Modulating Adsorption and Stability Properties in Pillared Metal

Nov 3, 2015 - Biography. Nicholas C. Burtch was born in Canada but also spent time living in the United States and Dubai before receiving his B.S.E. i...
0 downloads 3 Views 3MB Size
Download PDF
Article pubs.acs.org/accounts

Modulating Adsorption and Stability Properties in Pillared Metal− Organic Frameworks: A Model System for Understanding Ligand Effects Nicholas C. Burtch and Krista S. Walton* School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, 311 Ferst Drive NW, Atlanta, Georgia 30332, United States CONSPECTUS: Metal−organic frameworks (MOFs) are nanoporous materials with highly tunable properties that make them ideal for a wide array of adsorption applications. Through careful choice of metal and ligand precursors, one can target the specific functionality and pore characteristics desired for the application of interest. However, among the wide array of MOFs reported in the literature, there are varying trends in the effects that ligand identity has on the adsorption, chemical stability, and intrinsic framework dynamics of the material. This is largely due to ligand effects being strongly coupled with structural properties arising from the differing topologies among frameworks. Given the important role such properties play in dictating adsorbent performance, understanding these effects will be critical for the design of next generation functional materials. Pillared MOFs are ideal platforms for understanding how ligand properties can affect the adsorption, stability, and framework dynamics in MOFs. In this Account, we highlight our recent work demonstrating how experiment and simulation can be used to understand the important role ligand identity plays in governing the properties of isostructural MOFs containing interconnected layers pillared by bridging ligands. Changing the identity of the linear, ditopic ligand in either the 2-D layer or the pillaring third dimension allows targeted modulation of the chemical functionality, porosity, and interpenetration of the framework. We will discuss how these characteristics can have important consequences on the adsorption, chemical stability, and dynamic properties of pillared MOFs. The structures discussed in this Account comprise the greatest diversity of isostructural MOFs whose stability properties have been studied, allowing valuable insight into how ligand properties dictate the chemical stability of isostructural frameworks. We also discuss how functional groups can affect adsorbate energetics at their most favorable adsorption sites to elucidate how functional groups can affect the adsorptive performance of these materials in ways that are unexpected based on the isolated ligand’s properties. We then highlight a variety of simulation tools that not only can be used to understand the differing molecular-level behavior of the adsorbate and framework dynamics within these isostructural MOFs, but also can shed light on possible mechanisms that govern the differing chemical stability properties among these materials. Lastly, we provide perspective on the challenges and opportunities for utilizing the structure−property relationships arising from the ligand effects described in this Account for the design of further MOFs with enhanced chemical stability and adsorption properties.

1. INTRODUCTION Porous metal−organic frameworks, formed by the assembly of metal-containing units with organic linkers, exhibit enormous structural diversity due to the wide array of species available as chemical building blocks. Given this extreme tunability, numerous synthetic strategies have been employed for their targeted design based on the desired pore characteristics.1,2 Beyond such strategies, further pore control can also be realized through postsynthetic modification of the linkers to incorporate additional organic and metal−organic complexes into the structure.3 This high degree of control over structural and chemical properties makes MOFs promising candidates for a wide array of adsorption applications, ranging from gas separation and storage to chemical sensing and catalysis.4,5 However, among © 2015 American Chemical Society

the large number of MOFs reported in literature, the observed trends for how ligand identity affects the adsorption, chemical stability, and framework dynamic properties of different structures can widely vary. One large contributor to these discrepancies is the important role structural topology can also play in determining framework properties. For example, while one can attempt to predict framework adsorption properties from attributes inherent to the ligand and its functional groups in their isolated state, once incorporated within the MOF structure, the functional group’s environment can be significantly altered. This can make broad generalizations regarding the impact of specific functional groups on MOF Received: June 22, 2015 Published: November 3, 2015 2850

DOI: 10.1021/acs.accounts.5b00311 Acc. Chem. Res. 2015, 48, 2850−2857

Article

Accounts of Chemical Research

Figure 1. Schematic of the pillared topology (top left) and summary of the different pillaring (bottom left) and 2-D layer (right) ligands used in the synthesis of the structures discussed in this Account. ADB = 9,10-anthracenedibenzoic acid; ADC = 9,10-anthracenedicarboxylic acid; BDC = 1,4benzenedicarboxylic acid; BPY = 4,4′-bipyridine; DABCO = 4-diazabicyclo[2.2.2]octane; NDC = 1,4-napthalenedicarboxylic acid.

