Dalton Transactions View Article Online
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
PAPER
Cite this: DOI: 10.1039/c5dt00606f
View Journal
Hydrophobic and moisture-stable metal–organic frameworks† Carlos A. Fernandez,* Satish K. Nune,* Harsha V. Annapureddy, Liem X. Dang, B. Peter McGrail, Feng Zheng, Evgueni Polikarpov, David L. King, Charles Freeman and Kriston P. Brooks Metal–organic frameworks (MOFs) have proved to be very attractive for applications including gas storage, separation, sensing and catalysis. In particular, CO2 separation from flue gas in post-combustion processes is one of the main focuses of research among the scientific community. One of the major issues that are preventing the successful commercialization of these novel materials is their high affinity towards water that not only compromises gas sorption capacity but also the chemical stability. In this paper, we demonstrate a novel post-synthesis modification approach to modify MOFs towards increasing hydrophobic behaviour and chemical stability against moisture without compromising CO2 sorption capacity. Our approach consists of incorporating hydrophobic moieties on the external surface of the MOFs via physical adsorption. The rationale behind this concept is to increase the surface hydrophobicity in the porous materials without the need of introducing bulky functionalities inside the pore which compromises the sorption capacity toward other gases. We herein report preliminary results on routinely studied MOF materials [MIL-101(Cr) and NiDOBDC] demonstrating that the polymer-modified MOFs retain CO2 sorption capacity while reducing the water adsorption up to three times, with respect to the un-modified materials, via an equilibrium effect. Furthermore, the water stability of the polymer-functionalized MOFs is significantly higher than the water stability of the bare material. Molecular dynamic simu-
Received 10th February 2015, Accepted 1st May 2015
lations demonstrated that this equilibrium effect implies a fundamental and permanent change in the
DOI: 10.1039/c5dt00606f
water sorption capacity of MOFs. This approach can also be employed to render moisture stability and selectivity to MOFs that find applications in gas separations, catalysis and sensing where water plays a criti-
www.rsc.org/dalton
cal role in compromising MOF performance and recyclability.
Introduction Metal–organic frameworks (MOFs) are porous, crystalline 2D and 3D structures that possess very important features including high surface areas, controlled porosities, good thermal stability and the possibility of adjusting the pore size and chemical environment.1–6 For these reasons, they are increasingly in demand for applications in gas storage,7–9 separation,10–12 catalysis,13–15 imaging,16,17 proton conductors18 and drug delivery.17,19 Despite their potentials, there has been a limited success in the practical applications of MOFs, including CO2 removal in post-combustion processes.1 One of
Applied Functional Materials, Hydrocarbon Processing Group, Energy and Environment Directorate, Pacific Northwest National Laboratory, Richland, Washington 99352, USA. E-mail: [email protected], [email protected] † Electronic supplementary information (ESI) available: Detailed adsorption isotherms of water and CO2, SEM image of modified DOBDC, water sorption kinetics plot and simulation snapshots of NiDOBDC fragment in solvent chloroform and solution containing P123. See DOI: 10.1039/c5dt00606f
This journal is © The Royal Society of Chemistry 2015
the most significant drawbacks is their affinity towards water that renders mechanical and chemical instability. Particularly, zinc/carboxylate-based MOFs including isoreticular MOFs (IRMOFs) are extremely sensitive to moisture. This instability is related to the nature of the bonding (coordination bonding) between zinc atoms and carboxylate ligands.1 Other transition metal-based MOFs have also shown water instability.20–23 Copper-trimesic acid based MOF (HKUST-1) and Al(OH)(2,6ndc) (ndc = naphthalene dicarboxylate; DUT-4) are unstable in the presence of water vapor.21,23 Thorough understanding of the aspects behind the water stability of MOFs is still missing; this is demonstrated, e.g., by the fact that materials of the same family such as Al(OH)(1,4-ndc) and Al(OH)(2,6-ndc) (ndc = naphthalenedicarboxylate) show very different behavior, with the first material being stable upon water adsorption while the second decomposes.24 Some MOF materials, such as the MIL series (MIL101-Cr, MIL100-Fe), UIO-66 (containing ZrCl4 and 1,4-benzenedicarboxylate linkers (BDC) or zeolite imidazole frameworks (ZIFs) including the well-known ZIF-8 are resistant to moisture.25–30 However, their CO2 sorption capacities at low
Dalton Trans.
View Article Online
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Paper
pressure region (0 to 0.1 atm) is low in CO2/H2O gas mixtures.31 One possible reason for such low uptake in MILs is due to the competing selectivity of these materials toward water and non-availability of open metal centers that favor CO2 sorption. Stabilizing these MOF materials against ambient humidity and increasing the hydrophobic character would make MOFs more suitable for specialized and industrial applications including removal of CO2 from flue gas in postcombustion processes. A number of approaches were reported in the literature to stabilize these materials against hydrolysis. For instance, several Zn2+- and Cu2+-based frameworks have been synthesized with highly hydrophobic surfaces that adsorb only very small amounts of water at low pressures.32–38 Additionally, several MOFs have been synthesized with water-repellant groups incorporated directly into the organic ligands as a means of protecting the metal core from water.35,36,39 In either case, CO2 sorption properties were low or not reported. Introducing functional groups leads to complexity due to intricate coordination of functional groups to metal ions and poor thermal stability. The ease of introducing different organic functional groups onto organic ligands in MOFs compared to inorganic porous solids has triggered significant interest in post-synthetic modification. Post-synthetic modification (PSM) has also become an essential tool in developing systems with different functionalities inside the framework while introducing new chemical as well as physical properties that cannot be realized using simple MOFs.40–45 However, this approach doesn’t come without a cost, which sometimes involves a decrease in surface area and/or guest capacity.46,47 In one promising study, a MOF (Banasorb-22) isostructural to MOF-5 was synthesized, wherein each of the terephthalic acid linkers was modified with a trifluoromethoxy substituent.48 The material showed great stability in boiling water but no gas capture was reported. There are a number of publications where the effect of PSM on different MOF systems is studied. For instance, amine-containing MOFs can readily undergo PSM to form amide-functionalized MOFs.49,50 It has been hypothesized that the introduction of hydrophobic alkyl chains via PSM could improve the moisture resistance and changes the physical properties (i.e., hydrophobicity) of these MOFs.51 As an example, IRMOFs constructed of 1,4-benzenedicarboxylate (BDC, IRMOF-1), 2-amino-1,4-benzenedicarboxylate (NH2-BDC, IRMOF-3), and MIL-53(Al)-NH2 were functionalized with alkyl anhydrides via PSM showing high resistance to moisture.48 No CO2 adsorption was reported on these functionalized materials. Finally, it is important to note that the interaction of water and CO2 with the porous framework in most cases is via the metal coordination center facing the cavity. Both interactions are electrostatic in nature; with water it is a metal iondipole interaction while for CO2 is metal ion-quadrupole interaction. Consequently, one has to be very careful in protecting these metal centers with chemical functionalities since, although it will decrease the framework hydrophilicity, it will also decrease pore volume (MOF sorption capacity) thus compromising the material affinity toward CO2. Herein, we intro-
Dalton Trans.
Dalton Transactions
duce a simple approach to modify MOF materials to increase their moisture stability as well as minimize water adsorption without compromising CO2 sorption capacity of these materials.
Results and discussion Preliminary experimental results demonstrate that it is possible that MOFs can be modified to achieve increased hydrophobicity and stability towards water without the need of complex post-functionalization chemistries or introducing chemical functionalities inside the MOF cavities. Our approach consists of incorporating amphiphilic moieties via physical adsorption through their polar heads to the metal centers on the external surface of the MOFs. In this fashion, the pendant hydrophobic group on top of the surface of the MOF acts as a barrier against water and minimizes water vapor penetration/adsorption while maintaining the material’s CO2 capacity. The rationale behind this concept is the fact that the open metal centers and the ligand chemical functionalities inside the MOF framework are still available to interact with CO2 and that the surface area of the material would not be essentially compromised. To prove the concept, we synthesized different MOFs (MIL-101(Cr), and NiDOBDC using known procedures.27,52 NiDOBDC were chosen because of its decent CO2 capacities (∼28 wt% at 1 bar) however they suffer from moisture instability upon exposure to water for longer periods of time as it will be shown later.1,13,53 MIL-101(Cr) was chosen due to its characteristic high water sorption capacity (∼150 wt%) but poor CO2 sorption capacity at low pressure. Both porous materials were treated with commercially available Pluronic P123, (here onwards P1, average molecular weight 5800 Da)-chloroform solution. This non-ionic surfactant is a triblock copolymer based on poly(ethylene glycol)-poly( propylene glycol)-poly(ethylene glycol) mainly composed of hydroxyl groups at both ends of a hydrophobic ether-based long chain group (Fig. 1). Fig. 2A illustrates the room temperature water adsorption isotherms of untreated MIL101(Cr) as compared to the P1-treated material. The adsorption isotherm shows a significant decrease in water uptake on the P1-modified MIL101(Cr) with respect to the unmodified MIL101(Cr) at relative humidity (RH) above 50%. At 95% RH, P1-MIL101(Cr) showed nearly three times lower water uptake with respect to unmodified MIL101(Cr). P1-modified MIL101(Cr) was cycled three times for water sorption analysis without apparent change in the water uptake at room temperature, clearly demonstrating the stability of the
Fig. 1 Pluronic P123, X = 20, Y = 70, Z = 20 with average Mw = 5800 Da.
