M.T. Wharmby, P.A. Wright, S. Parsons, T. Düren, J. Am. Chem. Soc. 2011, 133, 8900-8902. High-pressure adsorption measurements were performed in a home- [20] M.E. Casco, Y.Q. Cheng, L. Daemen, D. Fairen-Jimenez, E.V. Ramos- made fully automated equipment designed and constructed by the LMA Fernandez, A.J. Ramirez-Cuesta, J. Silvestre-Albero, Chem. Commun. group. Before the adsorption measurements, samples were outgassed 2016, 52, 3639-3642. in-situ under UHV conditions at 200 ºC for 4 h. Excess adsorption [21] W.P. Mounfield, C. Han, S.H. Pang, U. Tumuluri, Y. Jiao, S. isotherms were obtained at 25ºC and up to 4.0 MPa for CH4 and CO2. Bhattacharyya, M.R. Dutzer, S. Nair, Z. Wu, R.P. Lively, D.S. Sholl, K.S. Walton, J. Phys. Chem. C 2016, 120, 27230-27240. [22] S. Bhattacharyya, S.H. Pang, M.R. Dutzer, R.P. Lively, K.S. Walton, Acknowledgements D.S. Sholl, S. Nair, J. Phys. Chem. C 2016, 120, 27221-27229. [23] Y.-C. Liang, C.-C. Wang, RSC Advances 2018, 8, 5063-5070. [24] M.V. Morales-Gallardo, A.M. Ayala, M. Pal, M.A. Cortes Jacome, J.A. JSA would like to acknowledge financial support from the Toledo Antonio, N.R. Mathews, Chem. Phys. Lett. 2016, 660, 93-98. MINECO (Project MAT2016-80285-p). I.I. and D.M. would like to [25] K. Xi, D. He, C. Harris, Y. Wang, C. Lai, H. Li, P.R. Coxon, S. Ding, C. acknowledge financial support from the Spanish MINECO Wang, R. Vasant Kumar, Adv. Sci. 2019, 6, 1800815. (Project RTI2018-095622-B-100), and Catalan AGAUR (Project [26] V. Presser, J. McDonough, S.-H. Yeon, Y. Gogotsi, Energy Environ. Sci. 2017 SGR 238), the ERC, under the EU-FP7 (ERC-Co 615954) 2011, 4, 3059-3066. and the CERCA Program/Generalitat de Catalunya. ICN2 is [27] M.E. Casco, M. Martínez-Escandell, J. Silvestre-Albero, F. Rodríguez- supported by the Severo Ochoa Program from the Spanish Reinoso, Carbon 2014, 67, 230-235. [28] H. Furukawa, N. Ko, Y.B. Go, N. Aratani, S.B. Choi, E. Choi, A.O. MINECO (Grant No. SEV-2017-0706). Yazaydin, R.Q. Snurr, M. O´Keeffe, J. Kim, O.M. Yaghi, Science 2010, 329, 424-428. Keywords: MOFs • Acid exposure • Structural stability • H2S • [29] A.R. Millward, O.M. Yaghi, J. Am. Chem. Soc. 2005, 127, 17998-17999. High-pressure [30] A. Dhakshinamoorthy, M. Alvaro, H. Chevreau, P. Horcajada, T. Devic, C. Serre, H. García, Catal. Sci. Technol. 2012, 2, 324-330. [1] J. Silvestre-Albero, F. Rodríguez-Reinoso in Nanoporous Materials for [31] H. Chevreau, A. Permyakova, F. Nouar, P. Fabry, C. Livage, F. Ragon, gas Storage, (Eds. K. Kaneko, F. Rodríguez-Reinoso), Green Energy A. García-Marquez, T. Devic, N. Steunou, C. Serre, P. Horcajada, and Technology Series, 2019, pp. 209-226. CrystEngComm. 2016, 18, 4094-4104. [2] M. Sevilla, G.A. Ferrero, A.B. Fuertes in Nanoporous Materials for gas Storage, (Eds. K. Kaneko, F. Rodríguez-Reinoso), Green Energy and Technology Series, 2019, pp. 287-330. [3] Y. He, W. Zhou, G. Qian, B. Chen, Chem. Soc. Rev. 2014, 43, 5657- 5678. [4] Y. Peng, V. Krungleviciute, I. Eryazici, J.T. Hupp, O.K. Farha, T. Yildirim, J. Am. Chem. Soc. 2013, 135, 11887-11894. [5] B. Lin, H.-M. Wen, W. Zhou, J.Q. Xu, B. Chen, Chem 2016, 1, 557-580. [6] A. Dailly, M. Beckner, in Nanoporous Materials for gas Storage, (Eds. K. Kaneko, F. Rodríguez-Reinoso), Green Energy and Technology Series, 2019, pp. 227-253. [7] Y. Ma, H. Tanaka, R. Matsuda in Nanoporous Materials for gas Storage, (Eds. K. Kaneko, F. Rodríguez-Reinoso), Green Energy and Technology Series, 2019, pp. 331-358. [8] J.H. Cavka, C.A. Grande, G. Mondino, R. Blom, Ind. Eng. Chem. Res. 2014, 53, 15500-15507. [9] D. Britt, H. Furukawa, B. Wang, T. Grant Glover, O.M. Yaghi, J. Halpern, PNAS 2009, 106, 20637-20640. [10] H. Li, K. Wang, Y. Sun, C.T. Lollar, J. Li, H.-C. Zhou, Mater. Today 2018, 21, 108-121. [11] M.E. Casco, M. Martínez-Escandell, E. Gadea-Ramos, K. Kaneko, J. Silvestre-Albero, F. Rodríguez-Reinoso, Chem. Mater. 2015, 27, 959- 964. [12] J. Silvestre-Albero, A. Wahby, A. Sepúlveda-Escribano, M. Martínez- Escandell, K. Kaneko, F. Rodríguez-Reinoso, Chem. Commun. 2011, 47, 6840-6842. [13] T. Tian, Z. Zeng, D. Vulpe, M.E. Casco, G. Divitini, P.A. Midgley, J. Silvestre-Albero, J.C. Tan, P.Z. Moghadam, D. Fairen-Jimenez, Nature Materials 2018, 17, 174-179. [14] B.M. Connolly, M. Aragones-Anglada, J. Gandara-Loe, N.A. Danaf, D.C. Lamb, J.P. Mehta, D. Vulpe, S. Wuttke, J. Silvestre-Albero, P.Z. Moghadam, A.E.H. Wheatley, D. Fairen-Jimenez, Nature Commun. 2019, 10, 2345. [15] S.H. Pang, C. Han, D.S. Sholl, C.W. Jones, R.P. Lively, Chem. Mater. 2016, 28, 6960-6967. [16] A. Dutta, N. Tyminska, G. Zhu, J. Collins, R.P. Lively, J.R. Schmidt, S. Vasenkov, J. Phys. Chem. C 2018, 122, 7278-7287. 8 Entry for the Table of Contents H2 S (211) (100) Well-faceted metal-organic Internal/external deterioration upon frameworks H2S exposure Exposure of MOFs to an acidic environment (H2S) give rise to important structural changes. Depending on the nature of the MOF and the crystallographic facets exposed, these changes include the development of cavities or holes in the external surface and/or the internal deterioration of the 3D MOF network. Furthermore, this structural deterioration gives rise to drastic changes in the adsorption performance of these materials for two relevant industrial processes: CO2 capture and CH4 storage. 9