and air pollution control where stability in the presence high humidity and acidic or basic conditions may also be needed.6,7 Relative to inorganic porous solids such as activated carbons and zeolites, MOFs tend to have lower chemical stability due to their metal−ligand coordination bonds. However, in recent years, a growing number of MOFs with high-valence metal ions such as Fe3+, Zr4+, and Hf4+ that exhibit stability even in the presence of acidic and basic conditions have emerged.8 Such metals allow high nuclearity among metal clusters along with strong electrostatic interactions in the metal−ligand bonds in order to form more hydrolytically stable coordination complexes. The focus of our recent work has been to develop a more diverse array of design strategies that can be used to improve MOF chemical stability without altering the identity of the metal in the metal cluster. To achieve this, our approach was to explore how functional groups and steric factors within the pore space can be used to tune stability properties. Such understanding would open further possibilities for simultaneously adjusting the stability and adsorption characteristics of MOFs that are of interest for target applications via careful choice of appropriate pore functionality. Table 1 summarizes the composition and relative stability of the structures explored in our past studies. The two strategies we exploited to increase the stability of the isostructural frameworks in Table 1 were to (i) incorporate bulky, nonpolar functional groups near the corner of each metal site in the pore space and (ii) introduce catenation, or interpenetration, within the frameworks. The structural characteristics of the zinc-based MOFs listed in Table 1 are quite diverse. These frameworks vary primarily in the type (polar versus nonpolar) and number of functional groups

properties problematic. To account for such effects, studies that elucidate the effect of ligand identity on framework properties within a single structural topology are optimal. In this Account, we discuss our studies on the array of (i) chemical stability, (ii) gas adsorption, and (iii) framework and adsorbate dynamic properties present in various isostructural, pillared MOFs. We also demonstrate how framework properties can be modulated by incorporating appropriate functional groups and steric characteristics around the linear, ditopic ligands in either the 2-D layer or the pillaring third dimension. In each structure, the ligands in the 2-D layer are connected to the metal sites via carboxylate groups whereas the pillaring ligands are coordinated via nitrogen bonds (Figure 1). We also address how complementary experimental and computational techniques can be used to understand these effects, as well as aid in the design of further functional materials.

2. CHEMICAL STABILITY To be considered viable candidates for real world applications, materials must go beyond possessing the intrinsic functionality needed to meet the performance requirements of the process. In addition, materials must also exhibit long-term stability under the operating conditions of the process. Depending on the application, the stability requirements will vary. For industrially relevant heterogeneous catalysis applications, high thermal stability may be required, whereas in applications requiring MOF powders to be processed into thin-films or pellets, high mechanical stability may be necessary. At a minimum, most adsorption applications require that the material maintain structural stability in the presence of air. Beyond this, more stringent chemical stability requirements may exist for processes such as postcombustion CO2 capture 2851

DOI: 10.1021/acs.accounts.5b00311 Acc. Chem. Res. 2015, 48, 2850−2857

Article

Accounts of Chemical Research

water or if the stability is merely a consequence of an increased hydrophobicity that prevents water from adsorbing into the pores. Figure 2 shows the water adsorption isotherms as a function of relative humidity at 298 K for all structures in Table 1 that

Table 1. Summary of the Compositions and Water Stability Characteristics for the Pillared Structures Characterized in Our Previous Work structure Zn-DMOF Zn-DMOF-Br Zn-DMOF-OH Zn-DMOF-NO2 Zn-DMOF-NH2 Zn-DMOFMM1/2 Zn-DMOF-MM Zn-DMOF-Cl2 Zn-DMOFDM1/2 Zn-DMOF-DM Zn-DMOF-TF Zn-DMOFTM1/2 Zn-DMOF-NDC Zn-DMOF-ADC Zn-DMOF-TMc Co-DMOF-TM Cu-DMOF-TM Ni-DMOF-TM Zn-MOF-508*d

compositiona Zn2(BDC)2(DABCO) Zn2(BDC-Br)2(DABCO) Zn2(BDC-OH)2(DABCO) Zn2(BDC-NO2)2(DABCO) Zn2(BDC-NH2)2(DABCO) Zn2(BDC-MM)(BDC) (DABCO) Zn2(BDC-MM)2(DABCO) Zn2(BDC-Cl2)2(DABCO) Zn2(BDC-DM)(BDC) (DABCO) Zn2(BDC-DM)2(DABCO) Zn2(BDC-TF)2(DABCO) Zn2(BDC-TM)(BDC) (DABCO) Zn2(NDC)2(DABCO) Zn2(ADC)2(DABCO) Zn2(BDC-TM)2(DABCO) Co2(BDC-TM)2(DABCO) Cu2(BDC-TM)2(DABCO) Ni2(BDC-TM)2(DABCO) Zn2(BDC)2(BPY)

stability at 80% RHb

ref

low low low low low low

9 10 10 10 9 11

low low low

11 10 11

low low partial

11 11 10

partial high high high high high high

10 10 11 12 12 12 13

a

Formula of the guest-free framework; ligand abbreviations are given in Figure 1. bStability under conditions of at least 80% RH in air at 298 K; low corresponds to a near complete loss of BET area after water exposure, partial indicates moderate BET area loss, and high indicates little to no loss of BET area after water exposure. cStructure also retains its surface area after three consecutive cycles of water adsorption and desorption until 90% RH. dThe asterisk indicates that the framework is 2-fold interpenetrated.