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Dalton Transactions
Fig. 2 RT water sorption (A) and low pressure RT CO2 sorption (B) in unmodified and P1-modified MIL101(Cr).
P1-modified MIL101(Cr) (Fig. 2). The treated materials were further activated by methanol exchange followed by heating under vacuum at 80 °C for 10 hours and tested for CO2 sorption. CO2 isotherms of P1-modified MIL101 (Cr) show no obvious alteration in CO2 uptake as compared to untreated MIL101(Cr) at both low and high CO2 pressure (Fig. 2B and S1†). Similarly, water adsorption on P1-modified NiDOBDC decreased by about 60% compared to the unmodified NiDOBDC (Fig. 3A) while maintaining the CO2 capacity (26% w/w at 1 bar) as shown in Fig. 3B. These results are of paramount importance and demonstrate for the first time that a MOF material can be readily functionalized on the surface with a surfactant that not only increases the hydrophobic behavior significantly but do not compromise its CO2 sorption capacity. To gain more insights on the hydrophobic/hydrophilic characteristics of P1-modified MOFs, we performed contact angle measurements on pelletized samples.38,47 Contact angle provides quantitative information on the wettability of a solid surface and it is measured at the liquid/solid surface interface. Materials with contact angles below 90° are considered hydrophilic materials, contact angles between 90–150° are considered hydrophobic materials and materials with contact angle values above 150° are considered as superhydrophobic materials. In the case of powder samples, determining contact angles is not a trivial task and sometimes leads to smaller contact angles due to the roughness and poro-
This journal is © The Royal Society of Chemistry 2015
Paper
Fig. 3 RT water adsorption isotherms on unmodified as well as P1modified NiDOBDC. Unmodified MOF captures twice the mass of water as compared to P1-NiDOBDC at RH above 20% (A). RT CO2 adsorption isotherms on unmodified as well as P1-modified NiDOBDC showing similar CO2 capacities in the entire pressure range (B).
sity of the pellets produced.54–57 Nevertheless, to provide a approximate estimation of the relative hydrophobicity pellets of unmodified MOFs as well as P1-modified MOFs were obtained by applying hydraulic pressure (up to 7 Ton cm−2 for 120 seconds). As prepared MOF pellets were used to measure contact angle using a Rame-Hart goniometer (details in ESI†). Table S1† and Fig. 4 show the results. Unmodified MIL101 and NiDOBDC exhibited significant wettability and near-zero water droplet life time absorbing water instantaneously making contact angle measurements very difficult to perform clearly illustrate their hydrophilic behavior. However, P1-modified MIL101(Cr) showed contact angles around 25–27° with a
Fig. 4 Photographs of P1-modified MIL101(Cr) (left) and unmodified MIL101(Cr) pellets during contact angle measurements.
Dalton Trans.
View Article Online
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Paper
droplet life time of about 30 seconds. P1-modified NiDOBDC pellets had a rough surface with contact angles 22–24° with a droplet life time about 40 seconds. Although the values of contact angle are smaller than expected, the relative lifetime of the water droplet in P1-modified MOFs as compared to unmodified MOFs clearly indicates the more hydrophobic nature of the P1-modified MOFs. These results confirm that an amphiphilic surfactant functionalizing the surface of MOF materials provides a hydrophobic barrier that delays the diffusion of water into the MOF. To learn about the MOFs stability toward water, P1-NiDOBDC was subjected to a number of water adsorption/ desorption cycles (with conditions identical to the ones shown on Fig. 3) followed by CO2 sorption analysis in between cycles. Details about the porous material activation followed by sorption analysis on each cycle are described in the Methods section. Fig. 5A illustrates the decreases in CO2 sorption capacity in unmodified-NiDOBDC with the number of water exposure cycles, particularly after the third cycle. After four water adsorption/desorption cycles, the CO2 uptake at 1 bar for NiDOBDC decreases nearly three times as compared to the original MOF sorption capacity. On the other hand, P1-modified NiDOBDC showed constant sorption values at all pressures independent of the number of water exposure cycles (Fig. 5B). This is a very interesting result because surface coating of MOFs with P1 not only results in increasing the materials hydrophobic behavior but also increases their stability while maintaining CO2 sorption capacity.
Fig. 5 CO2 adsorption isotherms on unmodified (A) and P1-functionalized NiDOBDC (B). Notice the decrease in CO2 capacity on unmodified NiDOBDC, particularly after four water exposure cycle.
Dalton Trans.
Dalton Transactions
Fig. 6 Powder XRD spectra of unmodified and P1-functionalized before and after water exposure showing the stability of the P1NiDOBDC after four CO2/water sorption/desorption cycles as compared to the unstable unmodified NiDOBDC.
Powder X-ray diffraction (PXRD) studies on NiDOBDC and P1-modified NiDOBDC before and after four water adsorption/ desorption cycles are shown in Fig. 6. It can be observed that for the original porous material there is no obvious difference in the crystalline structure after functionalization (Fig. 6A and 6C) More importantly, the PXRD pattern of P1-modified NiDOBDC shows no obvious change after four CO2/wateradsorption/desorption cycles (Fig. 6D). On the other hand, unmodified NiDOBDC does show indication of structural decomposition as evidenced by the disappearance of most of the peaks, particularly at large diffraction angles (Fig. 6B). These results are in complete agreement with the observed decrease in CO2 sorption capacity on unmodified NiDOBDC after the fourth water vapor exposure. Functionalization of the MOF material with the non-ionic surfactant proves not only to significantly reduce the MOF water sorption ability but, as importantly, to render stability against water-induced hydrolysis/decomposition. These two desirable properties can be achieved without compromising the CO2 sorption capacity of the materials in the case of NiDOBDC and MIL-101(Cr). Thermogravimetric (TG) analysis was performed on NiDOBDC and P1-modified NiDOBDC after methanol exchange and is shown in Fig. S2.† It is observed that unmodified NiDOBDC has a mass loss of 18% at low temperatures and then is stable to temperatures as high as 300 °C. P1NiDOBDC showed two mass losses, a 3% mass loss at low temperatures similarly to unmodified NiDOBDC followed by a gradual mass loss at higher temperatures (>230 °C) which can be assigned to Pluronic P123 desorption/decomposition from the MOF surface. Lower solvent loss was observed on P1NiDOBDC which could be attributed to the hydrophobic features of the functionalized material that minimize the incorporation of methanol molecules inside the framework cavities. From TG analysis it was estimated a 14% loading of Pluronic
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Dalton Transactions
P123 on the surface of NiDOBDC. Scanning electron microscopy (SEM) was performed on unmodified and P1-functionalized MIL-101(Cr), and NiDOBDC (Fig. S3†). There were no obvious differences observed in MOF surface morphology between the unmodified and P1-functionalized materials. However, electron charging on the surface of the samples modified with Pluronic P123 was observed in both MOF materials as expected due to the non-conducting nature of the adsorbed surfactants on the surface of the porous materials. This provides additional evidence that Pluronic P123 was indeed adsorbed on the surface of each MOF. Surface area analysis using nitrogen sorption isotherms collected at 77 K was conducted on NiDOBDC and P1-NiDOBDC to learn about possible pore obstruction by the surfactant on the P1-modified MOF (Fig. S4†). The results show that the permanent porosity is retained after functionalization and the surface area of functionalized P1-NiDOBDC (830 m2 g−1) is slightly lower than unmodified NiDOBDC (912 m2 g−1), but within the experimental error. Although slightly lower than the unmodified NiDOBDC, the surface area of modified P1DOBDC is significantly higher than many MOFs studied for CO2 separation.58,59 This confirms the hypothesis that deep penetration of surfactants inside the cavities is fairly unlikely compromising the overall pore capacity and framework selectivity during and after functionalization of the MOF. This is due to the fact that the length of Pluronic P123 (∼28 nm) is significantly larger than the pore size of NiDOBDC and MIL101(Cr) (nearly 20 X larger than the largest cavity size corresponding to MIL101(Cr), 1.6 nm). To further elucidate the effect of P123 on NiDOBDC and understand the possible mechanism involved during water sorption in these materials, the water uptake was plotted as a function of step changes in RH for both P1-NiDOBDC and unmodified NiDOBDC (Fig. 7). In addition, the rate of water
Fig. 7 Water isotherms as a function of time on unmodified NiDOBDC and on P1-modified NiDOBDC showing the times at which water sorption at a new %RH takes place. Note the nearly half of the water uptake of NiDOBDC occurs with 20 ≤ RH ≤ 30%.
