Subscriber access provided by READING UNIV
Article
On the Better Understanding of the Surprisingly High Performance of Metal-Organic Framework Based MixedMatrix Membranes Using the Example of UiO-66 and Matrimid Sebastian Friebe, Alexander Mundstock, Kai Volgmann, and Jürgen Caro ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b13037 • Publication Date (Web): 07 Nov 2017 Downloaded from http://pubs.acs.org on November 7, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces
On the Better Understanding of the Surprisingly High Performance of Metal-Organic Framework Based Mixed-Matrix Membranes Using the Example of UiO-66 and Matrimid Sebastian Friebe*, Alexander Mundstock, Kai Volgmann, Jürgen Caro Institute of Physical Chemistry and Electrochemistry, Gottfried Wilhelm Leibniz University Hannover, Callinstraße 3A, 30167 Hannover, Germany. ABSTRACT: Metal-Organic Frameworks feature a certain framework flexibility, mainly due to a linker mobility inside the lattice. The latter is responsible for effects like breathing or gate-opening, thus making predictions of the sorption and diffusion behavior quite difficult. Permeation measurements on supported UiO-66 membranes at low temperatures and on polymer-coated UiO-66 membrane layers as well as 2H-NMR line shape studies and nitrogen sorption measurements of UiO-66 with deuterated linkers in Matrimid as Mixed-Matrix Membrane (MMM) indicate that the 2-site 180° flips (π-flips) of the aromatic ring are hindered by the presence of (i) the surrounding polymer Matrimid and (ii) residual solvent molecules, thus giving profound insights in the molecular understanding of gas transport through Metal-Organic Framework based MMMs. Keywords: UiO-66, Metal-Organic Framework (MOF), Mixed-Matrix Membrane (MMM), Gas Separation Performance, Linker Mobility, Polymer Coating, NMR-Spectroscopy
INTRODUCTION: Metal-Organic Frameworks (MOFs) attracted a lot of interest during the last two decades, since they offer an almost unlimited stock for the synthesis of new materials1,2,3,4,5. Furthermore, the building-block principle and the resulting custom tailoring allow their versatile preparation in terms of desirable properties6,7,8. Thus, porous MOFs can feature functional groups or unsaturated metal sites for adsorption or separation applications9,10,11. Otherwise, they can be used for catalysis12,13,14 or medical drug release applications15,16. On top, by using appropriate starting materials, they can work as smart materials by re-/acting on external stimuli (e.g. UVirradiation)17,18. However, they can also be mixed with polymers, thus resulting in so called Mixed-Matrix Membranes (MMMs). This type of membranes combines all the positive properties of the filler component (tailorable features) and the easy processability and robustness of the polymer. Additionally and surprisingly, the composite material often features better properties as the two individual components19,20,21. Hence, these composite materials are interesting for industrial relevant separation processes like carbon capture and storage (CCS)22,23 or alcohol/water separation via pervaporation24. Furthermore, very recently several works reported the future perspectives, opportunities as well as the challenges for Mixed-Matrix Membranes and their successful preparation as a solution for versatile separation problems25,26,27,28,29,30. Nevertheless, if it is about the understanding or the prediction of the MMM performance, the common models (e.g. Maxwell, Bruggemann, etc.) often under- or overestimate the real values due to miraculous effects. The latter miscalculation is frequently attributed to different ramifications such as the filler embedment quality, polymer rigidification around the incorporated additives or reduced permeability regions inside the polymer or the porous filler31,32,33,34. However, another effect which ought to affect the performances of the MMM to a great degree is the reduced framework flexibility
of the MOF due to the rigid surroundings inside the polymer matrix. The latter is neither understood in detail nor sufficiently studied. In this work we investigated three different concepts to affect the linker mobility and to characterize the concomitant effects, respectively (c.f. Fig. 1).