Figure 2. Water adsorption isotherms as a function of relative humidity at 298 K in air for all structures in Table 1 that exhibit high stability under humid conditions. (a) Zinc-based MOFs with different ligands. The asterisk indicates that Zn-MOF-508 is a 2-fold interpenetrated structure. Isotherm data taken from refs 10, 11, and 13. (b) DMOF-TM variants with different metal identities. Isotherm data taken from refs 11 and 12.

grafted on the BDC ligand but also include instances where the number of rings on the layered or pillaring ligands are extended in either the axial or lateral direction. Within these structures, there are also frameworks that possess 50:50 mixtures of BDC with different mono-, di-, and tetra-methyl functionalized BDC ligands. We also report a structure exhibiting 2-fold interpenetration, along with variants of the tetramethyl functionalized DMOF structure possessing metals other than Zn, in order to illustrate the transferability of these ligand stability trends across isostructural MOFs with different metal identities. The stability of all frameworks in Table 1 were determined via a comparison of the BET areas derived from N2 adsorption at 77 K and the powder X-ray diffraction (PXRD) patterns obtained before and after sample exposure to 80% RH at 298 K in air. It is necessary that, at a very minimum, a combination of BET and PXRD pattern analysis before and after water exposure be considered in making such stability characterizations in order to avoid incorrect conclusions regarding the sample’s stability after water exposure.8 This is because PXRD only provides spatially averaged, long-range information about the crystallinity of the framework and can thus neglect critical information evident from BET analysis regarding local pore collapse at the crystal’s surface that may render the internal surface area of the structure inaccessible. Beyond this, measuring the adsorption isotherm during water exposure is also a useful practice because it allows one to understand whether the stability differences between materials can be attributed to an increased resiliency for the structure toward

exhibit high stability under humid conditions. For the unstable materials, structural breakdown during water adsorption results in isotherms with saturation uptakes that are well below what is expected based on their available porosity. However, among stable structures, the water loadings more closely correlate with porosity and can also be used to understand the inherent hydrophobicity of the structure. For the highly stable ZnDMOF-TM and Zn-DMOF-ADC structures, the isotherms in Figure 2a show that the structures adsorb significant amounts of water (greater than all of the less stable Zn-DMOF frameworks on a mole per kilogram basis9−11) at their saturation loadings. This indicates that their improvement in stability relative to the less stable structures is not a result of excluding water from the pore space and instead arises from an increased resiliency within the structures toward water. Furthermore, because this high level of stability is only exhibited in structures possessing the BDC-TM and ADC ligands but not those structures with the BDC-DM or NDC ligands, we can also conclude that it is critically important that the nonpolar groups be present near the corners of each metal cluster in the structure for this high level of stability to be present. The other strategy we exploited for increasing structural stability involved utilizing a pillaring ligand that enables framework catenation to occur during the synthesis process. In this case, replacing the DABCO ligand in Zn-DMOF with 2852

DOI: 10.1021/acs.accounts.5b00311 Acc. Chem. Res. 2015, 48, 2850−2857

Article

Accounts of Chemical Research

Figure 3. Experimental and molecular simulation results investigating the effect of ligand functionalization on low pressure CO2 affinity in ZnDMOF variants. Figure adapted from content in ref 21. (a) Experimental CO2 Henry’s coefficients at 298 K. Circled data points were chosen as representative structures for further analysis via molecular simulation. (b) CO2 density distributions at infinite dilution and 298 K obtained from molecular simulations. The x−y (left) and x−z (right) planes of each structure are shown. The black oval on the x−z plane of the structures highlights the concentrated CO2 density around its strongest 0 K binding site. (c) Contribution from van der Waals and Coulombic forces to the stronger 0 K CO2 binding site in each structure. Dotted green and red lines are a reference to contributions in the parent DMOF structure. (d) Two strong binding sites for CO2 found in each of the representative structures. The most favorable binding site (left) has CO2 located near the metal cluster, whereas the less favorable binding site (right) has CO2 positioned between the phenyl rings. Color code for framework atoms are red (oxygen), cyan (carbon), blue (nitrogen), and gray (zinc). BDC functional groups are omitted, and CO2 is highlighted with a yellow glow for clarity.

Ligand functionalization is one appealing approach for finetuning the adsorbent’s gas adsorption properties. In the low pressure regime, polar functional groups are well-known to enhance framework−CO2 adsorption properties.6,16 For instance, among ZIFs, polar −CN and −CHO groups were shown to create more favorable CO2 interaction sites than the nonpolar −CH3 functionality.17 Polar entities such as methoxy, amino, and nitro groups on the BDC ligand produced similar enhancements to the low pressure CO2 affinity of UiO-66.18 Density functional theory calculations on isolated benzene rings also predicted that polar functional groups such as −OH, −NO2, and −NH2 will enhance CO2 interactions due to lone pair and hydrogen bonding interactions that polarize CO2.19 Given the ultimate goal of developing both chemically stable and CO2 selective materials for use in CO2 capture processes where moisture will be present, strategies that simultaneously enhance both the stability and CO2 affinity of materials are of great interest. Expanding upon a number of previously synthesized Zn-DMOF variants,20 we recently showed that the introduction of nonpolar functional groups within these pillared structures is one such strategy to achieve this.21 Analysis of the experimental CO2 Henry’s coefficients obtained from functionalizing the BDC ligand on ten isostructural ZnDMOF variants demonstrated that nonpolar functional groups provide the greatest enhancement in the low pressure CO2 affinity whereas polar groups such as nitro, hydroxyl, chlorine, fluorine, and bromine provide little to no improvement. As a case study for how molecular simulation can be used to aid in the interpretation of such experiments, we will briefly discuss our approach for understanding these results at the molecular level.

the longer BPY ligand results in the formation of an isostructural, 2-fold interpenetrated Zn-MOF-508 structure. Despite DABCO having a higher basicity, which one would expect to increase the strength of the resulting coordination bond in the structure, Zn-DMOF is the less stable variant and exhibits a near complete loss of its surface area after exposure to 80% RH at 298 K. On the other hand, Zn-MOF-508 retains its complete surface area under these same conditions. However, the origin of this structure’s greater water stability is not as clear as in the former cases because of its much lower water uptake during adsorption. As a result, one cannot determine whether its stability improvement is due to catenation bringing an increased water resiliency to the structure or instead is simply a result of its lower water uptake.