This journal is © The Royal Society of Chemistry 2015
Paper
uptake (δ wt%/δt ) was plotted as a function of time (Fig. S5†). Water adsorption in P1-modified and unmodified NiDOBDC are essentially equivalent up to RH = 20%. However at higher RH, particularly between RH = 20–40%, the rates of water adsorption are up to five times higher in unmodified NiDOBDC as compared to P1-modified NiDOBDC at a given time. In addition, the water sorption rates in P1-modified NiDOBDC decay much faster (Fig. S5†) than the rates of unmodified NiDOBDC resulting in shorter equilibrium times (Fig. 7). The combination of lower rates of water adsorption and shorter equilibrium times translates to a significantly lower water uptake by P1-modified NiDOBDC compared to NiDOBDC (nearly half, 45 wt% vs. 23 wt%, Fig. 3 and 7). With these results in hand, we hypothesized that pore obstruction by the surfactant during water sorption might be responsible for the rapid sorption equilibrium and the resulting reduced water capacity of P1-modified NiDOBDC. This equilibrium effect which implies a fundamental and permanent change in the water sorption capacity is of great significance because the modified-MOF materials could be potentially used for thousands of cycles without structure impairment or pore saturation. Then the question is, if it is an equilibrium effect, what is the chemical mechanism? Since the surfactant should be constrained to the outer surface of the MOF material, how could such a surface effect impart a bulk structural change in the adsorption affinity? Molecular dynamics (MD) simulations60 were then performed using Amber 9 Package using the Force field parameters for MOF were taken from previous work to (1) understand the interaction between the NiDOBDC fragments and P-123 in chloroform during MOF functionalization and (2) to learn about the interaction of the P1-functionalized NiDOBDC with water and CO2.60,61 The molecular formula of polymer (P-123) unit used in these MD simulations is (HO(CH2CH2O)2(CH2CH(CH3)O)2(CH2CH2O)2H).62–64 To model the MOF functionalization process with P-123, a portion of a rigid fragment of MOF was taken from the NiMOF-74 crystal structure. The MOF fragment is then centered in a box of size 80 × 80 × 80 Å. To this fragment 20 units of P-123 were added and then solvated with chloroform (ESI†). Snapshots of simulation box containing NiDOBDC and P-123 solvated in chloroform are shown in Fig. S6.† Fig. S7† illustrates the radial distribution functions (RDF) computed between the Ni-atoms of MOF and the oxygen and carbon atoms of P-123. The NiMOF-OP-123 RDF has sharp and intense peak at ∼2.2 Å. As expected we observed strong interaction between the positive Ni-atoms of the MOF and the partially negative O-atoms of P-123. In Fig. S7† we show the RDF between the oxygen atoms of MOF linker units and the terminal hydrogen atoms of P-123. This RDF has a peak at ∼1.7 Å, indicating strong OMOF–HOP-123 interactions. These two kinds of interactions are primary governing factors for binding of P-123 to MOF material. Two systems were prepared by randomly adding three hundred molecules of water or carbon dioxide to MOF fragment equilibrated with the polymer molecules wrapped around it (Fig. S8†). Fig. 8A shows the front view of the initial
Dalton Trans.
View Article Online
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Paper
Dalton Transactions
It is interesting to note that although P1-modified NiDOBDC can still adsorb about 23 wt% of water at RH = 95%, upon regeneration we did not observe any loss in CO2 capacity after 4 cycles. However, we observed significant loss in CO2 capacity of unmodified NiDOBDC ( particularly after the third regeneration cycle) under identical conditions (Fig. 5). This is also evident by the crystalline structure loss of the unmodified NiDOBDC after the third water exposure cycle (Fig. 6). Experimental as well as MD results clearly suggest that the functionalization of the MOF with P123 not only reduces the water capacity of NiDOBDC significantly but it also brings stability upon water exposure without compromising CO2 capacity of the porous material.
Conclusions
Fig. 8 MD snapshot (A) initial frame with water, (B) final frame with water after 150 ps, (C) initial frame with CO2 and (D) final frame with CO2 after 2 ns.
snapshot of the MD simulation and Fig. 8B shows the front view of the MD snapshots after 150 ps of water exposure. Within 150 ps it is observed that the water gets adsorbed on the MOF surface. Also, it is interesting to note that when the water goes into the pore of the MOF fragment they also carry polymer units along with them. Polymer units that try to diffuse into the pores are colored in magenta and the units that remain at the outer surface are colored in white in Fig. 8. The polymer molecules then partially occupy the MOF cavities and restrict the number of water molecules diffusing and compromising the overall water uptake. However, the polymer molecules do not significantly compromise the overall pore capacity and volume. This seems to be the reason for the rapid water sorption equilibrium and the resulting decrease in water loading observed in the water isotherms of P1-NiDOBDC (Fig. 3 and 6). Therefore, the MD simulation studies presented in Fig. 8 are in agreement with the surface area measurements and CO2 sorption studies illustrated in Fig. 5. Now, it is expected that this competing mechanism sorbate/polymer observed for water would not occur during CO2 sorption since the CO2 isotherms obtained for P1-modified NiDOBDC and unmodified NiDOBDC are identical. Fig. 8C and 8D show similar snapshots for the case of exposure of the P1-functionalized NiDOBDC to CO2. It is observed that no P-123 units diffuse into the MOF pores during CO2 diffusion. To ensure a long enough exposure to CO2 MD simulations were ran for longer times. Fig. 8D shows a snapshot after 2 ns exposure of the functionalized MOF material to CO2 showing no polymer diffusing inside the pores during CO2 adsorption, i.e., no polymer/sorbate competing effect was identified when CO2 diffused inside the MOF pores.
Dalton Trans.
The present work provides experimental and modeling evidence that incorporating hydrophobic molecules on the external surface of MOFs is a very important approach for stabilization of these materials against water while maintaining the CO2 sorption properties. It was demonstrated that the surfactant adsorbed on the surface of a porous MOF material is carried into the pores when exposed to water thus compromising the volume available for water. This results in significant reduction of water sorption while bringing stability against water. This concept can be applied to increase stability of MOFs in gas separation, catalysis and sensing applications where water is present. For example, a frequent issue in catalysis is water poisoning.26–28 Indeed; catalytic reactions can be delayed or simply inhibited due to poisoning effects originating from moisture in the air. The incorporation of a hydrophobic shell would minimize the penetration of water to the catalytic sites, hence, rendering water stability and longer catalyst lifetimes.
Experimental Materials synthesis NiDOBDC and MIL-101(Cr) were synthesized following literature procedures.32,58,59 Surfactant functionalization was performed by introducing the evacuated MOF materials in P1 solutions for a couple of hours and then washed several times to remove excess of P1. The materials were evacuated and water and CO2 sorption isotherms were collected at RT. Activation of the modified and unmodified materials were performed at 80 °C after replacing solvents from the synthesis with low boiling point solvents, methanol or chloroform, to minimize potential surfactant desorption. Synthesis of P123 modified MOF materials was performed by simply immersing for 24 h at room temperature evacuated NiDOBDC or MIL-101(Cr) in a 10 wt% chloroform solution of P123. Then, the materials are rinsed with fresh chloroform followed by centrifugation (three times) and finally dried with a gentle flow of N2 or air.
This journal is © The Royal Society of Chemistry 2015
View Article Online
Dalton Transactions
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Characterization methods SEM images were obtained using FEI Helios 600 NanoLab FIB-SEM. CO2 and Nitrogen adsorption–desorption isotherms were collected using a Quantachrome autosorb-6 automated gas sorption. Brunauer–Emmet–Teller (BET) surface area was calculated from the nitrogen isotherm curves ranging from 0.1 to 0.3 of relative pressure. Water sorption isotherms were collected at RT using VTI-SA+ Vapor-sorption analyzer. Thermal stability of CMOFs was investigated using a TG (NETZSCH TG 209 F1) under nitrogen (N2). Powder X-ray diffraction (XRD) analysis was obtained from powder samples placed on 0.8 mm capillaries. A custom-built XYZ environmental stage inside a Bruker-AXS D8 Discover XRD unit equipped with a rotating Cu anode (1.54 Å), göbel mirror, 0.5 mm collimator, and 0.5 mm pin hole (Madison, WI). EMOF films were aligned using a video laser alignment system. Patterns were collected with a GADDS® area detector positioned at 21.5° 2θ with a measured distance from the sample of 15 cm. Collection of individual XRD tracings required 60 s with power settings of 45 kV and 200 mA. Images were processed with Bruker-AXS GADDS software before importing into JADE® XRD software to obtain peak positions and intensities. Contact Angle Goniometry. Contact angle measurements of water with the MOF pellet surfaces were made using a Rame-Hart goniometer. Reported contact angles are static.
Acknowledgements This work was supported by PNNL Energy Conversion Initiative and Office of Fossil Energy, U.S. Department of Energy (DOE). The U.S. Department of Energy (DOE), Office of Basic Energy Sciences (BES), Division of Chemical Sciences, Geosciences and Biosciences funded the work performed by L. X. D. PNNL is a multiprogramming laboratory operated by Battelle Memorial Institute for the Department of Energy under Contract DE-AC05-76RL01830.
Notes and references 1 J. Canivet, A. Fateeva, Y. M. Guo, B. Coasne and D. Farrusseng, Chem. Soc. Rev., 2014, 43, 5594–5617. 2 W. G. Lu, Z. W. Wei, Z. Y. Gu, T. F. Liu, J. Park, J. Park, J. Tian, M. W. Zhang, Q. Zhang, T. Gentle, M. Bosch and H. C. Zhou, Chem. Soc. Rev., 2014, 43, 5561–5593. 3 J. R. Li, Y. G. Ma, M. C. McCarthy, J. Sculley, J. M. Yu, H. K. Jeong, P. B. Balbuena and H. C. Zhou, Coord. Chem. Rev., 2011, 255, 1791–1823. 4 M. P. Suh, H. J. Park, T. K. Prasad and D. W. Lim, Chem. Rev., 2012, 112, 782–835. 5 E. Barea, C. Montoro and J. A. R. Navarro, Chem. Soc. Rev., 2014, 43, 5419–5430. 6 L. Heinke, Z. G. Gu and C. Woll, Nat. Commun., 2014, 5. 7 H. Furukawa, K. E. Cordova, M. O’Keeffe and O. M. Yaghi, Science, 2013, 341, 974–986.