Figure 1: Illustration of the different concepts affecting the linker mobility within Metal-Organic Frameworks. The rigidity of UiO66, for instance, can be increased by direct cooling (A), a polymer cover (B) and the embedment of particles in a polymeric matrix (C). All the concepts affect the 2-site 180° flips (π-flips) of the benzene rings and thereby the effective window size of the MOF.
1 Environment ACS Paragon Plus
ACS Applied Materials & Interfaces
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The first approach was to suppress the thermally triggered linker motion of a MOF via direct cooling to low temperatures. For instance, the MOF UiO-66 is known to show a temperature dependent linker motion35,36. Furthermore, the terephthalate linkers bridging the secondary building units (SBUs) have the lowest activation energy compared to other common MOFs. At 213 K, for example, the linkers are quasi-static while they perform fast and slow flips along the C2 symmetry axis at room temperature. Consequently, a supported UiO-66 membrane (for detailed synthesis see SI) was prepared as a thin layer on top of a porous alumina support and mixed gas permeation was measured in dependence of the temperature in a range from ≈ 200 K to room temperature. The second idea was to stop the linker motion by an additional polymer layer on top of the MOF layer as proposed in ref. 37, which should modify the framework flexibility of the MOF at the interface via some kind of “gumming up”. Therefore, a membrane of type UiO-66 was first synthesized on a porous ceramic support as described above and subsequently covered with a thin (15 µm) Matrimid layer (for detailed synthesis see SI). This top layer should affect the linker mobility in the interface MOF-polymer, thus giving increased separation selectivities in comparison to the uncoated MOF membrane layer, as previously investigated in different works38,39,40. Third, nanosized particles of the deuterated MOF UiO-66 were incorporated completely in a polymeric matrix in a 50/50 ratio (for detailed information see SI). This composite was subsequently investigated with 2 H-NMR to clarify the linker motion of the MOF inside the polymeric matrix. The reduced framework flexibility through direct cooling was investigated by measuring the gas separation performance of UiO-66 membranes for a binary H2/CH4 mixture between ≈ 200 K and room temperature. Therefore, the membrane module was encased with liquid nitrogen cooled copper band and Al-foil to maintain the temperature conditions (c.f. Figs. S1 & S2). The H2/CH4 mixed gas permeation results in Fig. 2 indicate a strong temperature influence on the separation results.
Fig. 2: Reversible H2/CH4 mixed gas separation results for the neat supported UiO-66 membrane (c.f. Fig. 1) between temperatures of ≈ 200 K and room temperature. Starting point at 293 K, then cooling to low temperatures (≈ 200 K) and warming up to room temperature again. Error bars are based on two membranes tested.
Page 2 of 8
At temperatures around 213 K the terephthalate ring motion in UiO-66 is known to be completely stopped35,36. This goes hand in hand with the mixed gas permeation results. The hydrogen permeance is two times smaller compared to room temperature. However, since methane shows a greater kinetic diameter, the stopped or hindered π-flips of the terephthalate linkers affects CH4 to a greater degree, meaning the CH4 permeance is a magnitude smaller compared to room temperature. Consequently, the mixed gas separation factor increases from α (H2/CH4) = 17 at room temperature to round about 40 at low temperatures (≈ 200 K). This great increase can not only be a result of the different diffusion coefficients of hydrogen or methane but should also be based on the hindered linker mobility, thus converting UiO-66 to a more rigid framework, as already shown for the corresponding powder by Jobic et al35. When the membrane is warmed up to higher temperatures, the terephthalate rings start to vibrate and swing until they can perform π-flips at room temperature again. This temperature dependent behaviour of the benzene ring motions inside the UiO-66 membrane can indirectly be traced by the permeabilities of the mixture components and the corresponding separation factors. At the coolest temperature (≈ 200 K), the separation factor