3. GAS ADSORPTION PROPERTIES Advanced sorbent technologies are attractive alternatives to industrial large-scale separation processes involving absorption and distillation. 14,15 This is due to the large energy consumption that is inherent to these traditional processes and the immense opportunities for designing tailored sorbents that possess the molecular sieving and selective gas adsorption attributes needed to achieve these same separations at lower operating costs. For applications such as postcombustion CO2 capture, a major challenge is to selectively capture CO2 at partial pressures near 0.1 bar from mixtures that contain a majority of nitrogen along with water, oxygen, and other trace contaminants. A fundamental understanding of how low pressure adsorption properties are affected by MOF structural features is therefore a critical step in designing materials that exhibit the performance needed for such applications. 2853

DOI: 10.1021/acs.accounts.5b00311 Acc. Chem. Res. 2015, 48, 2850−2857

Article

Accounts of Chemical Research

Within the pillared topology, the flexible “breathing” transformation between large and narrow pore structures is exhibited by various structures in the presence of guest molecules. Zn-DMOF is among these structures,24,25 and it has also been shown that its flexible breathing behavior in the presence of guest molecules such as benzene and isopropyl alcohol can be modeled using flexible force field models.26 Functionalization of the BDC ligand can strongly impact this behavior, and postsynthetic modification with linear alkyl anhydrides,27 amino-halo bifunctional groups,28 or dangling alkoxy groups with various functionalities29 can be used to drastically tune the structure’s ability to breathe. Building upon this work, we’ve recently studied how ligand functionalization can affect the adsorbate and ligand dynamic properties within various Zn-DMOF variants using both experiment and simulation. In particular, our recent studies have focused on understanding how functional groups on the BDC ligand affect the (i) dynamic properties of water, (ii) nanorotor dynamics of the DABCO ligand, and (iii) thermal expansion behavior of the evacuated framework. As we will discuss, these first two topics can also be useful in exploring possible mechanisms governing the water stability behavior of the highly stable Zn-DMOF-TM structure relative to the less stable Zn-DMOF variants. For all our molecular simulations, we used the RASPA molecular simulation package30 and applied Monte Carlo techniques31 to probe static properties along with molecular dynamics simulations to explore the dynamic behavior of the different Zn-DMOF variants. One water stability hypothesis we tested that is related to the dynamic properties of water is whether the presence of bulkier functional groups in the more stable structures prevents water from ever approaching the zinc metal sites. To test this hypothesis, we used molecular simulation to explore the behavior of the Zn-DMOF-TM and Zn-DMOF-ADC structures relative to the less stable ZnDMOF variants in the presence of water loadings that correspond to structural breakdown in the less stable frameworks.32 Using radial distribution functions, we quantified the distance between water and the zinc metals in the different structures and, as shown in Figure 4a, found that water is only able to get 0.5 Å closer to the metal sites in the unstable structures. We also explored the effect of framework flexibility on the Zn-DMOF and Zn-DMOF-TM modeling results and found that the flexible framework’s radial distribution functions at short distances are nearly identical to those of the rigid frameworks. In addition, we also measured how ligand identity can affect water’s reorientation behavior via the rotational autocorrelation function (Figure 4b). These results show that, relative to bulk water at the same approximate density as what is present in the frameworks, bulky ligands such as those present in Zn-DMOFTM and Zn-DMOF-A result in a longer molecular orientation memory for the water (except for at very short times). In other words, these bulky ligands create a confinement effect that causes water molecules near their proximity to be stabilized, therefore making it harder for them to change their orientation. Another water stability hypothesis we tested was related to the DABCO dynamics in the Zn-DMOF and Zn-DMOF-TM structures.33 It is known from experiment that DABCO is displaced from the Zn-DMOF framework in the presence of moisture, causing its 3-D structure to collapse into 2-D layers of BDC with the DABCO ligands absent.34 We therefore made it our second hypothesis to test whether the water stability