This journal is © The Royal Society of Chemistry 2015
Paper
8 L. J. Murray, M. Dinca and J. R. Long, Chem. Soc. Rev., 2009, 38, 1294–1314. 9 H. Li, M. Eddaoudi, M. O’Keeffe and O. M. Yaghi, Nature, 1999, 402, 276–279. 10 G. Q. Li, H. Kobayashi, J. M. Taylor, R. Ikeda, Y. Kubota, K. Kato, M. Takata, T. Yamamoto, S. Toh, S. Matsumura and H. Kitagawa, Nat. Mater., 2014, 13, 802–806. 11 B. L. Chen, S. C. Xiang and G. D. Qian, Acc. Chem. Res., 2010, 43, 1115–1124. 12 S. C. Xiang, Y. B. He, Z. J. Zhang, H. Wu, W. Zhou, R. Krishna and B. L. Chen, Nat. Commun., 2012, 3, 1–9. 13 L. Q. Ma, C. Abney and W. B. Lin, Chem. Soc. Rev., 2009, 38, 1248–1256. 14 P. K. Thallapally, C. A. Fernandez, R. K. Motkuri, S. K. Nune, J. Liu and C. H. F. Peden, Dalton Trans., 2010, 39, 1692–1694. 15 C. A. Fernandez, P. K. Thallapally, R. K. Motkuri, S. K. Nune, J. C. Sumrak, J. Tian and J. Liu, Cryst. Growth Des., 2010, 10, 1037–1039. 16 J. Della Rocca and W. B. Lin, Eur. J. Inorg. Chem., 2010, 3725–3734. 17 W. J. Rieter, K. M. L. Taylor and W. B. Lin, J. Am. Chem. Soc., 2007, 129, 9852–9853. 18 G. K. H. Shimizu, J. M. Taylor and S. Kim, Science, 2013, 341, 354–355. 19 C. Y. Sun, C. Qin, X. L. Wang and Z. M. Su, Expert Opin. Drug Delivery, 2013, 10, 89–101. 20 L. Bellarosa, S. Calero and N. Lopez, Phys. Chem. Chem. Phys., 2012, 14, 7240–7245. 21 J. J. Low, A. I. Benin, P. Jakubczak, J. F. Abrahamian, S. A. Faheem and R. R. Willis, J. Am. Chem. Soc., 2009, 131, 15834–15842. 22 A. Comotti, S. Bracco, P. Sozzani, S. Horike, R. Matsuda, J. Chen, M. Takata, Y. Kubota and S. Kitagawa, J. Am. Chem. Soc., 2008, 130, 13664–13672. 23 J. B. DeCoste, G. W. Peterson, B. J. Schindler, K. L. Killops, M. A. Browe and J. J. Mahle, J. Mater. Chem. A, 2013, 1, 11922–11932. 24 J. Ehrenmann, S. K. Henninger and C. Janiak, Eur. J. Inorg. Chem., 2011, 471–474. 25 J. B. DeCoste, G. W. Peterson, H. Jasuja, T. G. Glover, Y. G. Huang and K. S. Walton, J. Mater. Chem. A, 2013, 1, 5642–5650. 26 M. Kim, J. F. Cahill, H. H. Fei, K. A. Prather and S. M. Cohen, J. Am. Chem. Soc., 2012, 134, 18082– 18088. 27 L. Bromberg, Y. Diao, H. M. Wu, S. A. Speakman and T. A. Hatton, Chem. Mater., 2012, 24, 1664–1675. 28 X. L. Liu, Y. S. Li, Y. J. Ban, Y. Peng, H. Jin, H. Bux, L. Y. Xu, J. Caro and W. S. Yang, Chem. Commun., 2013, 49, 9140– 9142. 29 K. S. Park, Z. Ni, A. P. Cote, J. Y. Choi, R. D. Huang, F. J. Uribe-Romo, H. K. Chae, M. O’Keeffe and O. M. Yaghi, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 10186–10191. 30 K. Kida, M. Okita, K. Fujita, S. Tanaka and Y. Miyake, CrystEngComm, 2013, 15, 1794–1801.
Dalton Trans.
View Article Online
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Paper
31 J. A. Thompson, N. A. Brunelli, R. P. Lively, J. R. Johnson, C. W. Jones and S. Nair, J. Phys. Chem. C, 2013, 117, 8198– 8207. 32 S. Horike, D. Tanaka, K. Nakagawa and S. Kitagawa, Chem. Commun., 2007, 3395–3397. 33 N. M. Padial, E. Q. Procopio, C. Montoro, E. Lopez, J. E. Oltra, V. Colombo, A. Maspero, N. Masciocchi, S. Galli, I. Senkovska, S. Kaskel, E. Barea and J. A. R. Navarro, Angew. Chem., Int. Ed., 2013, 52, 8290–8294. 34 C. Yang, U. Kaipa, Q. Z. Mather, X. P. Wang, V. Nesterov, A. F. Venero and M. A. Omary, J. Am. Chem. Soc., 2011, 133, 18094–18097. 35 D. Y. Ma, Y. W. Li and Z. Li, Chem. Commun., 2011, 47, 7377–7379. 36 H. Jasuja, N. C. Burtch, Y. G. Huang, Y. Cai and K. S. Walton, Langmuir, 2013, 29, 633–642. 37 L. Bellarosa, J. J. Gutierrez-Sevillano, S. Calero and N. Lopez, Phys. Chem. Chem. Phys., 2013, 15, 17696–17704. 38 T. H. Chen, I. Popov, O. Zenasni, O. Daugulis and O. S. Miljanic, Chem. Commun., 2013, 49, 6846–6848. 39 R. Vaidhyanathan, S. Natarajan and C. N. R. Rao, Dalton Trans., 2003, 1459–1464. 40 A. D. Burrows, C. G. Frost, M. F. Mahon and C. Richardson, Angew. Chem., Int. Ed., 2008, 47, 8482–8486. 41 P. Roy, A. Schaate, P. Behrens and A. Godt, Chem. – Eur. J., 2012, 18, 6979–6985. 42 M. Servalli, M. Ranocchiari and J. A. Van Bokhoven, Chem. Commun., 2012, 48, 1904–1906. 43 W. T. Zhu, C. He, X. Wu and C. Y. Duan, Inorg. Chem. Commun., 2014, 39, 83–85. 44 J. D. Evans, C. J. Sumby and C. J. Doonan, Chem. Soc. Rev., 2014, 43, 5933–5951. 45 A. M. Shultz, A. A. Sarjeant, O. K. Farha, J. T. Hupp and S. T. Nguyen, J. Am. Chem. Soc., 2011, 133, 13252–13255. 46 S. Aguado, J. Canivet, Y. Schuurman and D. Farrusseng, J. Catal., 2011, 284, 207–214. 47 J. G. Nguyen and S. M. Cohen, J. Am. Chem. Soc., 2010, 132, 4560–4561.
Dalton Trans.
Dalton Transactions
48 T. J. Wu, L. J. Shen, M. Luebbers, C. H. Hu, Q. M. Chen, Z. Ni and R. I. Masel, Chem. Commun., 2010, 46, 6120– 6122. 49 Y. Cao, F. J. Song, Y. X. Zhao and Q. Zhong, J. Environ. Sci., 2013, 25, 2081–2087. 50 T. M. McDonald, W. R. Lee, J. A. Mason, B. M. Wiers, C. S. Hong and J. R. Long, J. Am. Chem. Soc., 2012, 134, 7056–7065. 51 K. K. Tanabe and S. M. Cohen, Chem. Soc. Rev., 2011, 40, 498–519. 52 S. R. Caskey, A. G. Wong-Foy and A. J. Matzger, J. Am. Chem. Soc., 2008, 130, 10870–10871. 53 A. C. Kizzie, A. G. Wong-Foy and A. J. Matzger, Langmuir, 2011, 27, 6368–6373. 54 E. Chibowski and R. Perea-Carpio, Adv. Colloid Interface Sci., 2002, 98, 245–264. 55 R. J. Hunter, Foundations of Colloid Science, Oxford University Press, 2nd edn, 2001. 56 P. Chassin, C. Jouany and H. Quiquampoix, Clay Miner., 1986, 21, 899–907. 57 E. Chibowski, L. Holysz, A. Zdziennicka and F. Gonzalezcaballero, Surfactants in Solution, Marcel Dekker, New York, 1996. 58 Y. Liu, Z. U. Wang and H.-C. Zhou, Greenhouse Gases: Sci. Technol., 2012, 2, 239–259. 59 J. Liu, P. K. Thallapally, B. P. McGrail, D. R. Brown and J. Liu, Chem. Soc. Rev., 2012, 41, 2308–2322. 60 X. Q. Sun, C. D. Wick, P. K. Thallapally, B. P. McGrail and L. X. Dang, J. Phys. Chem. B, 2011, 115, 2842– 2849. 61 H. J. C. Berendsen, J. R. Grigera and T. P. Straatsma, J. Phys. Chem., 1987, 91, 6269–6271. 62 J. M. Wang, R. M. Wolf, J. W. Caldwell, P. A. Kollman and D. A. Case, J. Comput. Chem., 2004, 25, 1157–1174. 63 J. M. Wang, R. M. Wolf, J. W. Caldwell, P. A. Kollman and D. A. Case, J. Comput. Chem., 2005, 26, 114–114. 64 J. M. Wang, W. Wang, P. A. Kollman and D. A. Case, J. Mol. Graphics Modell., 2006, 25, 247–260.