shows the highest value, whereas the mixture permeabilities are lowered, as a consequence of the stopped π-flips (c.f. Fig. 1). Going to medium temperatures, the gas mixture fluxes increase and the separation factors decrease steeply, since the benzene rings can vibrate and swing yet again. At room temperature, the separation factor exhibits the minimal value, whereas the gas mixture permeabilities are at the maximum. This measurement indicates that the linker motion inside a MOF membrane can successfully be suppressed by direct cooling. To the best of our knowledge, this is the first time a MOF membrane was modified in terms of its gas separation capabilities by cooling to low temperatures, thus affecting the framework flexibility. In a second approach, we suppressed linker motion of the MOF by covering UiO-66 with a thin polymer layer (Matrimid), thus increasing the gas separation capabilities (for preparation details see SI). This experimental finding is in complete accordance with former results for two other MOF/polymer sandwich composites (c.f. Fig. 3)37. The permeation patterns of sandwich membranes (polymer coating on top of a supported MOF layer) is another indirect hint for the modulated framework flexibility. The additional polymer layer ought to modify the rigidity of the MOF at the MOF/polymer interface, thus giving enhanced separation capabilities. This increased separation performance due to the polymer top layer is in good accordance to the results presented in Fig. 2 accomplished by direct cooling. The membrane materials shown in Fig. 3 exhibit a comparable tendency. All presented neat MOF membranes (ZIF-8, ZIF-90, UiO-66) feature similar selectivities for the separation of binary H2/CH4 mixtures but with varying permeabilities, most likely due to the different pore sizes and geometries of the MOF membranes (ZIF-8 ≈ 3.4 Å, ZIF-90 ≈ 3.2 Å, UiO-66 ≈ 6.0 Å) and their crystallinity.
2 Environment ACS Paragon Plus
Page 3 of 8
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
ACS Applied Materials & Interfaces thalate rings in dependence of the prevailing temperature35. This thermally triggered π-flip motion of the aromatic benzene ring can directly be seen in the 2H-NMR spectra of the free d4-UiO-66 powders (c.f. Fig. 4).
Fig. 3: Coating of thin-layer supported neat MOF membranes by a polymer Matrimid film reduces permeability and increases selectivity in H2/CH4 mixed gas separation. Wicke-Kallenbach measurement at room temperature and 1 bar. Note that the separation selectivity increases above the value of the pure 37 Matrimid . Error bars are based on three membranes tested. By covering these MOF membranes with thin Matrimid layers, the permeability of the resulting multilayer composite greatly decreased. At the same time, the mixture separation factor drastically increased for all materials. Although the latter can be a result of a plethora of reasons (defect healing, pore size reduction by polymer infiltration, etc.), we think these results also give an indirect hint for mutated linker mobilities, since the single material membrane layers (Matrimid or UiO-66) show lower separation factors α for H2/CH4 mixtures. In a third approach, we studied the reduced rotational linker motion when embedding d4-UiO-66 powder in a polymeric matrix as it is the case for an MMM. XRD measurements for the neat d4-UiO-66 powder and the corresponding 50/50 d4UiO-66/Matrimid composite show that both the latter did not suffer from the harsh activation procedure (c.f. Fig. S3) and are, therefore, suited for further analysis. To explore the origin of the selectivity increase in terms of the separation performance of an MMM, we prepared a deuterated UiO-66 (d4-UiO-66) powder (for detailed synthesis see SI). This d4UiO-66 powder was then mixed with Matrimid in a 50/50 ratio and the solid state 2H-NMR spectra (c.f. Fig. 4) were measured at room temperature (characterization details in SI; c.f. Figs. S3, S4, S5 & S6). Certain molecular motions of the aromatic benzene ring like fast C6 or a C2 rotation will reduce the quadrupolar splitting of the Pake doublets in a characteristic way as shown in a pioneering paper for deuterated benzene adsorbed in zeolites41. If the correlation time τC of this motion is sufficiently short compared to the 2H-NMR quadrupole frequency ωQ i.e. τCωQ