As shown in Figure 3a, a subset of these experimental structures (Zn-DMOF, Zn-DMOF-DM, Zn-DMOF-Cl2, ZnDMOF-TM, and Zn-DMOF-TF) that vary in terms of the type (polar vs nonpolar) and extent of ligand functionalization were first identified for molecular modeling. Strong agreement was obtained between the experimental and calculated Henry’s coefficients using these molecular models, demonstrating their ability to capture the experimentally observed trends. Because the Henry’s coefficient is related to the weighted, spatially averaged free energy of an isolated CO2 molecule within the structure, one can obtain valuable insight into the observed Henry’s coefficients by considering a combination of the (i) energetics for CO2 at its most favorable 0 K binding sites and (ii) probability that CO2 visits these sites at finite temperatures. By this approach, the interplay of intermolecular interactions governing these trends could be uncovered. In each of the Zn-DMOF variants, the most favorable 0 K binding site was found near the zinc metal cluster with the oxygen atoms in CO2 oriented in close proximity to the adjacent functional groups on the ligand (Figure 3d, left). A breakdown of the energetics at these most favorable sites showed that the main factor distinguishing the adsorption affinities among structures was their Coulombic interactions (Figure 3c). Given the close proximity of the negatively charged oxygen in CO2 to the functional groups in the structure, unfavorable repulsive Coulombic interactions result in those structures possessing polar functional groups whereas these same unfavorable repulsions are absent in structures with more nonpolar groups such as methyl. It should be noted that these unfavorable interactions are a direct result of the functional groups’ positions once incorporated within the pore space and therefore could not be predicted from isolated ligand calculations where CO2 can freely orient itself to achieve its most favorable interactions with the functional group. As mentioned earlier, the Henry’s coefficient is related to the weighted, spatially averaged free energy of an isolated adsorbate in the pore. Therefore, in addition to the binding site energetics, the second contribution to the Henry’s coefficient that must be considered is the probability that CO2 actually visits these most favorable adsorption sites. From this analysis, one can extract more quantitative insight into the observed Henry’s coefficient trends. For example, while the 0 K binding energetics for CO2 are very similar for the Zn-DMOF-DM and Zn-DMOF-TM structures, the calculated Henry’s coefficient is roughly two times greater within the Zn-DMOF-TM structure. By analyzing the CO2 density distributions at 298 K and infinite dilution within the structures (Figure 3b), one can then see that this result arises from the much higher probability for CO2 to visit these more favorable sites in the Zn-DMOF-TM versus the Zn-DMOF-DM structure.

4. FRAMEWORK AND ADSORBATE DYNAMICS The reversible framework dynamics exhibited in the subclass of MOFs termed soft porous crystals create unique opportunities for tailoring these materials for chemical sensing functions not attainable in rigid porous materials.22 Understanding the mechanism of these flexible responses to external stimuli can also be critical to understanding the material’s performance properties. For instance, the imidazolate linker reorientations found in ZIF-8 are responsible for its ability to perform kinetic separations that discriminate between adsorbates with molecular sizes that are above what is predicted from its rigid aperture diameter.23 2854

DOI: 10.1021/acs.accounts.5b00311 Acc. Chem. Res. 2015, 48, 2850−2857

Article

Accounts of Chemical Research

Zn-DMOF-TM still persists even after the structure has adsorbed significant amounts of water. Using in situ IR measurements, we were also able to show that while water interactions do perturb the DABCO environment, these perturbations are not occurring on a time scale that is detectable to 2H NMR. However, the time scale of our molecular simulation results did capture these perturbations and reflect this in the short-time behavior of the autocorrelation function for DABCO rotation around its N−N axis (Figure 5b). Lastly, we also explored the thermal expansion properties of the evacuated Zn-DMOF and Zn-DMOF-TM frameworks.32 This behavior was studied computationally via flexible framework simulations where fluctuations in the unit cell length and shape were allowed to occur. Contrary to the behavior of most materials, negative thermal expansion can be observed in nanoporous materials such as MOFs. This behavior has been shown in carboxylate-based MOFs to be due to an increased wiggling motion of the linkers that cause their total projected length to decrease at higher temperatures. On the other hand, lower temperatures cause the linkers to wiggle less and therefore increase their projected length.35 In agreement with this, we find negative thermal expansion behavior in the direction of the 2-D BDC layers of the Zn-DMOF structure, though not in the direction of the more rigid DABCO linkers (Figure 6). Interestingly, the Zn-DMOF-TM structure does not exhibit this same negative thermal expansion behavior in the direction of its 2-D layers. This is likely due to the bulkiness of the BDC-TM ligand and the resulting steric hindrance this creates, which severely limits the wiggling motions of the linker.

Figure 4. Molecular simulation results in different Zn-DMOF variants. Data taken from ref 32. (a) Radial distribution function of the zinc and oxygen (water) pair at a water loading that corresponds to structural breakdown in the unstable structures. (b) Rotational autocorrelation functions for water. The rotational autocorrelation functions of bulk water at 995 (liquid) and 80 kg/m3 (the approximate density of water within the frameworks) are also given as references.