This journal is © The Royal Society of Chemistry 2015
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
PAPER
Cite this: DOI: 10.1039/c5dt00606f
View Journal
Hydrophobic and moisture-stable metal–organic frameworks† Carlos A. Fernandez,* Satish K. Nune,* Harsha V. Annapureddy, Liem X. Dang, B. Peter McGrail, Feng Zheng, Evgueni Polikarpov, David L. King, Charles Freeman and Kriston P. Brooks Metal–organic frameworks (MOFs) have proved to be very attractive for applications including gas storage, separation, sensing and catalysis. In particular, CO2 separation from flue gas in post-combustion processes is one of the main focuses of research among the scientific community. One of the major issues that are preventing the successful commercialization of these novel materials is their high affinity towards water that not only compromises gas sorption capacity but also the chemical stability. In this paper, we demonstrate a novel post-synthesis modification approach to modify MOFs towards increasing hydrophobic behaviour and chemical stability against moisture without compromising CO2 sorption capacity. Our approach consists of incorporating hydrophobic moieties on the external surface of the MOFs via physical adsorption. The rationale behind this concept is to increase the surface hydrophobicity in the porous materials without the need of introducing bulky functionalities inside the pore which compromises the sorption capacity toward other gases. We herein report preliminary results on routinely studied MOF materials [MIL-101(Cr) and NiDOBDC] demonstrating that the polymer-modified MOFs retain CO2 sorption capacity while reducing the water adsorption up to three times, with respect to the un-modified materials, via an equilibrium effect. Furthermore, the water stability of the polymer-functionalized MOFs is significantly higher than the water stability of the bare material. Molecular dynamic simu-
Received 10th February 2015, Accepted 1st May 2015
lations demonstrated that this equilibrium effect implies a fundamental and permanent change in the
DOI: 10.1039/c5dt00606f
water sorption capacity of MOFs. This approach can also be employed to render moisture stability and selectivity to MOFs that find applications in gas separations, catalysis and sensing where water plays a criti-
www.rsc.org/dalton
cal role in compromising MOF performance and recyclability.
Introduction Metal–organic frameworks (MOFs) are porous, crystalline 2D and 3D structures that possess very important features including high surface areas, controlled porosities, good thermal stability and the possibility of adjusting the pore size and chemical environment.1–6 For these reasons, they are increasingly in demand for applications in gas storage,7–9 separation,10–12 catalysis,13–15 imaging,16,17 proton conductors18 and drug delivery.17,19 Despite their potentials, there has been a limited success in the practical applications of MOFs, including CO2 removal in post-combustion processes.1 One of
Applied Functional Materials, Hydrocarbon Processing Group, Energy and Environment Directorate, Pacific Northwest National Laboratory, Richland, Washington 99352, USA. E-mail: [email protected], [email protected] † Electronic supplementary information (ESI) available: Detailed adsorption isotherms of water and CO2, SEM image of modified DOBDC, water sorption kinetics plot and simulation snapshots of NiDOBDC fragment in solvent chloroform and solution containing P123. See DOI: 10.1039/c5dt00606f
This journal is © The Royal Society of Chemistry 2015
the most significant drawbacks is their affinity towards water that renders mechanical and chemical instability. Particularly, zinc/carboxylate-based MOFs including isoreticular MOFs (IRMOFs) are extremely sensitive to moisture. This instability is related to the nature of the bonding (coordination bonding) between zinc atoms and carboxylate ligands.1 Other transition metal-based MOFs have also shown water instability.20–23 Copper-trimesic acid based MOF (HKUST-1) and Al(OH)(2,6ndc) (ndc = naphthalene dicarboxylate; DUT-4) are unstable in the presence of water vapor.21,23 Thorough understanding of the aspects behind the water stability of MOFs is still missing; this is demonstrated, e.g., by the fact that materials of the same family such as Al(OH)(1,4-ndc) and Al(OH)(2,6-ndc) (ndc = naphthalenedicarboxylate) show very different behavior, with the first material being stable upon water adsorption while the second decomposes.24 Some MOF materials, such as the MIL series (MIL101-Cr, MIL100-Fe), UIO-66 (containing ZrCl4 and 1,4-benzenedicarboxylate linkers (BDC) or zeolite imidazole frameworks (ZIFs) including the well-known ZIF-8 are resistant to moisture.25–30 However, their CO2 sorption capacities at low
Dalton Trans.
View Article Online
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Paper
pressure region (0 to 0.1 atm) is low in CO2/H2O gas mixtures.31 One possible reason for such low uptake in MILs is due to the competing selectivity of these materials toward water and non-availability of open metal centers that favor CO2 sorption. Stabilizing these MOF materials against ambient humidity and increasing the hydrophobic character would make MOFs more suitable for specialized and industrial applications including removal of CO2 from flue gas in postcombustion processes. A number of approaches were reported in the literature to stabilize these materials against hydrolysis. For instance, several Zn2+- and Cu2+-based frameworks have been synthesized with highly hydrophobic surfaces that adsorb only very small amounts of water at low pressures.32–38 Additionally, several MOFs have been synthesized with water-repellant groups incorporated directly into the organic ligands as a means of protecting the metal core from water.35,36,39 In either case, CO2 sorption properties were low or not reported. Introducing functional groups leads to complexity due to intricate coordination of functional groups to metal ions and poor thermal stability. The ease of introducing different organic functional groups onto organic ligands in MOFs compared to inorganic porous solids has triggered significant interest in post-synthetic modification. Post-synthetic modification (PSM) has also become an essential tool in developing systems with different functionalities inside the framework while introducing new chemical as well as physical properties that cannot be realized using simple MOFs.40–45 However, this approach doesn’t come without a cost, which sometimes involves a decrease in surface area and/or guest capacity.46,47 In one promising study, a MOF (Banasorb-22) isostructural to MOF-5 was synthesized, wherein each of the terephthalic acid linkers was modified with a trifluoromethoxy substituent.48 The material showed great stability in boiling water but no gas capture was reported. There are a number of publications where the effect of PSM on different MOF systems is studied. For instance, amine-containing MOFs can readily undergo PSM to form amide-functionalized MOFs.49,50 It has been hypothesized that the introduction of hydrophobic alkyl chains via PSM could improve the moisture resistance and changes the physical properties (i.e., hydrophobicity) of these MOFs.51 As an example, IRMOFs constructed of 1,4-benzenedicarboxylate (BDC, IRMOF-1), 2-amino-1,4-benzenedicarboxylate (NH2-BDC, IRMOF-3), and MIL-53(Al)-NH2 were functionalized with alkyl anhydrides via PSM showing high resistance to moisture.48 No CO2 adsorption was reported on these functionalized materials. Finally, it is important to note that the interaction of water and CO2 with the porous framework in most cases is via the metal coordination center facing the cavity. Both interactions are electrostatic in nature; with water it is a metal iondipole interaction while for CO2 is metal ion-quadrupole interaction. Consequently, one has to be very careful in protecting these metal centers with chemical functionalities since, although it will decrease the framework hydrophilicity, it will also decrease pore volume (MOF sorption capacity) thus compromising the material affinity toward CO2. Herein, we intro-
Dalton Trans.
Dalton Transactions
duce a simple approach to modify MOF materials to increase their moisture stability as well as minimize water adsorption without compromising CO2 sorption capacity of these materials.
Results and discussion Preliminary experimental results demonstrate that it is possible that MOFs can be modified to achieve increased hydrophobicity and stability towards water without the need of complex post-functionalization chemistries or introducing chemical functionalities inside the MOF cavities. Our approach consists of incorporating amphiphilic moieties via physical adsorption through their polar heads to the metal centers on the external surface of the MOFs. In this fashion, the pendant hydrophobic group on top of the surface of the MOF acts as a barrier against water and minimizes water vapor penetration/adsorption while maintaining the material’s CO2 capacity. The rationale behind this concept is the fact that the open metal centers and the ligand chemical functionalities inside the MOF framework are still available to interact with CO2 and that the surface area of the material would not be essentially compromised. To prove the concept, we synthesized different MOFs (MIL-101(Cr), and NiDOBDC using known procedures.27,52 NiDOBDC were chosen because of its decent CO2 capacities (∼28 wt% at 1 bar) however they suffer from moisture instability upon exposure to water for longer periods of time as it will be shown later.1,13,53 MIL-101(Cr) was chosen due to its characteristic high water sorption capacity (∼150 wt%) but poor CO2 sorption capacity at low pressure. Both porous materials were treated with commercially available Pluronic P123, (here onwards P1, average molecular weight 5800 Da)-chloroform solution. This non-ionic surfactant is a triblock copolymer based on poly(ethylene glycol)-poly( propylene glycol)-poly(ethylene glycol) mainly composed of hydroxyl groups at both ends of a hydrophobic ether-based long chain group (Fig. 1). Fig. 2A illustrates the room temperature water adsorption isotherms of untreated MIL101(Cr) as compared to the P1-treated material. The adsorption isotherm shows a significant decrease in water uptake on the P1-modified MIL101(Cr) with respect to the unmodified MIL101(Cr) at relative humidity (RH) above 50%. At 95% RH, P1-MIL101(Cr) showed nearly three times lower water uptake with respect to unmodified MIL101(Cr). P1-modified MIL101(Cr) was cycled three times for water sorption analysis without apparent change in the water uptake at room temperature, clearly demonstrating the stability of the
Fig. 1 Pluronic P123, X = 20, Y = 70, Z = 20 with average Mw = 5800 Da.
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Dalton Transactions
Fig. 2 RT water sorption (A) and low pressure RT CO2 sorption (B) in unmodified and P1-modified MIL101(Cr).
P1-modified MIL101(Cr) (Fig. 2). The treated materials were further activated by methanol exchange followed by heating under vacuum at 80 °C for 10 hours and tested for CO2 sorption. CO2 isotherms of P1-modified MIL101 (Cr) show no obvious alteration in CO2 uptake as compared to untreated MIL101(Cr) at both low and high CO2 pressure (Fig. 2B and S1†). Similarly, water adsorption on P1-modified NiDOBDC decreased by about 60% compared to the unmodified NiDOBDC (Fig. 3A) while maintaining the CO2 capacity (26% w/w at 1 bar) as shown in Fig. 3B. These results are of paramount importance and demonstrate for the first time that a MOF material can be readily functionalized on the surface with a surfactant that not only increases the hydrophobic behavior significantly but do not compromise its CO2 sorption capacity. To gain more insights on the hydrophobic/hydrophilic characteristics of P1-modified MOFs, we performed contact angle measurements on pelletized samples.38,47 Contact angle provides quantitative information on the wettability of a solid surface and it is measured at the liquid/solid surface interface. Materials with contact angles below 90° are considered hydrophilic materials, contact angles between 90–150° are considered hydrophobic materials and materials with contact angle values above 150° are considered as superhydrophobic materials. In the case of powder samples, determining contact angles is not a trivial task and sometimes leads to smaller contact angles due to the roughness and poro-
This journal is © The Royal Society of Chemistry 2015
Paper
Fig. 3 RT water adsorption isotherms on unmodified as well as P1modified NiDOBDC. Unmodified MOF captures twice the mass of water as compared to P1-NiDOBDC at RH above 20% (A). RT CO2 adsorption isotherms on unmodified as well as P1-modified NiDOBDC showing similar CO2 capacities in the entire pressure range (B).
sity of the pellets produced.54–57 Nevertheless, to provide a approximate estimation of the relative hydrophobicity pellets of unmodified MOFs as well as P1-modified MOFs were obtained by applying hydraulic pressure (up to 7 Ton cm−2 for 120 seconds). As prepared MOF pellets were used to measure contact angle using a Rame-Hart goniometer (details in ESI†). Table S1† and Fig. 4 show the results. Unmodified MIL101 and NiDOBDC exhibited significant wettability and near-zero water droplet life time absorbing water instantaneously making contact angle measurements very difficult to perform clearly illustrate their hydrophilic behavior. However, P1-modified MIL101(Cr) showed contact angles around 25–27° with a
Fig. 4 Photographs of P1-modified MIL101(Cr) (left) and unmodified MIL101(Cr) pellets during contact angle measurements.
Dalton Trans.
View Article Online
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Paper
droplet life time of about 30 seconds. P1-modified NiDOBDC pellets had a rough surface with contact angles 22–24° with a droplet life time about 40 seconds. Although the values of contact angle are smaller than expected, the relative lifetime of the water droplet in P1-modified MOFs as compared to unmodified MOFs clearly indicates the more hydrophobic nature of the P1-modified MOFs. These results confirm that an amphiphilic surfactant functionalizing the surface of MOF materials provides a hydrophobic barrier that delays the diffusion of water into the MOF. To learn about the MOFs stability toward water, P1-NiDOBDC was subjected to a number of water adsorption/ desorption cycles (with conditions identical to the ones shown on Fig. 3) followed by CO2 sorption analysis in between cycles. Details about the porous material activation followed by sorption analysis on each cycle are described in the Methods section. Fig. 5A illustrates the decreases in CO2 sorption capacity in unmodified-NiDOBDC with the number of water exposure cycles, particularly after the third cycle. After four water adsorption/desorption cycles, the CO2 uptake at 1 bar for NiDOBDC decreases nearly three times as compared to the original MOF sorption capacity. On the other hand, P1-modified NiDOBDC showed constant sorption values at all pressures independent of the number of water exposure cycles (Fig. 5B). This is a very interesting result because surface coating of MOFs with P1 not only results in increasing the materials hydrophobic behavior but also increases their stability while maintaining CO2 sorption capacity.
Fig. 5 CO2 adsorption isotherms on unmodified (A) and P1-functionalized NiDOBDC (B). Notice the decrease in CO2 capacity on unmodified NiDOBDC, particularly after four water exposure cycle.
Dalton Trans.
Dalton Transactions
Fig. 6 Powder XRD spectra of unmodified and P1-functionalized before and after water exposure showing the stability of the P1NiDOBDC after four CO2/water sorption/desorption cycles as compared to the unstable unmodified NiDOBDC.
Powder X-ray diffraction (PXRD) studies on NiDOBDC and P1-modified NiDOBDC before and after four water adsorption/ desorption cycles are shown in Fig. 6. It can be observed that for the original porous material there is no obvious difference in the crystalline structure after functionalization (Fig. 6A and 6C) More importantly, the PXRD pattern of P1-modified NiDOBDC shows no obvious change after four CO2/wateradsorption/desorption cycles (Fig. 6D). On the other hand, unmodified NiDOBDC does show indication of structural decomposition as evidenced by the disappearance of most of the peaks, particularly at large diffraction angles (Fig. 6B). These results are in complete agreement with the observed decrease in CO2 sorption capacity on unmodified NiDOBDC after the fourth water vapor exposure. Functionalization of the MOF material with the non-ionic surfactant proves not only to significantly reduce the MOF water sorption ability but, as importantly, to render stability against water-induced hydrolysis/decomposition. These two desirable properties can be achieved without compromising the CO2 sorption capacity of the materials in the case of NiDOBDC and MIL-101(Cr). Thermogravimetric (TG) analysis was performed on NiDOBDC and P1-modified NiDOBDC after methanol exchange and is shown in Fig. S2.† It is observed that unmodified NiDOBDC has a mass loss of 18% at low temperatures and then is stable to temperatures as high as 300 °C. P1NiDOBDC showed two mass losses, a 3% mass loss at low temperatures similarly to unmodified NiDOBDC followed by a gradual mass loss at higher temperatures (>230 °C) which can be assigned to Pluronic P123 desorption/decomposition from the MOF surface. Lower solvent loss was observed on P1NiDOBDC which could be attributed to the hydrophobic features of the functionalized material that minimize the incorporation of methanol molecules inside the framework cavities. From TG analysis it was estimated a 14% loading of Pluronic
This journal is © The Royal Society of Chemistry 2015
View Article Online
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Dalton Transactions
P123 on the surface of NiDOBDC. Scanning electron microscopy (SEM) was performed on unmodified and P1-functionalized MIL-101(Cr), and NiDOBDC (Fig. S3†). There were no obvious differences observed in MOF surface morphology between the unmodified and P1-functionalized materials. However, electron charging on the surface of the samples modified with Pluronic P123 was observed in both MOF materials as expected due to the non-conducting nature of the adsorbed surfactants on the surface of the porous materials. This provides additional evidence that Pluronic P123 was indeed adsorbed on the surface of each MOF. Surface area analysis using nitrogen sorption isotherms collected at 77 K was conducted on NiDOBDC and P1-NiDOBDC to learn about possible pore obstruction by the surfactant on the P1-modified MOF (Fig. S4†). The results show that the permanent porosity is retained after functionalization and the surface area of functionalized P1-NiDOBDC (830 m2 g−1) is slightly lower than unmodified NiDOBDC (912 m2 g−1), but within the experimental error. Although slightly lower than the unmodified NiDOBDC, the surface area of modified P1DOBDC is significantly higher than many MOFs studied for CO2 separation.58,59 This confirms the hypothesis that deep penetration of surfactants inside the cavities is fairly unlikely compromising the overall pore capacity and framework selectivity during and after functionalization of the MOF. This is due to the fact that the length of Pluronic P123 (∼28 nm) is significantly larger than the pore size of NiDOBDC and MIL101(Cr) (nearly 20 X larger than the largest cavity size corresponding to MIL101(Cr), 1.6 nm). To further elucidate the effect of P123 on NiDOBDC and understand the possible mechanism involved during water sorption in these materials, the water uptake was plotted as a function of step changes in RH for both P1-NiDOBDC and unmodified NiDOBDC (Fig. 7). In addition, the rate of water
Fig. 7 Water isotherms as a function of time on unmodified NiDOBDC and on P1-modified NiDOBDC showing the times at which water sorption at a new %RH takes place. Note the nearly half of the water uptake of NiDOBDC occurs with 20 ≤ RH ≤ 30%.