5. PERSPECTIVE AND OUTLOOK In this Account, we have summarized our recent work toward understanding the important role ligand properties play in determining the adsorption, chemical stability, and dynamic properties of pillared MOFs. While the specific structure− property relationships arising from ligand effects may differ based on the structural topology, the experimental and molecular simulation approaches we present in this Account will nonetheless aid in the study of further materials. In our discussion, we highlighted instances where incorporating steric factors such nonpolar functionality into the pore space can be used to tune framework stability and adsorption properties. It has since been demonstrated that nonpolar

improvement in Zn-DMOF-TM might be due to its bulky functional groups locking the DABCO in place, thus preventing its displacement from the structure. DABCO has a globular shape that allows it to serve as a highly effective nanorotor that performs fast rotations around its longitudinal N−N axis (Figure 5a). Using a sample of the Zn-DMOF-TM structure containing deuterated DABCO, we were able to monitor this behavior in situ via 2H NMR measurements to observe any differences in the DABCO dynamics within Zn-DMOF-TM in the presence of water. These measurements ruled out a steric locking mechanism by showing that DABCO rotates freely inside both the Zn-DMOF and Zn-DMOF-TM structures and that DABCO rotation in the

Figure 5. Understanding DABCO nanorotor dynamics in Zn-DMOF and Zn-DMOF-TM. Figures taken from ref 33. (a) Schematic depicting DABCO rotation around its N−N axis. (b) Autocorrelation functions for DABCO rotation around its N−N axis in the evacuated structures (inset) and Zn-DMOF-TM as a function of water loading. 2855

DOI: 10.1021/acs.accounts.5b00311 Acc. Chem. Res. 2015, 48, 2850−2857

Accounts of Chemical Research

■

Article

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest. Biographies Nicholas C. Burtch was born in Canada but also spent time living in the United States and Dubai before receiving his B.S.E. in chemical engineering from the University of Michigan. He is currently an NSF fellow in Professor Krista Walton’s research laboratory at the Georgia Institute of Technology. His research interests include the experimental synthesis and computational design of nanoporous materials for adsorption applications.

Figure 6. Average unit cell length changes as a function of temperature, showing the thermal expansion behavior in Zn-DMOF and Zn-DMOF-TM observed from flexible NPT ensemble molecular simulations (relative to their minimum energy 0 K structures). The Bdirection corresponds to the crystallographic direction of the pillaring DABCO, whereas (A + C)/2 corresponds to the average length of the crystallographic directions corresponding to the 2-D layer of carboxylate linkers. Data taken from ref 32.

Krista S. Walton was born in Florence, Alabama, and grew up in nearby Elgin. She received a B.S.E. degree from the University of AlabamaHuntsville and Ph.D. from Vanderbilt University, both in chemical engineering. She joined the faculty of the Georgia Institute of Technology in the School of Chemical & Biomolecular Engineering in August 2009 and is now Associate Professor and Marvin R. McClatchey and Ruth McClatchey Cline Faculty Fellow. Her research involves the design and synthesis of next-generation multifunctional porous materials for adsorption applications.

methyl and alkyl groups also produce similar improvements to MOF stability and adsorption characteristics in widely different topologies,36,37 indicating that this strategy can be considered a promising general approach for frameworks outside of the pillared topology as well. We also demonstrated that framework interpenetration could be used as a strategy for improving MOF water stability. However, inherent to both of these approaches is the trade-off in pore volume that will practically limit the degree to which one can introduce such steric effects into the pore space. Despite these advances, there are still open questions regarding these chemical stability mechanisms. For example, there are clear limits in the ability for catenation to improve framework stability properties. In contrast to our stability findings in the 2-fold interpenetrated Zn-MOF-508 structure, there is the previously reported DUT-30(Zn) pillared structure that is also 2-fold interpenetrated but instead pillared by DABCO along with ADB (see Figure 1) as its carboxylatebased ligand. In this structure, framework interpenetration does not impart a significant level of water stability on the structure.38 Furthermore, there appear to be additional mechanisms, beyond those discussed in this Account, through which ligands can also improve MOF stability properties. For example, in a Zn-DMOF variant possessing the polar BDC-(NO2)2 ligand, an increased water stability in the material relative to the parent Zn-DMOF was reported.39 This was attributed to the symmetric NO2 groups within the pore space and their ability to create interaction sites for hydrogen bonding that attract water away from its metal hydrolysis sites. A similar stability mechanism has also been proposed computationally40 and therefore represents another strategy for increasing the chemical stability of materials that merits further investigation. Lastly, it is possible that defect incorporation during synthesis may also affect MOF stability properties. Given the difficulty in detecting and controlling defects within metal−organic frameworks,41 this presents an additional challenge in pinpointing the origin of the stability differences between materials.

■

ACKNOWLEDGMENTS This work was supported as part of the Center for Understanding and Control of Acid Gas-Induced Evolution of Materials for Energy (UNCAGE-ME), an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, under Award No. DESC0012577. N.B. acknowledges a fellowship through the National Science Foundation Graduate Research Fellowship and Graduate Research Opportunities Worldwide award under Grant Number DGE-1148903. We also thank Mariá Teresa Santos Fernández for help with the figure preparation.