This journal is © The Royal Society of Chemistry 2015
Paper
uptake (δ wt%/δt ) was plotted as a function of time (Fig. S5†). Water adsorption in P1-modified and unmodified NiDOBDC are essentially equivalent up to RH = 20%. However at higher RH, particularly between RH = 20–40%, the rates of water adsorption are up to five times higher in unmodified NiDOBDC as compared to P1-modified NiDOBDC at a given time. In addition, the water sorption rates in P1-modified NiDOBDC decay much faster (Fig. S5†) than the rates of unmodified NiDOBDC resulting in shorter equilibrium times (Fig. 7). The combination of lower rates of water adsorption and shorter equilibrium times translates to a significantly lower water uptake by P1-modified NiDOBDC compared to NiDOBDC (nearly half, 45 wt% vs. 23 wt%, Fig. 3 and 7). With these results in hand, we hypothesized that pore obstruction by the surfactant during water sorption might be responsible for the rapid sorption equilibrium and the resulting reduced water capacity of P1-modified NiDOBDC. This equilibrium effect which implies a fundamental and permanent change in the water sorption capacity is of great significance because the modified-MOF materials could be potentially used for thousands of cycles without structure impairment or pore saturation. Then the question is, if it is an equilibrium effect, what is the chemical mechanism? Since the surfactant should be constrained to the outer surface of the MOF material, how could such a surface effect impart a bulk structural change in the adsorption affinity? Molecular dynamics (MD) simulations60 were then performed using Amber 9 Package using the Force field parameters for MOF were taken from previous work to (1) understand the interaction between the NiDOBDC fragments and P-123 in chloroform during MOF functionalization and (2) to learn about the interaction of the P1-functionalized NiDOBDC with water and CO2.60,61 The molecular formula of polymer (P-123) unit used in these MD simulations is (HO(CH2CH2O)2(CH2CH(CH3)O)2(CH2CH2O)2H).62–64 To model the MOF functionalization process with P-123, a portion of a rigid fragment of MOF was taken from the NiMOF-74 crystal structure. The MOF fragment is then centered in a box of size 80 × 80 × 80 Å. To this fragment 20 units of P-123 were added and then solvated with chloroform (ESI†). Snapshots of simulation box containing NiDOBDC and P-123 solvated in chloroform are shown in Fig. S6.† Fig. S7† illustrates the radial distribution functions (RDF) computed between the Ni-atoms of MOF and the oxygen and carbon atoms of P-123. The NiMOF-OP-123 RDF has sharp and intense peak at ∼2.2 Å. As expected we observed strong interaction between the positive Ni-atoms of the MOF and the partially negative O-atoms of P-123. In Fig. S7† we show the RDF between the oxygen atoms of MOF linker units and the terminal hydrogen atoms of P-123. This RDF has a peak at ∼1.7 Å, indicating strong OMOF–HOP-123 interactions. These two kinds of interactions are primary governing factors for binding of P-123 to MOF material. Two systems were prepared by randomly adding three hundred molecules of water or carbon dioxide to MOF fragment equilibrated with the polymer molecules wrapped around it (Fig. S8†). Fig. 8A shows the front view of the initial
Dalton Trans.
View Article Online
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Paper
Dalton Transactions
It is interesting to note that although P1-modified NiDOBDC can still adsorb about 23 wt% of water at RH = 95%, upon regeneration we did not observe any loss in CO2 capacity after 4 cycles. However, we observed significant loss in CO2 capacity of unmodified NiDOBDC ( particularly after the third regeneration cycle) under identical conditions (Fig. 5). This is also evident by the crystalline structure loss of the unmodified NiDOBDC after the third water exposure cycle (Fig. 6). Experimental as well as MD results clearly suggest that the functionalization of the MOF with P123 not only reduces the water capacity of NiDOBDC significantly but it also brings stability upon water exposure without compromising CO2 capacity of the porous material.
Conclusions
Fig. 8 MD snapshot (A) initial frame with water, (B) final frame with water after 150 ps, (C) initial frame with CO2 and (D) final frame with CO2 after 2 ns.
snapshot of the MD simulation and Fig. 8B shows the front view of the MD snapshots after 150 ps of water exposure. Within 150 ps it is observed that the water gets adsorbed on the MOF surface. Also, it is interesting to note that when the water goes into the pore of the MOF fragment they also carry polymer units along with them. Polymer units that try to diffuse into the pores are colored in magenta and the units that remain at the outer surface are colored in white in Fig. 8. The polymer molecules then partially occupy the MOF cavities and restrict the number of water molecules diffusing and compromising the overall water uptake. However, the polymer molecules do not significantly compromise the overall pore capacity and volume. This seems to be the reason for the rapid water sorption equilibrium and the resulting decrease in water loading observed in the water isotherms of P1-NiDOBDC (Fig. 3 and 6). Therefore, the MD simulation studies presented in Fig. 8 are in agreement with the surface area measurements and CO2 sorption studies illustrated in Fig. 5. Now, it is expected that this competing mechanism sorbate/polymer observed for water would not occur during CO2 sorption since the CO2 isotherms obtained for P1-modified NiDOBDC and unmodified NiDOBDC are identical. Fig. 8C and 8D show similar snapshots for the case of exposure of the P1-functionalized NiDOBDC to CO2. It is observed that no P-123 units diffuse into the MOF pores during CO2 diffusion. To ensure a long enough exposure to CO2 MD simulations were ran for longer times. Fig. 8D shows a snapshot after 2 ns exposure of the functionalized MOF material to CO2 showing no polymer diffusing inside the pores during CO2 adsorption, i.e., no polymer/sorbate competing effect was identified when CO2 diffused inside the MOF pores.
Dalton Trans.
The present work provides experimental and modeling evidence that incorporating hydrophobic molecules on the external surface of MOFs is a very important approach for stabilization of these materials against water while maintaining the CO2 sorption properties. It was demonstrated that the surfactant adsorbed on the surface of a porous MOF material is carried into the pores when exposed to water thus compromising the volume available for water. This results in significant reduction of water sorption while bringing stability against water. This concept can be applied to increase stability of MOFs in gas separation, catalysis and sensing applications where water is present. For example, a frequent issue in catalysis is water poisoning.26–28 Indeed; catalytic reactions can be delayed or simply inhibited due to poisoning effects originating from moisture in the air. The incorporation of a hydrophobic shell would minimize the penetration of water to the catalytic sites, hence, rendering water stability and longer catalyst lifetimes.
Experimental Materials synthesis NiDOBDC and MIL-101(Cr) were synthesized following literature procedures.32,58,59 Surfactant functionalization was performed by introducing the evacuated MOF materials in P1 solutions for a couple of hours and then washed several times to remove excess of P1. The materials were evacuated and water and CO2 sorption isotherms were collected at RT. Activation of the modified and unmodified materials were performed at 80 °C after replacing solvents from the synthesis with low boiling point solvents, methanol or chloroform, to minimize potential surfactant desorption. Synthesis of P123 modified MOF materials was performed by simply immersing for 24 h at room temperature evacuated NiDOBDC or MIL-101(Cr) in a 10 wt% chloroform solution of P123. Then, the materials are rinsed with fresh chloroform followed by centrifugation (three times) and finally dried with a gentle flow of N2 or air.
This journal is © The Royal Society of Chemistry 2015
View Article Online
Dalton Transactions
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Characterization methods SEM images were obtained using FEI Helios 600 NanoLab FIB-SEM. CO2 and Nitrogen adsorption–desorption isotherms were collected using a Quantachrome autosorb-6 automated gas sorption. Brunauer–Emmet–Teller (BET) surface area was calculated from the nitrogen isotherm curves ranging from 0.1 to 0.3 of relative pressure. Water sorption isotherms were collected at RT using VTI-SA+ Vapor-sorption analyzer. Thermal stability of CMOFs was investigated using a TG (NETZSCH TG 209 F1) under nitrogen (N2). Powder X-ray diffraction (XRD) analysis was obtained from powder samples placed on 0.8 mm capillaries. A custom-built XYZ environmental stage inside a Bruker-AXS D8 Discover XRD unit equipped with a rotating Cu anode (1.54 Å), göbel mirror, 0.5 mm collimator, and 0.5 mm pin hole (Madison, WI). EMOF films were aligned using a video laser alignment system. Patterns were collected with a GADDS® area detector positioned at 21.5° 2θ with a measured distance from the sample of 15 cm. Collection of individual XRD tracings required 60 s with power settings of 45 kV and 200 mA. Images were processed with Bruker-AXS GADDS software before importing into JADE® XRD software to obtain peak positions and intensities. Contact Angle Goniometry. Contact angle measurements of water with the MOF pellet surfaces were made using a Rame-Hart goniometer. Reported contact angles are static.
Acknowledgements This work was supported by PNNL Energy Conversion Initiative and Office of Fossil Energy, U.S. Department of Energy (DOE). The U.S. Department of Energy (DOE), Office of Basic Energy Sciences (BES), Division of Chemical Sciences, Geosciences and Biosciences funded the work performed by L. X. D. PNNL is a multiprogramming laboratory operated by Battelle Memorial Institute for the Department of Energy under Contract DE-AC05-76RL01830.
Notes and references 1 J. Canivet, A. Fateeva, Y. M. Guo, B. Coasne and D. Farrusseng, Chem. Soc. Rev., 2014, 43, 5594–5617. 2 W. G. Lu, Z. W. Wei, Z. Y. Gu, T. F. Liu, J. Park, J. Park, J. Tian, M. W. Zhang, Q. Zhang, T. Gentle, M. Bosch and H. C. Zhou, Chem. Soc. Rev., 2014, 43, 5561–5593. 3 J. R. Li, Y. G. Ma, M. C. McCarthy, J. Sculley, J. M. Yu, H. K. Jeong, P. B. Balbuena and H. C. Zhou, Coord. Chem. Rev., 2011, 255, 1791–1823. 4 M. P. Suh, H. J. Park, T. K. Prasad and D. W. Lim, Chem. Rev., 2012, 112, 782–835. 5 E. Barea, C. Montoro and J. A. R. Navarro, Chem. Soc. Rev., 2014, 43, 5419–5430. 6 L. Heinke, Z. G. Gu and C. Woll, Nat. Commun., 2014, 5. 7 H. Furukawa, K. E. Cordova, M. O’Keeffe and O. M. Yaghi, Science, 2013, 341, 974–986.
This journal is © The Royal Society of Chemistry 2015
Paper
8 L. J. Murray, M. Dinca and J. R. Long, Chem. Soc. Rev., 2009, 38, 1294–1314. 9 H. Li, M. Eddaoudi, M. O’Keeffe and O. M. Yaghi, Nature, 1999, 402, 276–279. 10 G. Q. Li, H. Kobayashi, J. M. Taylor, R. Ikeda, Y. Kubota, K. Kato, M. Takata, T. Yamamoto, S. Toh, S. Matsumura and H. Kitagawa, Nat. Mater., 2014, 13, 802–806. 11 B. L. Chen, S. C. Xiang and G. D. Qian, Acc. Chem. Res., 2010, 43, 1115–1124. 12 S. C. Xiang, Y. B. He, Z. J. Zhang, H. Wu, W. Zhou, R. Krishna and B. L. Chen, Nat. Commun., 2012, 3, 1–9. 13 L. Q. Ma, C. Abney and W. B. Lin, Chem. Soc. Rev., 2009, 38, 1248–1256. 14 P. K. Thallapally, C. A. Fernandez, R. K. Motkuri, S. K. Nune, J. Liu and C. H. F. Peden, Dalton Trans., 2010, 39, 1692–1694. 15 C. A. Fernandez, P. K. Thallapally, R. K. Motkuri, S. K. Nune, J. C. Sumrak, J. Tian and J. Liu, Cryst. Growth Des., 2010, 10, 1037–1039. 16 J. Della Rocca and W. B. Lin, Eur. J. Inorg. Chem., 2010, 3725–3734. 17 W. J. Rieter, K. M. L. Taylor and W. B. Lin, J. Am. Chem. Soc., 2007, 129, 9852–9853. 18 G. K. H. Shimizu, J. M. Taylor and S. Kim, Science, 2013, 341, 354–355. 19 C. Y. Sun, C. Qin, X. L. Wang and Z. M. Su, Expert Opin. Drug Delivery, 2013, 10, 89–101. 20 L. Bellarosa, S. Calero and N. Lopez, Phys. Chem. Chem. Phys., 2012, 14, 7240–7245. 21 J. J. Low, A. I. Benin, P. Jakubczak, J. F. Abrahamian, S. A. Faheem and R. R. Willis, J. Am. Chem. Soc., 2009, 131, 15834–15842. 22 A. Comotti, S. Bracco, P. Sozzani, S. Horike, R. Matsuda, J. Chen, M. Takata, Y. Kubota and S. Kitagawa, J. Am. Chem. Soc., 2008, 130, 13664–13672. 23 J. B. DeCoste, G. W. Peterson, B. J. Schindler, K. L. Killops, M. A. Browe and J. J. Mahle, J. Mater. Chem. A, 2013, 1, 11922–11932. 24 J. Ehrenmann, S. K. Henninger and C. Janiak, Eur. J. Inorg. Chem., 2011, 471–474. 25 J. B. DeCoste, G. W. Peterson, H. Jasuja, T. G. Glover, Y. G. Huang and K. S. Walton, J. Mater. Chem. A, 2013, 1, 5642–5650. 26 M. Kim, J. F. Cahill, H. H. Fei, K. A. Prather and S. M. Cohen, J. Am. Chem. Soc., 2012, 134, 18082– 18088. 27 L. Bromberg, Y. Diao, H. M. Wu, S. A. Speakman and T. A. Hatton, Chem. Mater., 2012, 24, 1664–1675. 28 X. L. Liu, Y. S. Li, Y. J. Ban, Y. Peng, H. Jin, H. Bux, L. Y. Xu, J. Caro and W. S. Yang, Chem. Commun., 2013, 49, 9140– 9142. 29 K. S. Park, Z. Ni, A. P. Cote, J. Y. Choi, R. D. Huang, F. J. Uribe-Romo, H. K. Chae, M. O’Keeffe and O. M. Yaghi, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 10186–10191. 30 K. Kida, M. Okita, K. Fujita, S. Tanaka and Y. Miyake, CrystEngComm, 2013, 15, 1794–1801.
Dalton Trans.
View Article Online
Published on 01 May 2015. Downloaded by Freie Universitaet Berlin on 13/05/2015 18:05:25.
Paper
31 J. A. Thompson, N. A. Brunelli, R. P. Lively, J. R. Johnson, C. W. Jones and S. Nair, J. Phys. Chem. C, 2013, 117, 8198– 8207. 32 S. Horike, D. Tanaka, K. Nakagawa and S. Kitagawa, Chem. Commun., 2007, 3395–3397. 33 N. M. Padial, E. Q. Procopio, C. Montoro, E. Lopez, J. E. Oltra, V. Colombo, A. Maspero, N. Masciocchi, S. Galli, I. Senkovska, S. Kaskel, E. Barea and J. A. R. Navarro, Angew. Chem., Int. Ed., 2013, 52, 8290–8294. 34 C. Yang, U. Kaipa, Q. Z. Mather, X. P. Wang, V. Nesterov, A. F. Venero and M. A. Omary, J. Am. Chem. Soc., 2011, 133, 18094–18097. 35 D. Y. Ma, Y. W. Li and Z. Li, Chem. Commun., 2011, 47, 7377–7379. 36 H. Jasuja, N. C. Burtch, Y. G. Huang, Y. Cai and K. S. Walton, Langmuir, 2013, 29, 633–642. 37 L. Bellarosa, J. J. Gutierrez-Sevillano, S. Calero and N. Lopez, Phys. Chem. Chem. Phys., 2013, 15, 17696–17704. 38 T. H. Chen, I. Popov, O. Zenasni, O. Daugulis and O. S. Miljanic, Chem. Commun., 2013, 49, 6846–6848. 39 R. Vaidhyanathan, S. Natarajan and C. N. R. Rao, Dalton Trans., 2003, 1459–1464. 40 A. D. Burrows, C. G. Frost, M. F. Mahon and C. Richardson, Angew. Chem., Int. Ed., 2008, 47, 8482–8486. 41 P. Roy, A. Schaate, P. Behrens and A. Godt, Chem. – Eur. J., 2012, 18, 6979–6985. 42 M. Servalli, M. Ranocchiari and J. A. Van Bokhoven, Chem. Commun., 2012, 48, 1904–1906. 43 W. T. Zhu, C. He, X. Wu and C. Y. Duan, Inorg. Chem. Commun., 2014, 39, 83–85. 44 J. D. Evans, C. J. Sumby and C. J. Doonan, Chem. Soc. Rev., 2014, 43, 5933–5951. 45 A. M. Shultz, A. A. Sarjeant, O. K. Farha, J. T. Hupp and S. T. Nguyen, J. Am. Chem. Soc., 2011, 133, 13252–13255. 46 S. Aguado, J. Canivet, Y. Schuurman and D. Farrusseng, J. Catal., 2011, 284, 207–214. 47 J. G. Nguyen and S. M. Cohen, J. Am. Chem. Soc., 2010, 132, 4560–4561.
Dalton Trans.
Dalton Transactions
48 T. J. Wu, L. J. Shen, M. Luebbers, C. H. Hu, Q. M. Chen, Z. Ni and R. I. Masel, Chem. Commun., 2010, 46, 6120– 6122. 49 Y. Cao, F. J. Song, Y. X. Zhao and Q. Zhong, J. Environ. Sci., 2013, 25, 2081–2087. 50 T. M. McDonald, W. R. Lee, J. A. Mason, B. M. Wiers, C. S. Hong and J. R. Long, J. Am. Chem. Soc., 2012, 134, 7056–7065. 51 K. K. Tanabe and S. M. Cohen, Chem. Soc. Rev., 2011, 40, 498–519. 52 S. R. Caskey, A. G. Wong-Foy and A. J. Matzger, J. Am. Chem. Soc., 2008, 130, 10870–10871. 53 A. C. Kizzie, A. G. Wong-Foy and A. J. Matzger, Langmuir, 2011, 27, 6368–6373. 54 E. Chibowski and R. Perea-Carpio, Adv. Colloid Interface Sci., 2002, 98, 245–264. 55 R. J. Hunter, Foundations of Colloid Science, Oxford University Press, 2nd edn, 2001. 56 P. Chassin, C. Jouany and H. Quiquampoix, Clay Miner., 1986, 21, 899–907. 57 E. Chibowski, L. Holysz, A. Zdziennicka and F. Gonzalezcaballero, Surfactants in Solution, Marcel Dekker, New York, 1996. 58 Y. Liu, Z. U. Wang and H.-C. Zhou, Greenhouse Gases: Sci. Technol., 2012, 2, 239–259. 59 J. Liu, P. K. Thallapally, B. P. McGrail, D. R. Brown and J. Liu, Chem. Soc. Rev., 2012, 41, 2308–2322. 60 X. Q. Sun, C. D. Wick, P. K. Thallapally, B. P. McGrail and L. X. Dang, J. Phys. Chem. B, 2011, 115, 2842– 2849. 61 H. J. C. Berendsen, J. R. Grigera and T. P. Straatsma, J. Phys. Chem., 1987, 91, 6269–6271. 62 J. M. Wang, R. M. Wolf, J. W. Caldwell, P. A. Kollman and D. A. Case, J. Comput. Chem., 2004, 25, 1157–1174. 63 J. M. Wang, R. M. Wolf, J. W. Caldwell, P. A. Kollman and D. A. Case, J. Comput. Chem., 2005, 26, 114–114. 64 J. M. Wang, W. Wang, P. A. Kollman and D. A. Case, J. Mol. Graphics Modell., 2006, 25, 247–260.
This journal is © The Royal Society of Chemistry 2015