■

REFERENCES

(1) Yaghi, O. M.; O’Keeffe, M.; Ockwig, N. W.; Chae, H. K.; Eddaoudi, M.; Kim, J. Reticular Synthesis and the Design of New Materials. Nature 2003, 423, 705−714. (2) Guillerm, V.; Kim, D.; Eubank, J. F.; Luebke, R.; Liu, X.; Adil, K.; Lah, M. S.; Eddaoudi, M. A Supermolecular Building Approach for the Design and Construction of Metal-Organic Frameworks. Chem. Soc. Rev. 2014, 43, 6141−6172. (3) Cohen, S. M. Postsynthetic Methods for the Functionalization of Metal-Organic Frameworks. Chem. Rev. 2012, 112, 970−1000. (4) Furukawa, H.; Cordova, K. E.; O’Keeffe, M.; Yaghi, O. M. The Chemistry and Applications of Metal-Organic Frameworks. Science 2013, 341, 1230444. (5) Li, J.-R.; Sculley, J.; Zhou, H.-C. Metal-Organic Frameworks for Separations. Chem. Rev. 2012, 112, 869−932. (6) Sumida, K.; Rogow, D. L.; Mason, J. A.; McDonald, T. M.; Bloch, E. D.; Herm, Z. R.; Bae, T.-H.; Long, J. R. Carbon Dioxide Capture in Metal-Organic Frameworks. Chem. Rev. 2012, 112, 724−781. (7) DeCoste, J. B.; Peterson, G. W. Metal-Organic Frameworks for Air Purification of Toxic Chemicals. Chem. Rev. 2014, 114, 5695− 5727. (8) Burtch, N. C.; Jasuja, H.; Walton, K. S. Water Stability and Adsorption in Metal-Organic Frameworks. Chem. Rev. 2014, 114, 10575−10612. (9) Schoenecker, P. M.; Carson, C. G.; Jasuja, H.; Flemming, C. J. J.; Walton, K. S. Effect of Water Adsorption on Retention of Structure

2856

DOI: 10.1021/acs.accounts.5b00311 Acc. Chem. Res. 2015, 48, 2850−2857

Article

Accounts of Chemical Research and Surface Area of Metal-Organic Frameworks. Ind. Eng. Chem. Res. 2012, 51, 6513−6519. (10) Jasuja, H.; Huang, Y.-G.; Walton, K. S. Adjusting the Stability of Metal-Organic Frameworks under Humid Conditions by Ligand Functionalization. Langmuir 2012, 28, 16874−16880. (11) Jasuja, H.; Burtch, N. C.; Huang, Y.-G.; Cai, Y.; Walton, K. S. Kinetic Water Stability of an Isostructural Family of Zinc-based Pillared Metal-Organic Frameworks. Langmuir 2013, 29, 633−642. (12) Jasuja, H.; Jiao, Y.; Burtch, N. C.; Huang, Y.-G.; Walton, K. S. Synthesis of Cobalt-, Nickel-, Copper-, and Zinc-based, Water-stable, Pillared Metal−Organic Frameworks. Langmuir 2014, 30, 14300− 14307. (13) Jasuja, H.; Walton, K. S. Effect of Catenation and Basicity of Pillared Ligands on the Water Stability of MOFs. Dalton Trans. 2013, 42, 15421−15426. (14) Pimentel, B. R.; Parulkar, A.; Zhou, E.-k.; Brunelli, N. A.; Lively, R. P. Zeolitic Imidazolate Frameworks: Next-generation Materials for Energy-efficient Gas Separations. ChemSusChem 2014, 7, 3202−3240. (15) Torres-Knoop, A.; Dubbeldam, D. Exploiting Large-pore Metal−organic Frameworks for Separations through Entropic Molecular Mechanisms. ChemPhysChem 2015, 16, 2046−2067. (16) Bae, Y.-S.; Snurr, R. Q. Development and Evaluation of Porous Materials for Carbon Dioxide Separation and Capture. Angew. Chem., Int. Ed. 2011, 50, 11586−11596. (17) Morris, W.; Leung, B.; Furukawa, H.; Yaghi, O. K.; He, N.; Hayashi, H.; Houndonougbo, Y.; Asta, M.; Laird, B. B.; Yaghi, O. M. A Combined Experimental-Computational Investigation of Carbon Dioxide Capture in a Series of Isoreticular Zeolitic Imidazolate Frameworks. J. Am. Chem. Soc. 2010, 132, 11006−11008. (18) Cmarik, G. E.; Kim, M.; Cohen, S. M.; Walton, K. S. Tuning the Adsorption Properties of UiO-66 via Ligand Functionalization. Langmuir 2012, 28, 15606−15613. (19) Torrisi, A.; Mellot-Draznieks, C.; Bell, R. G. Impact of Ligands on CO2 Adsorption in Metal-organic Frameworks: First Principles Study of the Interaction of CO2 with Functionalized Benzenes. II. Effect of Polar and Acidic Substituents. J. Chem. Phys. 2010, 132, 044705. (20) Chun, H.; Dybtsev, D. N.; Kim, H.; Kim, K. Synthesis, X-ray Crystal Structures, and Gas Sorption Properties of Pillared Square Grid Nets Based on Paddle-wheel Motifs: Implications for Hydrogen Storage in Porous Materials. Chem. - Eur. J. 2005, 11, 3521−3529. (21) Burtch, N. C.; Jasuja, H.; Dubbeldam, D.; Walton, K. S. Molecular-level Insight into Unusual Low Pressure CO2 Affinity in Pillared Metal-Organic Frameworks. J. Am. Chem. Soc. 2013, 135, 7172−7180. (22) Bureekaew, S.; Shimomura, S.; Kitagawa, S. Chemistry and Application of Flexible Porous Coordination Polymers. Sci. Technol. Adv. Mater. 2008, 9, 014108. (23) Zhang, K.; Lively, R. P.; Zhang, C.; Chance, R. R.; Koros, W. J.; Sholl, D. S.; Nair, S. Exploring the Framework Hydrophobicity and Flexibility of ZIF-8: From Biofuel Recovery to Hydrocarbon Separations. J. Phys. Chem. Lett. 2013, 4, 3618−3622. (24) Dybtsev, D. N.; Chun, H.; Kim, K. Rigid and Flexible: A Highly Porous Metal-organic Framework with Unusual Guest-dependent Dynamic Behavior. Angew. Chem., Int. Ed. 2004, 43, 5033−5036. (25) Uemura, K.; Yamasaki, Y.; Onishi, F.; Kita, H.; Ebihara, M. Twostep Adsorption on Jungle-gym-type Porous Coordination Polymers: Dependence on Hydrogen-bonding Capability of Adsorbates, Ligandsubstituent Effect, and Temperature. Inorg. Chem. 2010, 49, 10133− 10143. (26) Grosch, J. S.; Paesani, F. Molecular-level Characterization of the Breathing Behavior of the Jungle-gym-type DMOF-1 Metal-organic Framework. J. Am. Chem. Soc. 2012, 134, 4207−4215. (27) Wang, Z.; Cohen, S. M. Modulating Metal-organic Frameworks to Breathe: a Postsynthetic Covalent Modification Approach. J. Am. Chem. Soc. 2009, 131, 16675−16677. (28) Kim, M.; Boissonnault, J. A.; Dau, P. V.; Cohen, S. M. Metalorganic Framework Regioisomers Based on Bifunctional Ligands. Angew. Chem., Int. Ed. 2011, 50, 12193−12196.

(29) Henke, S.; Schneemann, A.; Wuetscher, A.; Fischer, R. A. Directing the Breathing Behavior of Pillared-layered Metal-organic Frameworks via a Systematic Library of Functionalized Linkers Bearing Flexible Substituents. J. Am. Chem. Soc. 2012, 134, 9464− 9474. (30) Dubbeldam, D.; Calero, S.; Ellis, D.; Snurr, R. RASPA: Molecular Software Package for Adsorption and Diffusion in Flexible Nanoporous Materials. Mol. Simul. 2016, 42, 81. (31) Dubbeldam, D.; Torres-Knoop, A.; Walton, K. S. On the Inner Workings of Monte Carlo Codes. Mol. Simul. 2013, 39, 1253−1292. (32) Burtch, N. C.; Dubbeldam, D.; Walton, K. S. Investigating Water and Framework Dynamics in Pillared MOFs. Mol. Simul. 2015, 41, 1379. (33) Burtch, N. C.; Torres-Knoop, A.; Foo, G. S.; Leisen, J.; Sievers, C.; Ensing, B.; Dubbeldam, D.; Walton, K. S. Understanding DABCO Nanorotor Dynamics in Isostructural Metal-organic Frameworks. J. Phys. Chem. Lett. 2015, 6, 812−816. (34) Chen, Z.; Xiang, S.; Zhao, D.; Chen, B. Reversible Twodimensional-three Dimensional Framework Transformation within a Prototype Metal-organic Framework. Cryst. Growth Des. 2009, 9, 5293−5296. (35) Dubbeldam, D.; Walton, K. S.; Ellis, D. E.; Snurr, R. Q. Exceptional Negative Thermal Expansion in Isoreticular Metal-organic Frameworks. Angew. Chem., Int. Ed. 2007, 46, 4496−4499. (36) Liu, L.; Telfer, S. G. Systematic Ligand Modulation Enhances the Moisture Stability and Gas Sorption Characteristics of Quaternary Metal-organic Frameworks. J. Am. Chem. Soc. 2015, 137, 3901−3909. (37) Wang, C.; Li, L.; Tang, S.; Zhao, X. Enhanced Uptake and Selectivity of CO2 Adsorption in a Hydrostable Metal-organic Frameworks via Incorporating Methylol and Methyl Groups. ACS Appl. Mater. Interfaces 2014, 6, 16932−16940. (38) Hauptvogel, I. M.; Biedermann, R.; Klein, N.; Senkovska, I.; Cadiau, A.; Wallacher, D.; Feyerherm, R.; Kaskel, S. Flexible and Hydrophobic Zn-based Metal-organic Framework. Inorg. Chem. 2011, 50, 8367−8374. (39) Uemura, K.; Onishi, F.; Yamasaki, Y.; Kita, H. Syntheses, Crystal Structures, and Water Adsorption Behaviors of Jungle-gym-type Porous Coordination Polymers Containing Nitro Moieties. J. Solid State Chem. 2009, 182, 2852−2857. (40) Bellarosa, L.; Gutierrez-Sevillano, J. J.; Calero, S.; Lopez, N. How Ligands Improve the Hydrothermal Stability and Affect the Adsorption in the IRMOF Family. Phys. Chem. Chem. Phys. 2013, 15, 17696−17704. (41) Sholl, D. S.; Lively, R. P. Defects in Metal-organic Frameworks: Challenge or Opportunity? J. Phys. Chem. Lett. 2015, 6, 3437−3444.

2857

DOI: 10.1021/acs.accounts.5b00311 Acc. Chem. Res. 2015, 48, 2850−2857